Briefly describe the implications to scholars in academia (so what? in what ways you can utilize the ideas in the readings in your professional development as a scholar).
Summarize the readings: (1,2,3)
Summarize and Reaction the readings and make a presentation (PowerPoint)
Two files
Your research project: chapter 5
Doing research in the business world: chapter 5
Research in education: chapter 4
***There are two main parts to the summary: ***
1. Summary of each chapter/article: Includes Title of Article, Author(s), Source, and Date of Article using APA style. In your OWN WORDS describe what the article is about, with major details or points, and should be easy to read (i.e. interesting and flow well!)
2. Reaction: Briefly describe the implications to scholars in academia (so what? in what ways you can utilize the ideas in the readings in your professional development as a scholar).
***There are main part in the presentation (PowerPoint): ***
I will lead the discussions of the readings. The student leading the discussion is must create a PowerPoint presentation, with bullet points that I want to talk about.
##provide preliminary questions for students to answer in class, and may require additional readings, among other responsibilities.
***Use easy words and easy Sentence***
Requirements: 1 day
Your Research Project.
Your Research Project.Designing, Planning, and GettingStartedNicholas WallimanFourth EditionLos AngelesLondonNew DelhiSingaporeWashington DCMelbourne
SAGE Publications Ltd1 Oliver’s Yard55 City RoadLondon EC1Y 1SPSAGE Publications Inc.2455 Teller RoadThousand Oaks, California 91320SAGE Publications India Pvt LtdB 1/I 1 Mohan Cooperative Industrial AreaMathura RoadNew Delhi 110 044SAGE Publications Asia-Pacific Pte Ltd3 Church Street#10-04 Samsung HubSingapore 049483© Nicholas Walliman 2020First published 2020Apart from any fair dealing for the purposes of research or privatestudy, or criticism or review, as permitted under the Copyright,Designs and Patents Act, 1988, this publication may bereproduced, stored or transmitted in any form, or by any means,only with the prior permission in writing of the publishers, or in the
case of reprographic reproduction, in accordance with the terms oflicences issued by the Copyright Licensing Agency. Enquiriesconcerning reproduction outside those terms should be sent to thepublishers.Library of Congress Control Number: 2019943692British Library Cataloguing in Publication dataA catalogue record for this book is available from the British LibraryISBN 978-1-5264-4119-5ISBN 978-1-5264-4120-1 (pbk)Editor: Michael AinsleyEditorial assistant: Amber Turner-FlandersProduction editor: Victoria NicholasMarketing manager: Ben Griffin-SherwoodCover design: Lisa Harper-WellsTypeset by: C&M Digitals (P) Ltd, Chennai, IndiaPrinted in the UKAt SAGE we take sustainability seriously. Most of our products are printed inthe UK using responsibly sourced papers and boards. When we print overseaswe ensure sustainable papers are used as measured by the PREPS gradingsystem. We undertake an annual audit to monitor our sustainability.
To my wife, Ursula
Summary of ContentsContentsAbout the AuthorAcknowledgementsIntroduction1 Types of Research and the Research Problem: BeginningYour Project2 Research Theory and the Nature of Knowledge:Understanding Philosophies and Approaches3 Using Language and Understanding Arguments4 Finding, Organizing and Retrieving Information5 Doing Your Literature Review, Forming Original Ideas andDefining Your Research Topic6 Honesty and Research Ethics: Establishing an Ethical Code7 Research Methods: Choosing the Best Methods for YourProject8 The Research Proposal: Planning and Structuring Your Ideas9 Writing Strategies: Getting Started and MaintainingMomentum Throughout Your ProjectAnswers to ExercisesKey Words GlossaryReferencesIndex
ContentsAbout the AuthorAcknowledgementsIntroductionParticular features of the bookStructure and outline of the book1 Types of Research and the Research Problem: BeginningYour ProjectAimsIntroductionWhat is research?What it is for – the objectives of researchTypes of researchHistoricalComparativeDescriptiveCorrelationExperimentalEvaluationActionEthnogenicFeministCulturalThe research processDesirable characteristics of research findingsStarting your own researchFinding and defining a research problemSome common mistakesAids to locating and analysing problemsInitial literature review and defining the problem areaResearch problem definitionThe sub-problemsSecond review of literaturePlanning a research projectExplorationTesting outProblem-solving
Which type of research lends itself best to gaining aresearch degree?Your project planThe next steps: finding your research problem areaConsolidation and assessmentFurther reading2 Research Theory and the Nature of Knowledge:Understanding Philosophies and ApproachesAimsIntroductionScientific enquiryPure inductionDeduction and the principle of falsificationHypothesisFalsificationHypothetico-deductive or scientific methodCritical rationalismRelativism‘Anti-method’ stanceApproaches to social researchThe positivist approachThe interpretivist approachThe reconciliatory approachCritical realismStructuralism, post-structuralism and postmodernismConclusionsThe next steps: the theoretical basis of your researchprojectConsolidation and assessmentFurther reading3 Using Language and Understanding ArgumentsAimsIntroductionConcepts and theoryConceptsAbstract and concrete conceptsConcept measurementLevels of measurementTheoryTheoretical statements
Language and argumentStatements: existent and relationalDiscourse and argumentUsing argumentDeductive and inductive argumentsLogic in argumentFallacies in argumentClassification and analogy in argumentConclusionsThe next steps: which type of research for your topic?Consolidation and assessmentFurther reading4 Finding, Organizing and Retrieving InformationAimsIntroductionFinding the literatureSources of informationThe InternetEvaluating Web sourcesLibrariesReadingNote-takingReasons for taking notesNote-taking techniquesCollecting and ordering your notesPersonalized bibliographiesConclusionsThe next steps: devising your own information system andwriting your literature reviewConsolidation and assessmentFurther reading5 Doing Your Literature Review, Forming Original Ideas andDefining Your Research TopicAimsIntroductionDoing a literature reviewCritical reading skillsDifferent strategies explained in more detailCritical appraisalState of the art
Different and competing theoriesMethodology reviewCross-disciplinary explorationHistorical overviewDoing the reviewStyle and content of the reviewForming original ideas and defining your research projectConclusionsThe next steps: Writing your literature review and definingyour research project, and exploring appropriate researchmethodsConsolidation and assessmentFurther reading6 Honesty and Research Ethics: Establishing an Ethical CodeAimsIntroductionHonesty in your work and avoiding plagiarismIntellectual ownership and plagiarismAcknowledgement of other people’s work throughcitation and referencingResponsibility and accountability of the researcherData and interpretationsWhere do you stand?Situations that raise ethical issuesResearch aimsMeans and endsEthics in relation to other peopleTerminologyRoles of researcher and researchedParticipantsCarrying out the researchPotential harm and gainInterviews and questionnairesParticipant involvement: experiments, observations,groupsSensitive materialHonesty, deception and covert methodsUsing assistants or delegating tasksSocial media researchDealing with data
Storing and transmitting dataAnalysing the data and getting resultsEthics policies, permissions and committees inorganizationsEthics committeesConclusionsThe next steps: plan your code of ethicsConsolidation and assessmentFurther reading7 Research Methods: Choosing the Best Methods for YourProjectAimsIntroductionData collection and analysisQuantitative, qualitative and mixed methods researchResearch strategiesCollecting secondary dataTypes and sources of secondary dataSuitability of data for your projectAuthentication and credibilityAnalysing secondary dataContent analysisData miningMeta-analysisCollecting primary dataSurvey researchSamplingQuestionnairesDiariesInterviews: structured, semi-structured and openStandardized scales and testsAccountsObservations and physical surveysUsing the Internet for primary researchCombined data generation and analysisLaboratory and field experimentsTypes of experimentModelsAnalysing primary dataQuantitative analysis
StatisticsQualitative analysisPreliminary analysis during data collectionTypologies, taxonomies and codingPattern coding, memoing and interim summaryFurther techniques during data collectionMain analysis during and after data collectionQualitative analysis of texts, documents and discourseMixed methods and interdisciplinary researchMixed methodsHow to use mixed methods researchData management and mixingInterdisciplinary researchConclusionsThe next steps: which research methods will you use?Consolidation and assessmentFurther reading8 The Research Proposal: Planning and Structuring Your IdeasAimsIntroductionDefining your research projectThe recipe for a successful research proposalTypes of research proposalThe main ingredients and sequenceThe titleAims of the researchThe background and literature reviewThe research problemOutline of methodsPossible outcomes of the researchAdditional information for proposals for funded researchFinalizing your proposalSuccessful proposalsConclusionsThe next steps: your research proposalConsolidation and assessmentFurther reading9 Writing Strategies: Getting Started and MaintainingMomentum Throughout Your ProjectAims
IntroductionWhere to begin on your dissertation or thesisThe writing processForming the structure and preparing an outlineRetrieving and organizing notesDrafting and redraftingParagraphsIllustrationsQuotationsPunctuationSpellingTables and figuresBibliographies, references and footnotesMotivation and maintaining momentumOther types of writing you might need to doPresentationsPostersConference papersResearch papers for refereed academic journalsProgress reportsArticlesConclusionsThe next steps: getting to write your thesis or dissertation orresearch reportConsolidation and assessmentFurther readingAnswers to ExercisesKey Words GlossaryReferencesIndex
About the AuthorNicholas Wallimanis a qualified architect and Associate Lecturer in the School ofthe Built Environment at Oxford Brookes University and was aresearch associate in the Oxford Institute for SustainableDevelopment. After many years of practice in architecture inthe UK and abroad, he returned to academic life to do his PhD.This experience raised his interest in research theory andmethods, and he was subsequently asked by the university towrite a distance learning course to guide postgraduatestudents embarking on research degrees. This courseprovided the raw material and incentive for writing this book.He has subsequently published several other books onresearch theory and methods for students and practitioners atvarious levels of expertise.He has been engaged in nationally and internationally fundedprojects on a range of aspects of building technology, such asenergy saving building envelope design, and mitigation of theeffects of floods on buildings and advanced constructionmethods. He has published numerous research papers onaspects of architectural technology. He is also supervisingseveral PhD, master’s and undergraduate students for theirtheses and dissertations. Despite this emphasis on scienceand technology, his work with research students covers manyother aspects of architecture and its relationship to culture andsociety, such as vernacular architecture, the effects ofwesternization, architectural education, conservation,administration and sustainable design.
AcknowledgementsMy grateful thanks go to Dr Rowland Newman and Professor MikeJenks, who originally gave me inspiration to write this book; to DrBousmaha Baiche for his help; to Mrs Margaret Ackrill, Mrs ValBacon and the postgraduate research students of the Departmentof Architecture, Oxford Brookes University, for their comments andsuggestions; and to my family for their unfailing support. WithoutUrsula, my wife, I could not have completed this book nor its latereditions, as she fed me, motivated me and cared for me in everyway.My thanks go to the following people and organizations forpermission to reproduce material in this book: Dr RolandNewman, Mrs Margaret Ackrill, Prof David Bonnett, Dr RobertIlles, Dr Alison Chisholm, Dr Louise Waite, Dr Adi Walker, Dr MinaSamangooei, Catalina Moralez Maya, Fatima Hashmi and HarryDodd. Also to fellow authors at Sage, R. Yin, E. Guba, Y. Lincoln,C. Hewson, P. Yuyle, D. Laurent, C. Vogel, L. Locke, W. Spirdusoand S. Silverman. And to all the authors that I have cited whohave made such valuable contributions to the debate aboutresearch. Specific acknowledgement is made where the materialappears.I have greatly appreciated the guidance and help afforded by allthe editorial team at Sage, in particular Michael Ainsley, AlyshaOwen, Patrick Brindle, Vanessa Harwood, Claire Lipscomb,Rachel Burrows, David Hodge, Katie Metzler and VictoriaNicholas for their encouragement and patience.
IntroductionAs a student or practitioner embarking on a research project forthe first time, it is difficult for you to know how to begin and how todevelop a research proposal in your chosen subject which willsatisfy the requirements of your superiors, educational and otherorganizations, and funding bodies. Most students attempting theirfirst dissertation or embarking on a research degree have littleknowledge or experience of research and are often not clear as tothe exact subject they wish to research. Contract researchers,practitioners and managers are also under pressure to workefficiently and in a well-focused manner. In general, supervisorsand bosses have little or no time to instruct their students andemployees in the theory and practice of research, so it is left toyou, the novice researcher, to wade through the bewilderingvariety of theoretical and technical books about research in orderto try to develop a credible research proposal related to yourinterest. This must be done when your skills are often veryrudimentary, and when you find it difficult to make the connectionbetween the general theory and practice of research and yourown research interests.This book provides you with a hands-on guide when you arebeginning to do research in any subject to do with social sciences,the environment, business studies, education and the humanities.It guides you to writing a successful research proposal – a crucialdocument, as its approval is the condition for continuing researchand often for obtaining funding. It leads you through the differentstages in planning a research project in a way that encouragesyou to progress your own work in tandem with your reading.Hence, this book focuses on the first stages of research: theselection of a topic and the planning of the project. In order toachieve these stages, however, it is necessary for you tounderstand the nature and purpose of research, the choice ofresearch methods, how to structure an argument and theimportant issue of research ethics – essentially, all the basicinformation about research and how to do it. Equipped with this
knowledge, and having prepared a sound research proposal, youwill be equipped with a solid foundation for carrying out theresearch itself. This book also provides advice on how to thenstart and stay motivated through your research writing and doingthe other written work associated with conducting a researchproject.The objective of this book is achieved by systematically impartinga basic understanding of the theory of and approaches toresearch while at the same time helping you to develop thesubject of your research, encouraging the formation of a high levelof trained intellectual ability, critical analysis, rigour andindependence of thought, fostering your individual judgement andskill in the application of research theory and methods, anddeveloping your skills required in writing research proposals,reports and theses.Particular features of the bookThe particular features of this book are the way it:combines the explanation of practical and theoretical aspectsof research directly with the progressive development of yourideas about your individual research topicalways gets you to relate this to your own subject of study,with no dry theorizing which is difficult to relate to yourindividual research interestsis divided into chapters with clearly limited objectives,requiring you to apply the aspects of research you havelearned in each chapter to the next stage in developing yourproposalregularly provides you with checklists of issues you shouldconsider or tasks you should undertake in order to progressyour worksets points for contemplation about the applicability of thelearned aspects to your individual project and gives aframework for issues to be discussed with your tutors andcolleagues
uses a direct approach, leading you step-by-step through thebook with interesting and amusing self-assessed exercises totest and develop your knowledge.Structure and outline of the bookThe book is structured around three sequential approaches, thefirst being a cumulative approach, which introduces, step-by-step,the features of, and debate about, the academic subject ofresearch theory and methods. This is offered in parallel with aproblem-centred sequence that involves you in the practical workof developing the skills needed to devise a good-quality researchproposal. There is, thirdly, an element of a spiral sequence,whereby concepts and techniques are reiterated and developedduring the book as your understanding increases.There are nine chapters in the book, which are designed to beworked through consecutively. The main sections in each chapterare devised to explain a major aspect of research theory orapproach. These sections contain informative, discursive text,regularly interspersed with checklists and exercises for you toconsolidate and assess your understanding of the subjectspresented.There then follows a section (‘the next steps’) which consists ofapplication to your own area of research of the specific topicsdiscussed in the earlier sections. Aims are set out and tasks aredefined. This section is always exploratory in nature, and it isexpected that you will devise a range of alternative solutions tothe tasks. The results of this work should, ideally, then bediscussed with your fellow students, colleagues, a tutor or asupervisor or team leader, who will wish to see that you haveunderstood the issues in the chapter and how they can be appliedin practice. Decisions made at the end of each chapterconsolidate progress towards writing your research proposal andstarting your research. Each chapter concludes with suggestionsfor further reading. The answers to the assessment exercises aregiven at the end of the book.
The wide range of approaches to the subject of research makeswriting an introduction to research rather difficult. It is not possiblewithin the scope of this book to cover every research strategy thatmight be relevant to your interests. Therefore the book has beenlimited to explaining the characteristics of the major researchapproaches, and mentioning alternatives where appropriate. Youwill become aware that the subject of research is widely debated,and therefore a prescriptive approach is not appropriate. There is,however, a range of basic research techniques which you need tobecome good at, whatever your subject: for example, analyticalreading and thinking, note-taking and referencing skills, andwriting skills and the ability to construct a sound argument.Depending on your previous research experience and languageskills, particularly if English is a foreign language, you might needto spend 10–20 hours on each chapter. It is important for you toremember that you should continue to read widely in your chosenresearch subject while following this book, as specialistknowledge on the chosen subject will be required in order to fulfilthe tasks set.In this fourth edition, more stress has been given to the checklistsin order to engage you more actively in progressing your work.More guidance on doing a literature review has been added, andthe use of the Internet in research both from a practical andethical view has been included. More examples of successfulwork done by research students have been included to providepointers to the form and quality of writing required. The section onresearch methods has been expanded to include mixed methodsapproaches and multidisciplinary work. Advice on doing all sortsof written and presentation work normally required as part of yourresearch has been added in an extended section.Although it is assumed that most of you readers are new to theresearch process and to the demands made on researchers, thisbook also will provide a useful source of information for you if youare more experienced but wish to revise your skills or developthem further.
1 Types of Research and theResearch Problem: BeginningYour ProjectAimsTo explain what research is, and what it is not, and theobjectives of researchTo outline the different types of researchTo discuss the research processTo introduce the concept at the heart of any research project –the research problem – and to discuss what a researchableproblem isTo warn of common mistakesTo describe how to choose your research strategy and plan yourresearch projectIntroductionThe shortest way of describing the contents of this chapter is tosay that it provides a starting point for your research efforts.It introduces the concept of research as understood in theacademic world, and contrasts it to the loose way the word‘research’ is used in everyday speech. However, even in theacademic world, the nature of research is the subject of a greatdeal of debate. The characteristics of scientific method are brieflyexplained, and the interpretivist alternative is discussed as one ofthe aspects of the debate about research methods. This debate istreated in much greater detail in Chapter 2. An overview of theresearch process is given showing various ways to illustrate it.An essential early step in the process of research is to find aresearch problem. What a research problem is and how tofind one are explained. The nature of your problem will, in its turn,
influence the form of your research. It is this quest for a problemwhich forms the task in the final section, where what you havelearned in the earlier sections is applied to your own subject.Research problemKey words are shown in bold and are repeated in the margin soyou can scan through the chapter to check up on their meaning.What is research?‘Research’ is a term loosely used in everyday speech to describea multitude of activities, such as collecting masses of information,delving into esoteric theories, and producing wonderful newproducts. It is important that a student or practitioner embarkingon a programme of academic or practical research has a clearidea of what the word ‘research’ really means, and clears awayany misconceptions that might exist owing to the word’s commonuse in other fields.TermIt is, therefore, worth looking at a few of the ways that the word isused in common language to describe activities, often calledresearch, which are not research in its real meaning, and also atsome of the emotive language that surrounds the term.These are some of the ways in which the term ‘research’ iswrongly used:As a mere gathering of facts or information: ‘I’ll go and do a bitof research into the subject.’ This usually means quicklyreading through a few books or magazines to become betterinformed about something. Such information can be collectedin other ways too, e.g. by asking people questions in the streetor by recording the number of vehicles driving along a road.This kind of activity may more accurately be called ‘collectionof information’, and can be carried out in a systematic and
thorough way. It certainly can be seen as an important part ofresearch.Moving facts from one situation to another: ‘I have done myresearch, and come up with this information which I present inthis paper.’ It is easy to collect information and reassemble it ina report or paper, duly annotated and referenced, and think ofit as research. However, even if the work is meticulouslycarried out, and brings enlightenment about the subject to theauthor and the reader, one vital ingredient of the researchprocess is missing – the interpretation of the information. Onemight call this form of activity ‘assembly of information’. This is,as with the collection of information, an important componentof research, but not its entirety.As an esoteric activity, far removed from practical life: ‘He’s justgone back into his laboratory to bury himself in his researchinto the mysterious processes of bimolecular fragmentation.’While many research projects deal with abstract andtheoretical subjects, it is often forgotten that the activity ofresearch has greatly influenced all aspects of our daily livesand created our understanding of the world. It is an activity thatis prompted by our need to satisfy our natural curiosity and ourwish to make sense of the world around us.As a word to get your product noticed: ‘Years of painstakingresearch have produced this revolutionary, labour-savingproduct!’ Very often the term ‘research’ is used in an emotivefashion in order to impress and build confidence. If you ask forevidence of the research process and methodology, you arelikely to be faced with incomprehension, muddled thinking, andpossibly even worse: the product may be the outcome of mereguesswork!So how can true research be defined? Box 1.1 suggests somealternatives.Box 1.1 Definitions of researchThe Oxford Encyclopaedic English Dictionary defines research as:the systematic investigation into the study of materials, sourcesetc. in order to establish facts and reach new conclusions
an endeavour to discover new or collate old facts etc. by thescientific study of a subject or by a course of critical investigation.(OEED, 1991, p. 1228)Leedy and Ormrod define it from a more utilitarian point of view:Research is a systematic process of collecting, analyzing, andinterpreting information – data – in order to increase ourunderstanding of a phenomenon about which we are interested orconcerned. (2015, p. 20)Dominowski is so terse in his definition that he seems to miss thepoint (see above):Research is a fact-finding activity. (1980, p. 2)Kerlinger uses more technical language to define it as:the systematic, controlled, empirical and critical investigation ofhypothetical propositions about presumed relations among naturalphenomena. (1999, p.108)Creswell clearly mentions the steps in the process of research, hereabout educational research:Research is a cyclical process of steps that typically begins withidentifying a research problem or issue of study. It then involvesreviewing the literature, specifying a purpose for the study,collecting and analysing data, and forming an interpretation of theinformation. This process culminates in a report, disseminated toaudiences, that is evaluated and used in the educationalcommunity (2014, p.11)You could go on finding definitions of research, which would, as inthe examples in the box, differ in emphasis and scope. What iscertain is that there are many different opinions about, andapproaches to, research. However, as a means of achieving agreater comprehension of our world, research distinguishes itself
from the two other basic and more ancient means, those ofexperience and reasoning.Briefly, experience results in knowledge and understandinggained either individually or as a group or society, or shared byexperts or leaders, through day-to-day living. Reflectiveawareness of the world around us, present to a degree even inother mammals, provides invaluable knowledge. The mostimmediate form of experience is personal experience, the body ofknowledge gained individually through encountering situationsand events in life. A child learns to walk by trial and error, and anadult gets adept at decorating jobs in the house after renovatingseveral rooms. When solutions to problems are not to be foundwithin the personal experience of an individual, then he or shemay turn to those who have wider or more specialist experiencefor advice, for example a solicitor in legal matters. Beyond this arethe ‘experts’ who have written books on particular subjects, e.g.healthcare or the finer points of playing golf.ExperienceFigure 1.1 Knowledge gained from experience forms an essentialaid to our understanding and activities in everyday life
Knowledge gained from experience forms an essential aid to ourunderstanding and activities in everyday life. However, it doeshave severe limitations as a means of methodically and reliablyextending knowledge and understanding of the world. This isbecause learning from experience tends to be rather haphazardand uncontrolled. Conclusions are often quickly drawn and notexhaustively tested, ‘common sense’ is invoked as self-evident,and the advice of experts is frequently misplaced or seen asirrelevant. Despite these shortcomings, experience can be avaluable starting point for systematic research, and may provide awealth of questions to be investigated and ideas to be tested.
Reasoning is a method of coming to conclusions by the use oflogical argument. There are three basic forms of argument:deductive, inductive and a combination of both calledinductive/deductive (or hypothetico-deductive, or scientificmethod). Deductive reasoning was first developed by the AncientGreeks, and was refined by Aristotle through his deductivesyllogisms. An argument based on deduction begins with generalstatements and, through logical argument, comes to a specificconclusion. A syllogism is the simplest form of this kind ofargument and consists of a major general premise (statement),followed by a minor, more specific premise, and a conclusionwhich follows logically. Here is a simple example:ReasoningArgumentAll live mammals breathe. – general premiseThis cow is a live mammal. – specific premiseTherefore, this cow breathes. – conclusionInductive argument works the other way round. It starts fromspecific observations and derives general conclusions therefrom.Its logical form cannot be so neatly encapsulated in a three-lineformat, but a simple example will demonstrate the line ofreasoning:All swans that have been observed are white in colour. –specific observationsTherefore one can conclude that all swans are white. –general conclusion
The value of inductive argument was revealed by Bacon in the1600s. By careful and systematic observation of the events in theworld around us, many theories have been evolved to explain therules of nature. Darwin’s theory of evolution and Mendel’sdiscovery of genetics are perhaps the most famous theoriesclaimed (even by their authors) to be derived from inductiveargument.However, deductive reasoning was found to be limiting because itcould only handle certain types of statement, and could becomeincreasingly divorced from observation and experience. Purelyinductive reasoning proved to be unwieldy and haphazard, and inpractice was rarely applied to the letter. Medawar (2009, pp. 10–11) quoted Darwin writing in his sixth edition of Origin of Species,where he said of himself that he ‘worked on true Baconianprinciples, and without any theory collected facts on a wholesalescale’, but later on he admitted he could not resist forming ahypothesis on every subject.When inductive and deductive argument were combined to forminductive/deductive argument, the to-and-fro process ofdeveloping hypotheses (testable theories) inductively fromobservations, charting their implications by deduction, and testingthem to refine or reject them in the light of the results, formed apowerful basis for the progress of knowledge, especially ofscientific knowledge, and is now commonly referred to asscientific method.It is the combination of experience with deductive and inductivereasoning which is the foundation of modern scientific research.Three characteristics of research can be seen to distinguish itfrom gaining knowledge either purely by experience or byreasoning, as shown in Box 1.2.Box 1.2 Three characteristics of research1. Gaining experience is an uncontrolled and haphazard activity,while research is systematic and controlled.2. Reasoning can operate in an abstract world, divorced fromreality, while research is empirical and turns to experience andthe world around us for validation.
3. Unlike experience and reason, research aims to be self-correcting. The process of research involves rigorously testingthe results obtained, and methods and results are open to publicscrutiny and criticism.But note that in addition:… research, far from being a mechanistic exercise, is adeliberative, complex, challenging, subtle activity and often amessier process than researchers would like it to be. (Cohen,Manion and Morrison, 2017, p. 3)When we talk about scientific and systematic research, it isusually assumed that it makes use of the rigorous and questioningtechniques of scientific enquiry. This form of enquiry is calledscientific method.What it is for – the objectives ofresearchResearch can have several legitimate objectives, either singly orin combination. The main, overriding objective must be that ofgaining useful or interesting knowledge. Reynolds (2016, pp. 4–11) listed five things that he believed most people expectedscientific knowledge to provide. These, together with one that Ihave added myself, can conveniently be used as the basis for alist of the possible objectives of research, as in Checklist 1.1.Checklist 1.1 What will be your research objective?To categorizeTo explainTo predictTo create a sense of understandingTo provide potential for controlTo evaluate
Categorization involves forming a typology of objects, eventsor concepts. This can be useful in explaining what ‘things’ belongtogether and how. One of the main problems is to decide on themost useful methods of categorization, depending on the reasonsfor attempting the categorization in the first place. Following fromthis is the problem of determining what criteria to use to judge theusefulness of the categorization. Two obvious criteria arementioned by Reynolds: that of exhaustiveness, by which all itemsshould be able to be placed into a category, without any being leftout; and that of mutual exclusiveness, by which each item should,without question, be appropriately placed into only one category.Finally, it should be noted that the typologies must be consistentwith the concepts used in the theoretical background to the study.CategorizationThere are many events and issues that we do not fully, or evenpartly, understand. The objective of providing an explanation ofparticular phenomena has been a common one in many forms ofresearch.ExplanationOn the basis of an explanation of a phenomenon it is oftenpossible to make a prediction of future events related to it. Inthe natural sciences these predictions are often made in the formof abstract statements, for example given C1, C2, … , Cn, if X,then Y. More readily understood are predictions made in text form,for example: if a person disagrees with a friend about his attitudetowards an object, then a state of psychological tension isproduced.PredictionWhilst explanation and prediction can reveal the inner workings ofphenomena, i.e. what happens and when, they do not alwaysprovide a sense of understanding of phenomena – how or
why they happen. A complete explanation of a phenomenon willrequire a wider study of the processes which surround thephenomenon and influence it or cause it to happen.Sense of understandingA good level of understanding of a phenomenon might lead to thepossibility of finding a way to control it. Obviously, not allphenomena lend themselves to this: for example, it is difficult toimagine how the disciplines of astronomy or geology could includean element of control. But all of technology is dependent on theability to control the behaviour, movement or stability of things.Even in society there are many attempts, often based on scientificprinciples, to control events such as crime, poverty, the economy,etc., though the record of success is more limited than in thenatural sciences, and perhaps there are cases of attempting theimpossible. The problem is that such attempts cannot be trulyscientific as the variables cannot all be controlled, nor can one becertain that all relevant variables have been considered. Thecrucial issue in control is to understand how certain variablesaffect one another, and then be able to change the variables insuch a way as to produce predictable results.ControlEvaluation is making judgements about the quality of objects orevents. Quality can be measured either in an absolute sense oron a comparative basis. To be useful, the methods of evaluationmust be relevant to the context and intentions of the research. Forexample, level of income is a relevant variable in the evaluation ofwealth, while degree of marital fidelity is not. Evaluation goesbeyond measurement, as it implies allotting values to objects orevents. It is the context of the research which will help to establishthe types of values that should be used.Evaluation
Types of researchThe different kinds of questions which instigate research requireapproaches to research that are distinguished by their theoreticalbackground and methodologies. A brief summary of various typesof research will illustrate the possibilities for your research efforts.Several major types of research can be identified, as in ChecklistBox 1.2. Writers differ in how they distinguish between them, andsome catalogue many more types than those listed.Checklist 1.2 Choose which type of research you will undertakeHistoricalComparativeDescriptiveCorrelationExperimentalEvaluationActionEthnogenicFeministCulturalIn order to help you choose, I will use these types as convenientoverall headings to explain their nature and include under them avariety of approaches which share some common features.HistoricalHistorical research has been defined as the systematic andobjective location, evaluation and synthesis of evidence in order toestablish facts and draw conclusions about past events (Gall, Galland Borg, 2006).It involves exploring the meaning and relationship of events, andas its resource it uses primary historical data in the form of historicartefacts, records and writings. It attempts to find out whathappened in the past and to reveal reasons for why and howthings happened. An interesting aspect of the values of historical
research as categorized by Hill and Kerber (1967), listed in Box1.3, is the relationship the past can have with the present andeven the future.Box 1.3 Values of historical researchIt enables solutions to contemporary problems to be sought in thepast.It throws light on present and future trends.It stresses the relative importance and the effects of theinteractions that are found within all cultures.It allows for the revaluation of data supporting selectedhypotheses, theories and generalizations that are presently heldabout the past.Historical evidence, consisting of primary historical data, must bescrutinized from two points of view. The first is to ascertainwhether the artefact or document to be studied is genuine. Therehave been many mistakes made in the past, either through a lackof analytical rigour by over-enthusiastic researchers, or throughfraud. (You might remember the Piltdown Skull, fraudulent skullbones which researchers long believed to be the ‘missing link’ inhuman history.) The second is to examine, in written evidence inthe form of historic documents etc., the authenticity of thecontents. What is the meaning of what is written, and howaccurate is it? For example, many authentic medieval texts areknown to be wildly inaccurate and vague in their descriptions ofevents.Figure 1.2 The first is to ascertain whether the artefact to bestudied is genuine
According to Gottschalk (1951), the questions of where, which,when and what are crucial in identifying the four aspects ofhistorical research which determine the scope of a study, asshown in Box 1.4.Box 1.4 Aspects of historical research that determine scope1. Where the events took place2. Which people were involved3. When the events occurred4. What kind of human activity was involvedThe degree to which an aspect is studied can be varied, i.e. thenumber of human activities examined can be increased ordecreased, the time-span covered can be extended or contractedetc. It must be remembered that the mere collection of historical
facts, or the setting up of chronologies of events, does notconstitute research. Although these are a necessary part ofhistorical research, an interpretation of the meanings and anassessment of the significance of the events are required.InterpretationHistorical research is not based purely on scientific method. Forinstance, the data used are seldom based on direct observation orexperimentation. But it should share many of the disciplines ofscientific method, such as objectivity and the desire to minimizebias and distortion, the use of scientific techniques such aschemical and radioactive analysis, and statistics. The problem forhistorians tends to be the paucity of information, while scientistsare often overwhelmed by it!All research students, whatever their chosen field of study, have toundertake a review of the literature. This is a study of what hasbeen done and written in the past, and so the principles ofhistorical research can be seen to be of direct relevance to thispart of their work.ComparativeComparative research is often used together with historicalresearch. Researchers compare people’s experience of differentsocieties, either between times in the past or in parallel situationsin the present. These studies can be on the macro level, e.g.studying the role of revolutions in class struggle, or on the microlevel, e.g. individual experiences in different types of marriage.It is often easier to understand phenomena when they arecompared with similar phenomena from another time or place.Culture and society rely heavily on what has gone before andoften use references from the past to justify the present. Theconstitution, the tax system, social mores are all rooted in theirown histories. Similarly, place also determines that phenomenadevelop differently.
Many social theories are presented as if the generalizations thatthey embody are valid for all times and places, when in fact theywere arrived at on the basis of limited contemporary Westernexperience. The study and comparison of differences help toreveal the origins and development of social phenomena, locatingthem in a certain time and place, and thus defeating claims thatthey are universal and atemporal.We can also learn by making comparisons both with the past andwith experiences elsewhere. It would be foolish for politicians tointroduce, say, sweeping changes to the electoral system, withoutcarefully studying the effects of such changes in the past and inother situations.Experimental research (described below), where the researchercan artificially control causal factors, is not really possible in socialresearch. However, the idea is put forward that history andcomparison can often supply the researcher with what is a naturalexperiment. According to Mill’s method of agreement (one of hisfive ‘methods of experimental enquiry’ devised in the nineteenthcentury), ‘If two or more instances of the phenomenon underinvestigation have only one circumstance in common, thecircumstances in which alone all the instances agree is the cause(or effect) of the given phenomenon’ (1973, p. 390). Using thistest it is possible to compare the suggested causes of severalinstances of a phenomenon (e.g. an industrial strike) andeliminate those that are not present in all instances as being non-essential to the occurrence of the phenomenon. For example,reasons for striking could be trade union power struggles, poorworking conditions, resistance to change, low pay, unfair labourrelations, etc. If, say, one cause only is present in all cases, e.g.unfair labour relations, then one could conclude that this is likelyto be the determining cause. One could then check to see if asituation where unfair labour relations did not result in a strikecould be found. If not, then this would support the foregoingconclusion.This kind of comparative exercise to explore and test causalfactors is an emblem of good research of this type, and helps to
overcome the fact that the researcher has no control over theavailable variables.DescriptiveInstead of examining records or artefacts, descriptive researchrelies on observation as a means of collecting data. It attempts toexamine situations in order to establish what is the norm, i.e. whatcan be predicted to happen again under the same circumstances.‘Observation’ can take many forms. Depending on the type ofinformation sought, people can be interviewed, questionnairesdistributed, visual records made, even sounds and smellsrecorded. The important point is that the observations are writtendown or recorded in some way, in order that they can besubsequently analysed. It is important that the data so collectedare organized and presented in a clear and systematic way, sothat the analysis can result in valid and accurate conclusions.The scale of the research is influenced by two major factors,identified in Box 1.5.Box 1.5 Influence on scale of descriptive research1. The level of complexity of the survey2. The scope of the surveyFor example, seeking relationships between specific eventsinevitably requires a more complex survey technique than aimingmerely to describe the nature of existing conditions. Likewise,surveying a large number of cases over a wide area will requiregreater resources than a small, local survey.In order both to save on unnecessary work and to give accurateinformation on the subject of your research, the sample of peopleor events surveyed (technically called the population) must becarefully chosen and delineated. To do this, it is necessary to beaware of the precise subject focus of the research so that specificobjectives can be formulated.
As descriptive research depends on human observations andresponses, there is a danger that distortion of the data can occur.This can be caused, among other ways, by inadvertently includingbiased questions in questionnaires or interviews, or throughselective observation of events. Although bias cannot be whollyeliminated, an awareness of its existence and likely extent isessential.CorrelationThe information sought in correlation research is expressed not inthe form of artefacts, words or observations, but in numbers.While historical and descriptive approaches are predominantlyforms of qualitative research, analytical survey or correlationresearch is principally quantitative. ‘Correlation’ is another word todescribe the measure of association or the relationships betweentwo phenomena.In order to find meaning in the numerical data, the techniques ofstatistics are used. What kind of statistical tests are used toanalyse the data depends very much on the nature of the data.This form of quantitative research can be broadly classified intotwo types of studies, as shown in Box 1.6.Box 1.6 Types of quantitative studies1. Relational studies2. Prediction studiesThe first is an investigation of possible relationships betweenphenomena to establish if a correlation exists and, if so, its extent.This exploratory form of research is carried out particularly wherelittle or no previous work has been done, and its outcomes canform the basis for further investigations.Prediction studies tend to be carried out in research areas wherecorrelations are already known. This knowledge is used to predictpossible future behaviour or events, on the basis that if there has
been a strong relationship between two or more characteristics orevents in the past, then these should exist in similarcircumstances in the future, leading to predictable outcomes.In order to produce statistically significant results, quantitativeresearch demands data from a large number of cases. Greaternumbers of cases tend to produce more reliable results; 20–30 isconsidered to be about the minimum, though this depends on thetype of statistical test applied. The data, whatever their originalcharacter, must be converted into numbers.One of the advantages of correlation research is that it allows forthe measurement of a number of characteristics (technically calledvariables) and their relationships simultaneously. Particularly insocial science, many variables contribute to a particular outcome(e.g. satisfaction with housing depends on many factors). Anotheradvantage is that, unlike other research approaches, it produces ameasure of the amount of relationship between the variablesbeing studied. It also, when used in prediction studies, gives anestimation of the probable accuracy of the predictions made. Onelimitation to what can be learned from correlation research is that,while the association of variables can be established, the causeand effect relationships are not revealed.ExperimentalExperimental research differs from the other research approachesnoted above through its greater control over the objects of itsstudy. The researcher strives to isolate and control every relevantcondition that determines the events investigated, so as toobserve the effects when the conditions are manipulated.Chemical experiments in a laboratory represent one of the purestforms of this research type.At its simplest, an experiment involves making a change inthe value of one variable – called the independent variable –and observing the effect of that change on another variable –called the dependent variable. (Cohen et al., 2017, p. 391)
Thus, the most important characteristic of the experimentalapproach is that it deals with the phenomenon of ‘cause andeffect’.However, the actual experiment is only a part of the researchprocess. There are several planned stages in experimentalresearch. When the researcher has established that the study isamenable to experimental methods, a prediction (technicallycalled a hypothesis) of the likely cause and effect patterns of thephenomenon has to be made. This allows decisions to be madeas to what variables are to be tested and how they are to becontrolled and measured. This stage, called the design of theexperiment, must also include the choice of relevant types of testand methods of analysing the results of the experiments (usuallyby statistical analysis). Pre-tests are then usually carried out todetect any problems in the experimental procedure.Only after this is the experiment proper carried out. Theprocedures decided upon must be rigorously adhered to and theobservations meticulously recorded and checked. Following thesuccessful completion of the experiment, the important task – thewhole point of the research exercise – is to process and analysethe data and to formulate an interpretation of the experimentalfindings.Not all experimental research has to, or even can, take place in alaboratory. The experimental methods used must take account ofhow much it is possible to control the variables. Writers oftextbooks on research have classified experimental designs indifferent ways. As an example, Campbell and Stanley (1966)make their categorization into four classes, as shown in Box 1.7,which can be regarded as a useful starting point for discussingtheir different characteristics.ClassesFigure 1.3 Not all experimental research has to, or even can, takeplace in a laboratory
Box 1.7 Classes of experiments1. Pre-experimental2. True experimental3. Quasi-experimental4. Correlation and ex post factoPre-experimental designs are unreliable and primitiveexperimental methods in which assumptions are made despite thelack of essential control of variables. An example of this is thesupposition that, faced with the same stimulus, all samples willbehave identically to the one tested, despite possible differencesbetween the samples.True experimental designs are those that rigorously check theidentical nature of the groups before testing the influence of avariable on a sample of them in controlled circumstances. Paralleltests are made on identical samples (control samples) which arenot subjected to the variable.
In quasi-experimental designs, not all of the conditions of trueexperimental design can be fulfilled. The nature of theshortcomings is however recognized, and steps are taken tominimize them or predict a level of reliability of the results. Themost common case is when a group is tested for the influence of avariable and compared with a non-identical group with knowndifferences (control group) which has not been subjected to thevariable. Another, in the absence of a control group, is repeatedtesting over time of one group, with and without the variable (i.e.the same group acts as its own control at different times).ReliabilityCorrelation design looks for cause and effect relationshipsbetween two sets of data, while ex post facto designs turnexperimentation into reverse, and attempt to interpret the natureof the cause of a phenomenon by the observed effects. Both ofthese forms of research result in conclusions which are difficult toprove and they rely heavily on logic and inference.EvaluationThis is a descriptive type of research specifically designed to dealwith complex social issues. It aims to move beyond ‘just gettingthe facts’ in order to make sense of the myriad human, political,social, cultural and contextual elements involved. The latest formof this type of research, named by Guba and Lincoln (1989) asfourth-generation evaluation, has, according to them, sixproperties, as in Box 1.8.Box 1.8 Properties of evaluation research1. The evaluation outcomes are not intended to represent ‘the waythings really are, or how they work’, but present the meaningfulconstructions which the individual actors or groups of actorscreate in order to make sense of the situations in which they findthemselves.2. In representing these constructions, it is recognized that they areshaped to a large extent by the values held by the constructors.
This is a very important consideration in a value-pluralisticsociety, where groups rarely share a common value system.3. These constructions are seen to be inextricably linked to theparticular physical, psychological, social and cultural contextswithin which they are formed and to which they refer. Thesesurrounding conditions, however, are themselves dependent onthe constructions of the actors which endow them withparameters, features and limits.4. It is recognized that the evaluation of these constructions ishighly dependent on the involvement and viewpoint of theevaluators in the situation studied.5. This type of research stresses that evaluation should be action-oriented, define a course that can be practically followed, andstimulate the carrying out of its recommendations. This usuallyrequires a stage of negotiation with all the interested parties.6. Due regard should be given to the dignity, integrity and privacy ofthose involved at any level, and those who are drawn into theevaluation should be welcomed as equal partners in everyaspect of design, implementation, interpretation and resultingaction. (Guba and Lincoln, 1989, pp. 8–11).There are a range of different approaches or evaluation models.Two of them are systems analysis and responsive evaluation.ModelsSystems analysis is a holistic type of research, which reverses thethree-stage order of thinking which is typical of scientific enquiry,i.e. breaking the problem or phenomenon to be investigated downinto researchable parts, then separately evaluating the parts, andfinally aggregating these evaluations into an explanation of thewhole. In systems analysis, there are also three stages, but theystart from appraising the whole, as in Box 1.9.Box 1.9 Stages of systems analysis1. Identifying an encompassing whole (system) of which thephenomenon or problem is a part2. Evaluating the behaviour or properties of the encompassingwhole
3. Explaining the behaviour or properties of the phenomenon orproblem in terms of its roles or functions within theencompassing wholeSystems analysis lends itself to creating understanding incomplicated situations, particularly those involving people andorganizations; such problems are often referred to as ‘messes’because of their indeterminate nature and large number ofinterconnected variables. Modelling and diagramming are two ofthe principal techniques used to describe systems.In the responsive evaluation model a series of investigative stepsis undertaken in order to evaluate how responsive a programme is(e.g. an advertising campaign, a new degree course or anexperimental traffic scheme) to all those taking part in it. Typicalsteps are shown in Box 1.10.Box 1.10 Steps in responsive evaluation1. Data collection: identifying issues from the people directlyinvolved in the programme; identifying further issues from theprogramme documents; observing how the programme isactually working2. Evaluation: the design of an evaluation based on the datacollected and reporting findings3. Suggesting changes: informing the participants of the findings inways specifically designed for each type of audienceA common purpose of evaluation research is to examineprogrammes or the working of projects from the point of view oflevels of awareness, costs and benefits, cost-effectiveness,attainment of objectives and quality assurance. The results aregenerally used to prescribe changes to improve and develop thesituation, but in some cases might be limited to descriptions givinga better understanding of the programme (Robson and McCartan,2016).Action
This can be seen as related to experimental research, though it iscarried out in the real world rather than in the context of a closedexperimental system. A basic definition of this type of research is:‘a small-scale intervention in the functioning of the ‘real’ world anda close examination of the effects of such an intervention’ (Cohenet al., 2017, p. 441).Its main characteristic is that it is essentially an ‘on the spot’procedure, principally designed to deal with a specific problemevident in a particular situation. No attempt is made to separate aparticular feature of the problem from its context in order to studyit in isolation. Constant monitoring and evaluation are carried out,and the conclusions from the findings are applied immediately,and further monitored. Action research depends mainly onobservation and behavioural data. As a practical form of research,aimed at a specific problem and situation and with little or nocontrol over independent variables, it cannot fulfil the scientificrequirement for generalizability. In this sense, despite itsexploratory nature, it is the antithesis of experimental research.EthnogenicIn this approach, the researcher is interested in how the subjectsof the research theorize about their own behaviour rather thanimposing a theory from outside. The test of success is that thesubjects themselves recognize the description of familiar featuresof their culture. As a process of studying human behaviour,according to Goetz and LeCompte (1993), the ethnogenicapproach has three characteristic features: it aims to represent aview of the world as it is structured by the participants underobservation by eliciting phenomenological data; it takes place inthe undisturbed natural settings of the subjects; and it attempts torepresent the totality of the social, cultural and economic situation,regarding the context to be equally important as the action.This is a difficult form of research for several reasons. As so muchof culture is hidden and rarely made explicit, the data beingsought by the researcher need to be pursued by delving deep intothe language and behaviour of the subjects of the study, and ofthe surrounding conditions in which they live. There is an ever-
present danger that the cultural background and assumptions ofthe researcher will unduly influence the interpretations anddescriptions made on the basis of the data collected. In addition tothis, there can be confusions produced by the use of languageand the different meanings which may be given to words by therespondents and researcher.The accounts of events in the past can never capture the infinitecontents of history. Historical knowledge, however wellauthenticated, is always subject to the biases and memory of itschronicler. It is also very difficult for one living in the twenty-firstcentury to understand a world outside the framework ofcontemporary beliefs, values and attitudes.Apart from these problems of interpretation of data, there is thefact that when working in a naturalistic setting, with social groupsengaged in everyday activities, it is impossible to repeat thesituation in order to verify the research. Social reality is not stable:a thing never ‘is’, as it is always changing into something else. It istherefore of great importance that multi-method and confirmatorydata sources are used to capture the moment.FeministFeminist research is a particular model of social research whichinvolves theory and analysis that highlight the differences betweenmen’s and women’s lives. It claims that researchers who ignorethese differences have invalid knowledge, as non-feministparadigms usually ignore the partiality of researchers’ ideas aboutthe social world. Value neutrality is impossible as no researcherpractises research outside his or her system of values and nomethods of social science can guarantee that knowledge isoriginated independently of values.No specific methods are seen to be particularly feminist, but themethodology used is informed by theories of gender relations.However, feminist research is undertaken with a politicalcommitment to the identification and transformation of genderrelations. This tends to reveal that this form of research is not
uniquely political, but rather exposes all methods of socialresearch to be political.CulturalMany of the prevailing theoretical debates (e.g.postmodernism, post-structuralism) are concerned withthe subjects of language and cultural interpretation, with the resultthat these issues have frequently become central to sociologicalstudies. The need has therefore arisen for methodologies thatallow analysis of cultural texts to be compared, replicated,disproved and generalized. From the late 1950s, language hasbeen analysed from several basic viewpoints: the structuralproperties of language (notably Chomsky, Sacks, Schegloff),language as an action in its contextual environment (notablyWittgenstein, Austin and Searle) and sociolinguistics and the‘ethnography of speaking’ (Hymes, Bernstein, Labov and manyothers).PostmodernismPost- structuralismHowever, the meaning of the term ‘cultural texts’ has beenbroadened from that of purely literary works to that of the manymanifestations of cultural exchange, be they formal such asopera, TV news programmes, cocktail parties, or informal such ashow people dress or converse. The main criterion for cultural textsis that one should be able to ‘read’ some meanings into thephenomena. Texts can therefore include tactile, visual and auralaspects, even smells and tastes. Three approaches to theconsistent interpretation of cultural texts can be mentioned herebriefly: content analysis, semiotics and discourse analysis.Content analysis was developed from the mid 1900s, chiefly inAmerica, and is a rather positivistic attempt to apply order to thesubjective domain of cultural meaning. A quantitative approach istaken by counting the frequency of phenomena within a case inorder to gauge its importance in comparison with other cases. As
a simple example, in a study of racial equality one could comparethe frequency of the appearance of black people in televisionadvertisements in various European countries. Much importanceis given to careful sampling and rigorous categorization andcoding in order to achieve a level of objectivity, reliability andgeneralizability and the development of theories.OrderSemiotics takes an almost opposite approach by attempting togain a deep understanding of meanings by the interpretation ofsingle elements of text rather than to generalize through aquantitative assessment of components. The approach is derivedfrom the linguistic studies of Saussure, in which he saw meaningsbeing derived from their place in a system of signs. Words areonly meaningful in their relationship with other words, e.g. we onlyknow the meaning of ‘horse’ if we can compare it with differentanimals with different features.This approach was further developed by Barthes and others toextend the analysis of linguistic-based signs to more general signsystems in any sets of objects:semiotics as a method focuses our attention on to the task oftracing the meanings of things back through the systems andcodes through which they have meaning and make meaning.(Slater, 1995, p. 240)Hence the meanings of a red traffic light can be seen asembedded in the system of traffic laws, colour psychology, codesof conduct and convention, etc. (which could explain why in Chinaa red traffic light means ‘go’). A strong distinction is thereforemade between denotation (what we perceive) and connotation(what we read into) when analysing a sign.Discourse analysis studies the way that people communicate witheach other through language within a social setting. Language isnot seen as a neutral medium for transmitting information; it is
bedded in our social situation and helps to create and recreate it.Language shapes our perception of the world, our attitudes andidentities. While a study of communication can be simply brokendown into four elements (sender, message code, receiver andchannel), or alternatively into a set of signs with both syntactical(i.e. orderly or systematic) organization and semantic (i.e.meaningful and significant) relationships, such simplistic analysisdoes not reflect the power of discourse.It is the triangular relationship between discourse, cognition andsociety that provides the focus for this form of analysis (van Dijk,1994, p. 122). Two central themes can be identified: theinterpretive context in which the discourse is set, and therhetorical organization of the discourse. The former concentrateson analysing the social context, for example the power relationsbetween the speakers (perhaps due to age or seniority) or thetype of occasion where the discourse takes place (at a privatemeeting or a party). The latter investigates the style and schemeof the argument in the discourse, for example a sermon will aim toconvince the listener in a very different way to a lawyer’spresentation in court.Post-structuralist social theory, and particularly the work of theFrench theorist Michel Foucault, has been influential in thedevelopment of this analytical approach to language. According toFoucault, discourses are ‘practices that systematically form theobjects of which they speak’ (1972, p. 43). He could thusdemonstrate how discourse is used to make social regulation andcontrol appear natural.Space does not allow the description of other types of research.Different disciplines, such as philosophy, theology and metaphysics,have types of research which are specifically suited to their purposes,but are beyond the scope of this book. It is important to point out thatthe above types of research are not generally mutually exclusive in aresearch project. More than one of these approaches may berelevantly used in order to achieve the outcomes aimed at in theresearch.
The research processWhichever type of research you choose, it will be useful tounderstand something of the process of research. This can helpyou to form a framework for your activities.Figure 1.4 Sitting down to write a 30,000 to 60,000 word thesis orresearch report is no simple taskSitting down to write a 30,000 to 60,000 word thesis or researchreport is no simple task. The research on which it is based doesnot develop in a linear fashion, any more than does the writing ofthe report itself. So how does one go about doing research? Youwill have undoubtedly noticed by now that the acquisition ofknowledge and the questioning of what to do with it is a complexprocess. From the numerous books on research methods, threeinterpretations of how the activities of research interweave with
each other have been selected, each viewing the process at adifferent level of detail.A simple summary of the relationships between five mainelements of the research process can be mapped (Diagram 1.1).This compact diagram stresses the circularity of the process andthe central role of research theory. Is it clear to you how progressis achieved, and at which point you can enter the system? Oneshould point out that this diagram makes research look a very tidyand logical process, but in reality you may find that it involvesguesses, intuition and intellectual cul-de-sacs.The spiral diagram that I have developed from the rather two-dimensional circular representation by Leedy and Ormrod (2015,p. 21) illustrates even more strongly the cyclical nature of theresearch process (Diagram 1.2). The division of the segmentsclearly indicates where you get on board. Notice how each turnthrough the spiral repeats the basic process. The knowledgegained and questions raised at each turn provide the basis for thenext cycle. To view research this way is to invest it with a dynamicquality rather than a one-time act – static, self-contained, and anend in itself. You will soon learn that genuine research oftencreates more problems than it resolves!The diagram developed from that of Newman (1989) concentrateson the first stages in the process. It shows a clear direction insequence of time, and displays how the process involvessuccessive widening and narrowing of knowledge bands (Diagram1.3). As each level of knowledge is achieved, the subject area isnarrowed down to become more specific, followed by subsequentwidening of knowledge as that specific area is researched indetail. This sequence of moving into more specific, yet morewidely researched subject areas could be extended right throughthe project, culminating in the specifically narrow conclusions andfinally widening out into recommendations which are of moregeneral significance.Diagram 1.1 The research process
Diagram 1.2 The research process (based on Leedy and Ormrod,2015, p. 21)Diagram 1.3 The research process (based on Newman, 1989, p.28)
Exercise 1.1Sketch the continuation of Diagram 1.3 using the following stages, andshow what gets rejected every time the subject is narrowed down:Definition of problem areaResearch into areaDefinition of research problemInvestigation into relevant concepts, theories and research methodsResearch proposalData-gathering and analysisFindings and conclusionsRecommendations
An alternative way of looking at it is as a series of stages that areinterrelated and are sometimes revisited in an iterative fashion during theproject (see Diagram 1.4). The teaching of research methods usuallyrelates to these stages and reflects the practical nature of the subject.To be able to design and plan your own research project you will haveto use your understanding of the process of research. The steps totake in planning the project will be explained later in this chapter.Desirable characteristics of research findingsThere is an untold mass of information in the world. By doingresearch, you will be adding to this plethora of information. Whatis it that will make your efforts worthwhile? What should thecharacteristics of your findings be to make your contributionvaluable? Reynolds (2016) identified four desirable characteristicsof scientific knowledge which we can use as a good guide and asa basis for discussion, as shown in Box 1.11.Box 1.11 Desirable characteristics of scientific knowledgeAbstractnessIntersubjectivity (meaning)Intersubjectivity (logical rigour)Empirical relevanceThe common thread between these is that the findings should berelevant to a wider sphere than the specific cases in yourresearch, and that they should be based on a research processthat is both accessible to and understandable by others. It is worthconsidering these characteristics in more detail.Diagram 1.4 The research process
The characteristic of abstractness is independence from aspecific time and place. Research findings are useful if they canbe applied in other situations, and can lead to the development ofgeneral theories. To discover the causes of a particularphenomenon that occurred in a particular time at a particularplace is of little general value if the knowledge gained is notrelevant to any other phenomena at different times and in differentplaces. There are two reasons for this.AbstractnessFirst, no future predictions about future events can be made usingthis knowledge, as the phenomenon can only be seen as a uniquehistorical event. As seen above, one of the important objectives ofresearch is to provide predictions about the future. Resulting fromthis lack of predictability is the inability to affect any control oversimilar future events.Secondly, by being restricted to a phenomenon in a particularplace, it will be impossible to generalize from the results of thisdiscovery to events which happen elsewhere.There are cases where the study of a particular event is bothuseful and unavoidable, for instance in historical and ethnographic
research. Historians are unlikely to feel competent to makepredictions of future events (e.g. election results) on the basis ofhistorical studies. The main aim of this kind of research is toanalyse, explain and gain a sense of understanding. With a betterunderstanding of a social phenomenon, interventions to alleviateproblems are likely to be more effective and have morepredictable outcomes. Similarly, in investigations following anaccident, the findings aim to explain events, understand theircauses and invite predictions: for example, a railway signallingfault discovered in an enquiry may cause more accidents if it isnot rectified.Intersubjectivity may be understood in two senses. First, toensure that everyone has the same understanding of words andevents there must be agreement as to the meaning of conceptsused in statements. This intersubjectivity of meaning, i.e.agreement between people about meaning, is attained by precisedefinition of concepts. Secondly, any statement describes therelationship of at least two concepts. Often, many connectedstatements are used in a research project to make predictions, orto explain a theory. To avoid ambiguity and disagreement aboutthe appropriate combination of statements to use, logical systemshave been evolved such as mathematics, statistics, symboliclogic. These are used to promote intersubjectivity, i.e. agreementabout use, at a logical level.IntersubjectivityIf scientists cannot agree on the predictions derived fromcombinations of statements, then there can be no agreementas to the usefulness of the statements for predicting orexplaining phenomena. (Reynolds, 2016, p. 15)Most of science and all technology is based on empiricalfoundations, i.e. built on, or guided by, the results of observationand experimentation. The basic purpose of a scientific theory is toexplain what causes an event or why one event is associated withanother. The basis for these explanations is the recorded
measurements made by the researcher of the events. Empiricalrelevance is a measure of the correspondence between aparticular theory and what is taken to be objective empirical data,which enable other scientists to verify the results of the researchfor themselves. The greater the relevance of the empirical data,the more confidence can be put in the veracity of the theory.Empirical relevanceStarting your own researchThe common element in student academic research at everylevel, from undergraduate to doctorate, is that it is an exercise inthe doing of research. The student will have to demonstrateknowledge of research theory and methods and the ability toapply these in an appropriate and successful manner relevant tothe chosen topic. You might consider that the topic itself servesmerely as a vehicle in order to make this demonstration possible.That is perhaps too cynical a view. The topic must be the drivingforce behind the project and, particularly at PhD level, theresearch must make some contribution to knowledge about thesubject. But without a proper understanding of research and itsapplication, this knowledge will not be discovered.But perhaps you are embarking on a research project as part ofyour work. Most of the issues that you will face will be similar tothose faced by academic researchers. The major differencesmight be the greater resources available to you, the lack of accessto supervision and advice, and the stresses of work in aprofessional context. What will be the same, however, is therequirement that the research has clear and achievable goals andis carried out efficiently using the appropriate research methods.Finding and defining a research problemIt should be evident from what you have read so far that in orderto carry out research, you need to start by identifying a questionthat demands an answer, or a need that requires a resolution, or a
riddle that seeks a solution, which can be developed into aresearch problem: the heart of the research project.Students starting their research degree course, and practitionerswishing to become involved in research, tend to come from widelydifferent backgrounds, and are equipped with varied amounts ofknowledge and degrees of experience in their chosen field ofstudy. While most are fairly sure of the subject they want toresearch, many are uncertain of the exact problem they wish toaddress.One of the first tasks, therefore, on the way to deciding on thedetailed topic of research is to find a question, an unresolvedcontroversy, a gap in knowledge or an unrequited need within thechosen subject. This search requires an awareness of currentissues in the subject and an inquisitive and questioning mind.Although you will find that the world is teeming with questions andunresolved problems, not every one of these is a suitable subjectfor research. So what features should you look for which couldlead you to a suitable research problem? Checklist 1.3 lists themost important.Figure 1.5 The world is teeming with questions and unresolvedproblems
Checklist 1.3 Features of a suitable research problemIt should be of great interest to you. You will have to spend manymonths investigating the problem. A lively interest in the subjectwill be an invaluable incentive to persevere.The problem should be significant. It is not worth time and effortinvestigating a trivial problem or repeating work that has alreadybeen done elsewhere.The problem should be delineated. Consider the time you haveto complete the work, and the depth to which the problem will beaddressed. You can cover a wide field only superficially, and themore you restrict the field, the more detailed the study can be.You should also consider the cost of necessary travel and otherexpenses.You should be able to obtain the information required. Youcannot carry out research if you fail to collect the relevantinformation needed to tackle your problem, either because youlack access to documents or other sources, and/or because you
have not obtained the cooperation of individuals or organizationsessential to your research.You should be able to draw conclusions related to the problem.The point of asking a question is to find an answer. The problemshould be one to which the research can offer some solution, orat least the elimination of some false ‘solutions’.You should be able to state the problem clearly and concisely. Aprecise, well thought out and fully articulated sentence,understandable by anyone, should normally clearly be able toexplain just what the problem is.It is not easy to decide on and define a research problem, and youwill not be expected to do so immediately. The important thing, atthis stage, is to know what you are looking for, and to explore yoursubject for suitable possibilities.The problem can be generated either by an initiating idea, or by aperceived problem area. For example, investigation of‘rhythmic patterns in settlement planning’ is the product of an ideathat there are such things as rhythmic patterns in settlementplans, even if no one has detected them before. This kind of ideawill then need to be formulated more precisely in order to developit into a researchable problem. We are surrounded by problemsconnected with society, the built environment, education, etc.,many of which can readily be perceived. Take, for example, socialproblems such as poverty, crime, unsuitable housing anduncomfortable workplaces, technical problems such as designdeficiencies, organizational problems such as business failuresand bureaucratic bungles, and many subjects where there may bea lack of knowledge that prevents improvements being made,such as the influence of parents on a child’s progress at school, orthe relationship between designers and clients. Obviously, it is notdifficult to find problem areas. The difficulty lies in choosing anarea that contains possible specific research problems suitable forthe subject of a research project or degree.Problem areaSome common mistakes
It is worth warning you at this stage of some common mistakesmade when a research problem is chosen. These mistakes arisemainly from the failure to grasp the necessity for the interpretationof data in the research project; Checklist 1.4 shows four commonmistakes.Checklist 1.4 Common mistakes when choosing a research problem1. Making the choice of a problem an excuse to fill in gaps in yourown knowledge. We all welcome the chance to learn more forourselves, but the point of research is not just personalenlightenment, but making a contribution to public knowledge.Anyone can find a problem that involves the gathering andduplication of information, but it requires an additional effort tofind one that requires data to be analysed and conclusions to bedrawn which are of wider interest.2. Formulating a problem that involves merely a comparison of twoor more sets of data. A comparison of sets of data or recordsmight fill up many pages (e.g. the average age of marriagethrough the centuries), but without any effort to reveal somethingnew from the information, there is no research activity. Theproblem should clearly state the objectives behind making thecomparison.3. Setting a problem in terms of finding the degree of correlationbetween two sets of data. Comparing two sets of data to revealan apparent link between them (e.g. the average age ofmarriage and the size of families) might be interesting, but theresult is only a number, and does not reveal a causalconnection. This number, or coefficient of correlation, revealsnothing about the nature of the link, and invites the question – sowhat?4. Devising a problem to which the answer can be only yes or no.In order to improve on our knowledge of the world we need toknow why things are as they are and how they work. A yes–nosolution to a problem skirts the issues by avoiding the search forthe reasons why yes or no is the answer, and the implicationswhich the answer has.Exercise 1.2Consider the following short sentences claiming to be research problemsand decide whether they are researchable, and are a feasible proposition
for an individual student, like yourself, to undertake for a research degreeor as a research project. Respond first with the answers ‘yes’, ‘no’ or‘possibly’. Then, if you think that the research problem is not viable or willpresent difficulties, briefly give your reasons.1. An enquiry into the history of the building of the Channel Tunnel2. A study to compare the results in school history exams for 16-year-olds throughout Europe between 1970 and 19803. The effects of parent unemployment on their children’s attitude toschoolwork4. The relationship between temperature, humidity and air movement inthe cooling effect of sweating on the human skin5. The effects of using glass of different thickness and qualities insingle, double and triple glazing6. What factors must be evaluated, and what is their relativeimportance, in constructing a formula for allotting grants to universitystudents in Scotland7. An analysis of the influence of Palladio’s villa designs on largecountry houses built in Britain in the eighteenth century8. Whether the advantages of foreign borrowing by developingcountries outweigh the disadvantages9. The composition of prefabricated elements of buildings in theconstruction of multistorey car parks in tight urban situations in largeconurbations of the United States of America during the 1970s10. A study of how hospital patients’ recovery is affected by the colour oftheir surroundings and of how they react to the effects of differentlight levels after major operations11. An enquiry to identify and evaluate the causes of ‘sick buildingsyndrome’ in order to indicate possible methods of avoiding theoccurrence of this ‘syndrome’ in new buildings12. The impact of local tax and exaction policies on the Londoncommercial office sector13. Economic implications of the programme of rental increases andhousing sales in China14. How the career plans of school leavers compare with theirsubsequent careers in terms of self-satisfaction and self-adjustment,and what information the analysis of the difference between plannedand realized careers provides to assist in career planningAs you can see, it requires a good deal of thought and knowledge ofyour chosen topic of study in order to isolate a suitable researchproblem. Unless you have come to do your research with a particular
detailed problem already identified (probably following on from someprevious research which you have done), you will need to narrowdown to a specific problem from a wider problem area.Aids to locating and analysing problemsBooth, Colomb and William (2016, Chapter 3) suggest that theprocess for focusing on the formulation of your research problemis as shown in Checklist 1.5. As you can see, they recommendthat, apart from simply narrowing down the object of study, youshould carefully scrutinize the resultant topic in the light of whatyou have found out in your background studies.Checklist 1.5 How to focus on a research problem1. Find an interest in a broad subject area (problem area).2. Narrow the interest to a plausible topic.3. Question the topic from several points of view.4. Define a rationale for your project.Initially, it is useful to define no more than a problem area, ratherthan a specific research problem, within the general body ofknowledge that interests you, e.g. housing and homelessness,parks in cities, building regulations and historic conservation. Youraim should be to subsequently narrow down the scope of the ideaor problem until it becomes a highly specific research problem.This narrowing process will require a lot of background reading inorder to discover what has been written about the subject already,what research has been carried out, where further work needs tobe done and where controversial issues still remain.You should keep in mind three questions when engaged in thepreliminary exploratory work. The first is, what is your motivationfor doing the research? A major motivation should be curiosityabout the research results. Another will undoubtedly be thefulfilment of the requirements of a research degree. Learningabout the process of research – practical knowledge that can beused in the future – is also likely to be a motivation. The choice ofproblem is likely to be influenced by these motivational factors.
The second question is, what relevant interest, experience orexpertise do you bring to bear on the subject? Obviously, interestin a subject is essential if you are to concentrate happily on it for ayear or more. Although experience or expertise in a subject is nota prerequisite to doing research in that field, it does have an effecton the preliminary and information-gathering stage of the work, asyou will be familiar with the literature and the potential problemareas. However, a ‘new light’ may be cast on a subject bysomeone looking at it with ‘fresh eyes’.Figure 1.6 What are you going to produce?The third question is, what are you going to produce? As aresearcher, your priority will be to produce a defendable thesis oruseful research report within your time limit. If you are a researchstudent, you should check the requirements of your university orcollege in the regulations issued about the nature of suitableresearch topics. (It might be a good idea to do that now. You will
find the information in the latest university research degreeregulations kept in the library. You should also be issued with yourown copy.) If you are doing a dissertation as part of a course,check the course notes for guidance. If you are doing a fundedresearch project, then you will need to know the requirements ofthe likely funders or of the policy of the organization for which youwork.Initial literature review and defining theproblem areaThe objective of the initial review of the literature is to discoverrelevant material published in the chosen field of study and tosearch for a suitable problem area.Fox (1969) mentions two kinds of literature that should bereviewed. The first is ‘conceptual literature’. This is written byauthorities on the subject you have in mind, giving opinions, ideas,theories or experiences, and published in the form of books,articles and papers. The second is ‘research literature’, whichgives accounts and results of research that has been undertakenin the subject, often presented in the form of papers and reports.Chapter 2 in this book tells you how you can effectively carry outthis search through the literature.As every piece of research contributes only a small part to agreater body of knowledge or understanding, researchers must beaware of the context within which their research work is to becarried out. At this stage it is important to get an overview of thesubject, rather than knowledge in depth. This will provide you withan understanding of the principal issues and problems orcontroversies, and the opportunity to select a problem area withina frame of reference. Within this problem area, it is important thatyou familiarize yourself with those aspects that have already beenwell established by previous research, and are generally acceptedas true. These ‘truths’ can then be assumed to need no furtherproof, and the research problem simply uses them. It is notpossible for a researcher to question absolutely everything inhis/her investigations. Alternatively the research problem can bein the form of a challenge to veracity of one or more of these
‘truths’. Advances in wisdom are only made by building on thesolid foundations of previous knowledge. Obviously, someonewho is already familiar with the subject investigated will tend to bequicker to advance through this stage.At this early stage in your research programme you are exploring yoursubject field only to identify a problem area, and do not need to try todefine your research problem in any detail. All the same, I think it isuseful to know what the next steps will be so that you can see thedirection in which you will be moving. This might well help you tochoose a problem area.Research problem definitionFrom the interest in the wider issues of the chosen subject, andafter the selection of a problem area, the next step is to define theproblem more closely so that it becomes a specific researchproblem, with all the characteristics already discussed. This stagerequires an enquiring mind, an eye for inconsistencies andinadequacies in current theory, and a measure of imagination. It isoften useful in identifying a specific problem to pose a simplequestion, for example, ‘Does the presence of indoor plants affectpeople’s frame of mind?’ or ‘How can prevention measuresreduce vandalism?’ or ‘Can planning and building regulationsprevent the destruction of indigenous architecture?’Such a question can provide a starting point for the formulation ofa specific research problem, whose conclusion should aim toanswer the question. At this stage, the nature of the question willgive some indication of the type of research approach (orapproaches) that could be appropriate. Will it be a historical studyor a descriptive enquiry, an analysis of correlations or anexperimental exercise, or a combination of more than one ofthem? Seemingly simple questions are riddled with ambiguities,which must be cleared up by careful definition. For example, in theabove questions, what does ‘frame of mind’ mean, what sort of‘prevention measures’ are envisaged, and does the question
embrace all types of ‘indigenous architecture’ everywhere? It islikely that the problem is too broad if you can state it in less thanhalf a dozen words. A few additional questions posed againsteach word can help to delineate the problem – where, who, what,which, when? Break the problem down into short sentences, notworrying at this stage about the overall length of the problemstatement. It is a useful trick to put each sentence on a separateslip of paper, so that they can be put into order in differentsequences. When the best logical progression from sentence tosentence is achieved, the statement can be edited into a moreelegant form. (Chapter 5 deals in more detail with the techniquesof the problem statement.)While developing a specific research problem, keep in mind theskills you will require to carry out the research posed by theproblem. Fox (1969, p. 39) defines five types of skill which areessential: research design, instrument development, datacollection, data analysis and research writing.Designing research can be learned, in consultation with your tutoror supervisor (just wait till Chapters 5 and 6). Instrumentdevelopment is, however, a highly specialized skill, so it isadvisable to formulate the problem so that you can usestandardized or previously developed instruments. The skillsrequired by data collection techniques are generally readilyacquired (introduced in Chapter 7), though consideration must begiven to the extent of data needed. Data analysis does requirespecialist skills, which can be of a highly sophisticated nature(specialist help is on hand when you get that far). It will definitelybe worth your while to consult your tutor or supervisor on theimplications for data analysis that the research problem mighthave. Skills in research writing will be developed in Chapter 9, andby consultation with your tutors or supervisors over the nextmonths (or years). Careful consideration of these points willensure that the planned research is practicable and has a goodchance of success.The sub-problems
Most research problems are difficult, or even impossible, to solvewithout breaking them down into smaller problems. The shortsentences devised during the problem formulation period can givea clue to the presence of sub-problems. Does one aspect haveto be researched before another aspect can be begun? Forexample, in one of the research questions asked above – thekinds of prevention measures that can be used against vandalism– how the measures can be employed and for what types ofvandalism they are suitable, will have to be examined. The sub-problems should delineate the scope of the work and, takentogether, should define the entire problem to be tackled assummarized in the main problem.Sub-problemsFollowing on from their recommended steps for narrowing downthe scope of your study to one topic, as shown in Checklist 1.5,Booth et al. (2016, p. 40) elaborate on how you can organize yourquestions to define the sub-problems by looking at your topic fromthe four perspectives shown in Checklist 1.6. It is interesting tonote that the usefulness of the topic is also an issue that shouldbe taken into account – but does this exclude blue-sky research? Ihope not!Checklist 1.6 Questions used to define sub-problems1. What are the parts of your topic and what larger whole is it a partof?2. What is its history and what larger history is it a part of?3. What kind of categories can you find in it, and to what largercategories of things does it belong?4. What good is it? What can you use it for?Second review of literatureA more focused review of the literature follows the formulation ofthe research problem. The purpose of this review is to learn about
research already carried out into one or more of the aspects of theresearch problem, as shown in Box 1.12.Box 1.12 Purposes of a literature review1. To summarize the results of previous research to form afoundation on which to build your own research2. To collect ideas on how to gather data3. To investigate methods of data analysis4. To study instrumentation that has been used5. To assess the success of the various research designs of thestudies already undertakenA full introduction to the techniques of literature review,information storage and information retrieval is given in Chapter 4.Exercise 1.3In order to reflect on what you have learned about the characteristics ofthe research problem and how it should be presented, here is part of aresearch proposal written by a postgraduate research student. It aimed todescribe accurately and succinctly the relevant background, the problemto be researched and its importance. Obviously, you are not required atthis stage to write anything as detailed as this yourself. The point of thisexercise is for you to examine this text to see how a research problem canbe extracted out of a context and defined and described in such a way asto convince the reader that the project is both worthwhile and possible tocarry out.After reading the following short research proposal, check the reportagainst the following criteria:1. Is the research problem clearly stated? What is it? Write it out. If it isnot clear, try to detect what it probably is and then summarize it.2. Does the problem seem to arise naturally from the backgroundinformation and questions? Summarize the main points of theargument which lead up to the problem. If you have difficulty findingthe relevant background information and argument, explain whereyou see the gaps.3. Are any sub-problems stated? If so, what are they? Write them out.Do they really form parts of the main problem?
4. Is the proposed research limited in scope? What are the limitations?(It will help you if you think of different aspects of the research, e.g.time, place, etc.)5. Did the researcher state what type of research approach would beused? If so, write a summary of the research activities to beundertaken.6. Is there any indication of the importance of the study? Describe how,if at all, this is conveyed.7. Is there any reference to, or discussion of, related literature orstudies by other researchers? If so, which?Research proposalA study of group-living accommodation foryoung physically disabled peopleThe aims of this study are to investigate different forms of group-livingaccommodation designed for people with physical disabilities; and toevaluate their effectiveness in meeting requirements for independentliving, particularly for young severely disabled people.The ethos behind segregation of disabled people has been that those whoare incapable of managing their own lives might reasonably be placed ininstitutions that can take over those responsibilities. Admission into suchinstitutions has for a long time implied, by circumstance or design, arelinquishment of certain rights, most particularly that of independentliving.As distinctions between those who are dependent on others have becomemore clear – the poor, sick, old and abandoned – so institutions andbuildings, such as workhouses, orphanages and asylums, have evolved toprovide for them. Their common ethos was segregation. After World WarII, that acceptability of segregated institutions was called into question andalternatives to institutional living were sought for those dependent onothers for their care. The response of the caring institutions was to shiftaway from segregation and towards the integration of people withdisabilities into mainstream society. The underlying problem for architectswas how buildings would need to change to accommodate this shift.Architects needed to devise a diversified range of buildings that widenedthe options for independent living for people with differing degrees ofdisability.In the 1950s and 1960s new building forms such as sheltered and specialneeds housing were developed, but these were predominantly for the
elderly. For younger disabled people there continued to be few alternativesbetween admission to an institution or staying at home.However, by 1970 new concepts were developed; most striking wereyoung disabled units (YDUs) for severely disabled people of working agewho had to leave home.Over 320 YDUs and similar buildings have been built in the last twodecades, providing places for 10,500 people. Some are built in thegrounds of hospitals and some in the community; they generallyaccommodate 30 residents with their own bedroom and shared commonfacilities. Their objective has been to meet requirements for independentliving, across the age range of residents, from school leaving age toretirement. However, research on the effectiveness of these schemes issparse. Investigation so far suggests that their design has been moresuccessful at accommodating the needs of older residents and lesssuccessful at accommodating the requirements for independent living ofyounger disabled people.The focus of this study will therefore be to investigate the influences ondifferent YDU built forms, and evaluate their effectiveness in meeting theindependent-living needs and aspirations of the young people with severedisabilities who live in them.Indicators of independent living established early in the study will be usedto measure the effectiveness of independent living attained in the differentbuilding types, all purpose-designed to wheelchair parameters. Data willbe collected by undertaking detailed multi-method surveys of differentYDU-type group-living schemes. The surveys will include detailedappraisal of plans and measurements of buildings, observation of thebuilding in use and structured interviews with residents across the agerange.The findings of the study are intended to make an original contribution toresearch in this area, and provide recommendations of practical value forthe design of independent-living schemes which set out to optimize theindependence of young people with severe physical disabilities.(Proposal by David Bonnett – who successfully completed his PhD threeyears later.)Are you finding it quicker to analyse a given text now? The examplegiven above is the first part of a research proposal for an MPhil with
intention to transfer to a PhD. What has been left out in this exampleis the detailed methodology, explaining exactly how the research willbe carried out. As already mentioned, you are not expected to be ableto write anything as detailed as that at this stage. However, afterChapter 7, you should be able to write something comparable and thisextract gives you some idea of what you are aiming at. Of course,your subject may be completely different, but the criteria listed abovewill be the same.Planning a research projectThe purpose of the research plan is to take the initial researchproblem and decide how it will be researched. A clearly definedand expressed research problem is one important prerequisite forevolving a research plan. Important facts to be considered whendesigning the project are: available time, financial resources,facilities, availability of data, possible methods of analysis, andyour own developing skills as a researcher.Remember that you do not have a team of researchers to supportyou, and that you have only a few weeks to complete adissertation, about one year to complete an MPhil or about threeyears to complete a PhD. All other research projects are similarlylimited in their time-frame. There will be some hard choices tomake; however fascinating your subject and however importantthe expected outcomes, it is essential to limit the area of yourinvestigation and keep it within manageable proportions. Keep inmind that working towards a research degree is also a trainingexercise to develop research skills, and your thesis will finallydemonstrate that you have acquired them sufficiently.What sort of research will you pursue? It is worth rememberingthe different overall aims that could be at the centre of yourproject. Phillips and Pugh (2015, pp. 58–9) identified three basicaims of research, as shown in Checklist 1.7 and discussed in thetext that follows.Checklist 1.7 What will be the basic aims of your research?Exploration
Testing outProblem-solvingExplorationThis kind of research delves into the unknown, tackling newproblem issues or topics. As little or no previous research hasbeen done on those topics, it will be impossible to delineateprecisely the scope of the research or to predict its outcomes.Because it will be in a relatively unexplored domain, a necessarypart of the research is to explore what existing theories, conceptsand methodologies might be used or adapted, or failing those, todevise new ones. It pushes out the boundaries of knowledge inthe anticipation that the outcomes will be of value.Testing outA common feature of such research is that it makesgeneralizations from specific instances. But how far are thegeneralizations valid? Testing out research explores the validity ofthe generalizations in other circumstances, and tries to definetheir limits. This basic scientific activity leads to the refinement oftheories. There are a host of opportunities in this approach:testing the generalizations in different locations, under differentsocial or physical conditions, in different contexts, etc.Problem-solvingThis type of research identifies a ‘real-life problem’. Its aim is tofind possible solutions to the problem by using techniques ofsystematic appraisal and analysis. As ‘real-life problems’ tend tobe complex, the study might involve several disciplines and avariety of methods, requiring a great deal of backgroundknowledge. Although it is possible to pursue this kind of researchon a theoretical level, commonly practical benefits flow from it.However, solutions are unlikely to be obvious and clear-cut.
Which type of research lends itself best togaining a research degree?Phillips and Pugh (2015) pragmatically suggest that the safestoption is to be recommended, that is, the one with the fewestunknown factors. Testing out research, based on known theoriesand established methods of testing, avoids the unpredictability(though it might miss out on the excitement) of the other tworesearch approaches. It is probably better to keep your feet on thebottom until you are able to swim! You will still have to introducesome new insights or methods into the subject to make theresearch worthwhile (rather than just replication), and it can beargued that this mainstream type of research will usefully producemore readily publishable and quotable results than the other twotypes.There are greater risks and unknowns in the exploratory andproblem-solving approaches. They undoubtedly require moreexpertise and experience on the part of the researcher anddemand the enthusiastic support of the supervisor. In suchinnovative and original research, it is more difficult to achieve theauthority in the subject required for it to be publishable, whichmight, in turn, impede a career in research.You should now think about your own research interest and reflect onwhich of the above types of research might be considered to beappropriate for your own work.Your project planAny research project requires planning so that the researcher’stime is used efficiently in pursuance of the research objectives.Much effort can be wasted and frustration incurred by haphazardreading and collecting of notes and references, sundry informationand opinions. This form of activity might be ‘very interesting’, butleads in no particular direction and hence does nothing to
advance the progress of the research. Similarly, concentrating onthe research methodology will not lead to the formulation of aviable plan of action.Do not confuse overall research planning with researchmethodology. Whereas the general approach to planning aresearch study may be similar across disciplines, the techniquesone uses to collect and analyse data – that is, the methodology –may be specific to a particular academic discipline (Leedy andOrmrod, 2015, p. 93).According to Leedy and Ormrod (2015), all research has a basicformat. Whatever subjects or disciplines are its focus, they allshare the need of a central research problem, a search for andcollation of data, appropriate methods of analysis and theformulation of substantiated conclusions. This is not to say thatthe methodology will be similar in all disciplines. On the contrary,much of the planning of research projects is taken up withdeciding on the most appropriate techniques for data collectionand analysis. The underlying dynamics of the process also includethe features of the researcher, such as:motivationexperience and skills.They also include aspects of the research situation, such as:costtimefacilities and equipmentsituation and locationneeds and demands of the respondents or others cooperatingin the research.Boxes 1.13–1.15 give three examples of research plans, two forPhD theses and the third for a funded research programme. Notethat references in these plans are not included in the reference listfor this book.Box 1.13 Example research plan 1
Oxford city primary care group: a casestudy of interagency collaborationPromoting independence in older peopleFor this part of the study, fracture of the neck of the femur (hipfracture) will be used as a tracer condition. Care provision for thisgroup potentially involves the whole system of health and social care,including prevention (accident reduction), trauma, rehabilitation,primary care, continuing community care, social services, thevoluntary sector, carers, day centres, residential care etc. If the PCGis to have an impact on interagency collaboration, it should beapparent in the treatment of this care group.The study will seek to identify the impact of the PCG on interagencycollaboration from the perspectives both of those at management leveland of service users. Semi-structured interviews will be conductedwith a purposive sample of representatives at middle managementand operational levels of the agencies involved. A SWOT (strengths,weaknesses, opportunities and threats) framework will be used. Asampling frame is being composed by asking senior people in relevantagencies to identify who in their organization would be the mostappropriate person to interview in terms of their proximity to interfaceswith other agencies. These will be followed by semi-structuredinterviews to explore the experiences of people who have returnedhome following a hip fracture and the experiences of their carers. Thiswill provide contextual data to corroborate (or not) the accounts of the‘professionals’. The methodological approach is to investigate theperceptions of professionals and users, including comparingexpectations with actuality one year on, rather than measuringimpacts directly. Service users and their carers will be accessedthrough the A&E department at the John Radcliffe Hospital. MedicalEthics Board approval will be sought. Fieldwork will be carried out attwo time points: at time 1 (April–June 2000) and at time 2 (April–June2001).Evaluation of the PCG’s substancemisuse servicesThis project is a PCG initiative that addresses one of its five statedpriorities: tackling the city’s drug and alcohol problems. The PCG hasput in post a Development Officer, Oxford Community SubstanceMisuse Services, for a year from January 2000 with a budget of
£100,000 to develop the initiative. The criteria measured in thepresent study will be largely the objectives of the initiative, and willtherefore be specified in consultation with its steering group.As with the previous part of the study, semi-structured interviews witha purposive sample of key informers at management and practitionerlevels of the relevant agencies will be conducted. A SWOT frameworkwill be used. Service users will also be interviewed to assess theimpact of the PCG initiative on their experiences of care. They will beaccessed through the street drug agency, Libra. Using the agency’sdirector as a ‘gatekeeper’, clients will be invited to take part in thestudy on a voluntary basis. The approval of the Medical Ethics Boardwill be required. Fieldwork will be carried out at two time points: attime 1 (April–June 2000) and at time 2 (April–June 2001).Monitoring structural changeThe third element of the study will investigate the PCG’s impact onpartnership working and monitor structural change in the PCG in threeways: through interviews, postal surveys and document analysis.Preliminary analysis of the interviews will inform the construction oftwo postal questionnaires. Comparisons will be made with the findingsof the National Tracking Project which is carrying out a national surveyof a 15% sample of PCGs, part of which will be looking particularly atthe development of partnerships with local authorities. A theme ofparticular interest will be the impact of the introduction of the PCG onformer fundholding GPs.1. Interviews. Each 6 months (June and November) interviews willbe carried out with key informers at the interfaces betweenagencies. A SWOT framework and/or Stephen Peckham’s(1999) separation–integration interview schedule will be used.Interviewees representing constituents of as many agencies aspossible will be interviewed.2. Postal questionnaires. The first questionnaire will be sent only toex-fundholders, and will aim to establish the prevalence of theexperiences expressed in the interviews with fundholders. Thedesign of the second questionnaire will be informed by thepreliminary analysis of all the interviews, and will be sent to eachpractice in Oxford. Support from the PCG will be provided toencourage the return of questionnaires.3. Document analysis. Ongoing analysis of PCG documents will becarried out. These documents will include the agendas andminutes of board meetings, subgroup meetings (HealthImprovement Plan Group, Communications and Public
Involvement Group, Mental Health Working Group, ClinicalGovernance) and other meetings, consultation exercises etc.The collection of baseline data during the first round of fieldwork willconstitute the MPhil stage of the study, while the PhD stage willcomprise the collection of data a year on and the analysis of how thePCG has impacted on interagency collaboration during its first periodof activity.(Alison Chisholm, Oxford Brookes University)Box 1.14 Example research plan 2Factors affecting the teaching andlearning of energy in science at key stage3Proposed planStage 1: Survey to find out about approaches to teaching energy atKS3. The focus will be curriculum organization, resources used andteachers’ subject specialization. This will be done initially with a semi-structured questionnaire aimed at heads of science. It is hoped toquickly follow this up with a short informal interview asking aboutopinions of the teaching of energy and the possibility for furtherresearch with science teachers in the schools. It is aimed initially tocontact a random sample of 100 secondary schools from a selectionof rural and urban settings over several shire counties andmetropolitan districts. This stage will be done as an initial contactingexercise and ice-breaker which will accumulate a useful base ofinformation on how energy is taught in schools. A semi-structuredquestionnaire is appropriate as the information asked for is fairlysuperficial and organizational and not of an emotive, sensitive nature.Existing contacts in the School of Education will be used initially. Datawill be analysed using a spreadsheet, e.g. Excel.Stage 2: Upon securing contact with some volunteer teachers, thisstage will be a short exercise with the aim of working with teachers toelicit and prioritize factors they feel important, in their experience, ineffecting conceptual understanding. Elicitation will be done through aconcept mapping exercise. Ranking and prioritizing will be donethrough a response scale to gauge actual perceived importance,triangulated with a prioritizing grid to gauge relative importance. This
stage will be done to gain a slightly deeper insight into teachingpractice within the curriculum framework established in stage 1; tofurther working relationships with schools; and to get to know teachersand open up the possibility for in-depth case study work, thusproviding a start for an action-research-based methodology byencouraging reflection in teachers and the researcher. It is aimed towork with twenty teachers. Data will be analysed with a spreadsheetby triangulating ranked and prioritized factors. The resulting individual‘picture’ of factors important to each teacher is intended to provideuseful material for reflection on conceptual understanding. ‘Pictures’for all the teachers as a group may be examined for evidence ofcorrelation. It is intended to pilot this stage with a group of PGCEstudents in order to refine the method of concept mapping andfacilitate reflection and also gain 35 further useful contacts.Stage 3: Case studies working with eight to ten teachers and theirpupils. The aim will be to assess the effectiveness of teaching energyunder particular sets of identified and reflected-upon groundedfactors, as elicited, ranked and prioritized in stage 2. This will be doneby working with teachers and their pupils from year 7, 8 or 9 classes,closely following a sequence of lessons on energy, or an energyrelated topic. Effectiveness will be assessed from the point of pupils’ apriori and a posteriori conceptual understandings. This will be done byfirst negotiating a set of criteria for pupils’ expected outcomes for thesequence of lessons. These will contain elements of the school’sscheme of work and hence National Curriculum references. Also, theexpected level of understanding of the energy concept will be clearlypredefined, such as in the model provided. However, it is anticipatedthat, given the complex nature of the energy concept coupled with therecent shift in its description, some teachers may need to challengetheir existing ideas about energy before the teaching sequencebegins. Hence it would be favourable to interview teachers in-depth toelicit any misconceptions and provide means for INSET (usingmethods and materials from the PSTS project). Pupils will be testedon their understandings of energy before and after the lessons usingstructured and open-ended questions. A sample will be interviewed in-depth to elicit richer meanings for their conceptions. Answers to testquestions and interview transcripts will be coded according to variousdescriptions of the energy concept. Overall, factors to be reflected onand critically examined may include curriculum organization, teachers’subject specialization and subject knowledge, effectiveness ofconstructivist teaching and learning, and recognition of the shift in thedescription of the energy concept in school science.In summary, the sequence for each case study, after securing contactthrough stage 2, may include:
interview the teacher on their understanding of energy using the‘interviews about instances and events’ techniquestest pupils and interview a sample on their preconceptions ofenergyprovide opportunity for INSET (if necessary) to the teacher onthe energy conceptreflect on and redesign elements of the teaching sequencetaking account of the above (constructivist principle)participant observation of sequence of lessons, taking fieldnotes, coding for reporting, reflectingtest pupils at the end, reinterview a sampledissemination, reflection on whole processpossibly retest pupils at a later date to test for consolidation ofconceptual understanding (constructivist principle).MethodologyIt is this researcher’s position that this broad and mixed methodologyis appropriate in addressing the aims set out for this study. A mixtureof qualitative and quantitative methods will be used. Stages 1 and 2are felt to be necessary prerequisites for stage 3, not only in theinformation they will yield but, importantly, in the personal realm. Thewhole design has Kelly’s personal construct theory in mind, inparticular the ‘subsuming of personal construing systems’ andsensitivity to ‘core constructs’. On initial contact with schools, the aimis to ask only superficial questions about curriculum organization.Then, only when sufficient rapport and trust have been built up withteachers, will it be considered appropriate to probe with deeperquestions. For example, the issue of teachers’ subject knowledge mayarise in stage 3 which, as mentioned, may be a sensitive area forsecondary school science teachers. This is especially so whenconsidering the position of a university researcher, not in the currentposition of teaching, coming in to work with teachers to criticallyexamine practice.Stage 3 has been designed on action research and constructivistprinciples. This researcher feels that AR is a good methodology forsuch in-depth educational research because of practitionerinvolvement and empowerment in the research process, the groundednature of the knowledge and the practical value of the researchgenerated. Also, because of the self-reflective process, it encouragesthe potential for a positive change in practice. Constructivism is felt tobe an appropriate pedagogy on grounds of its weighting in currentliterature, as reported.
View to a PhDIt is intended that this study will contribute to the knowledge base ofimproving the teaching and learning of the energy concept in scienceat KS3: in particular, by highlighting, critically assessing and reportingon factors crystallized from the research process that are believed tobe important in effecting the conceptual understanding of energy. Thisstudy will also demonstrate a mixed methodology, drawing on Kelly’stheory of personal constructs, action research, constructivism, andqualitative and quantitative methods. This may have the potential forextension into KS4, FE or other subject areas where conceptualunderstanding is an issue.Time-frame for studyYear 1 Term 1 literature survey Term 2 design study, fine detail Term 3 contacting, stage 1, pilot stage 2, PGCE studentsYear 2 Term 1 main study Term 2 main study Term 3 stages 1, 2 and 3Year 3 Term 1 stages 1, 2 and 3 Term 2 write up Term 3 write up(Robert Illes, Oxford Brookes University)Box 1.15 Example research plan 3Initial study for the production of adatabase of current and completedresearch in facilities management for the
use of managers, designers andresearchers working in NHS estatesObjectives and methodology1. Determination of necessary range and scope of the databasesubjects, present search methods, preferences andrequirements. Method: structured personal and telephoneinterviews with a small selection (approximately twenty) ofmanagers working in relevant fields on NHS projects. Result:formulation of database ‘brief’.2. Exploration of existing data sources which are being and can beused to locate the research. Method: library searches, telephoneand written enquiries to professional and research centres.Result: list of existing databases and sources of research activityinformation.3. Investigation of nature of access to these data sources, costs ofaccess, copyright situation with regard to use of data, likelycopyright costs. Method: written and telephonic communicationwith producers and managers of data sources. Result: detailedinformation of availability and costs of collection of data onresearch activities.4. Enquiry into suitable vehicles for the database, e.g. CD-ROM,Internet, intranet etc., including review of search engines toenable easy access to database by NHS staff and othersworking on NHS projects, including estimation of costimplications. Method: sourcing of specialist literature on thesubject, consultation with practitioners in the field. Result: list ofoptions with considered advantages/disadvantages and relativeestimated costs.5. Survey of suitable database computer programs, includingdesign features, formats, search methods, print options,updating characteristics and likely costs. Method: collection ofavailable program specifications, scrutiny of program reviews incomputer press, consultations with database designers andmanagers in university and professional fields. Result: draftreport on comparative features, suitability and costs of availableprograms.6. Review of possible methods of management, maintenance,quality control and periodical updating of database, ownershipand licensing options, and relative costs. Method: estimation ofrequirements depending on system, then consultations with NHSEstates strategic managers on funding and staffing optionswithin the NHS organization and/or use of outside consultants.
Indicative cost quotations from consultants. Result: list ofpossible options and costs with discussion of implications.7. Production of draft discussion paper together with feedbackquestionnaire seeking informed opinion from executives andmanagers. Submission to NHS Estates executive and peerreview. Method: preparation of paper containing collation ofinformation gained in stages 1–6, together with a list of optionsand reasoned recommendations made on the basis of theavailable information, expert and managerial views recordedduring consultations, policy and viability judgements by NHSEstates executives. Questionnaire formulation on the basis ofoptions suggested. Also, recommendations on dissemination ofdiscussion paper. Circulation to selected NHS managers andpeer researchers for comments. Corrections and adjustments todiscussion paper as a result of review. Result: an approveddiscussion paper ready for dissemination.8. Production and dissemination of discussion paper with attachedquestionnaire. Method: copying and postal dissemination.Result: facilities managers in the NHS informed of options andasked for opinions.9. Evaluation of questionnaire responses. Method: simple statisticalevaluation of questionnaire returns (using SPSS) to analyse andsummarize responses to options and recommendations.Summarization of any additional suggestions and commentsobtained. Result: summary of feedback from profession.10. Production of report with recommendations and possibly aproposal based on this for the production and management of aresearch database. Method: writing of report based ondiscussion paper and responses to it. Result: a well-informedreport aimed at policy and decision makers in NHS Estates onthe options, likely costs and acceptability of a database ofcurrent and completed research in facilities and estatemanagement. This could form the basis for a proposal for theproduction and maintenance of such a database.Project milestonesThe plan of work will be carried out in three stages:1. Formulation of design ‘brief’ for database and collection andassessment of information (eight weeks).2. Writing of discussion paper and consultation (five weeks).3. Analysis of questionnaire returns and production of report withrecommendations (six weeks).
The detailed tasks are scheduled in the project plan (not shown here).The milestones for reporting results are: end of collection andassessment of information (25 January 2019), dissemination ofdiscussion paper draft (15 February 2019) and submission of the finalreport (29 March 2019).The information-gathering, consultations and evaluations will becarried out by Nicholas Walliman with the occasional assistance(advice) of other members of the team. A casual clerical assistant willbe employed sporadically at appropriate times to help in theorganization of the data and writing up.Methods for disseminating andimplementing research to the NHSA discussion paper will be produced which sets out an evaluation ofthe advantages, options and costs of setting up and maintaining aresearch database. This will be disseminated to all relevant managersand executives within the Health Service. The paper will include afeedback questionnaire which, when returned, will be evaluated, anda report will be published which makes specific recommendations.This report will be sent to policy and decision makers in NHS Estatesand to all the managers and executives on the previous list.If the recommendations support the creation of a research databaseor some alternative to it, these will form the basis of a proposal for thenext stages in the production of a system for improving access to thelatest research information relevant to facilities management in theNHS, which will be submitted to the appropriate department of theNHS Estates Agency.(Nicholas Walliman, Oxford Brookes University)Phillips and Pugh (2015, p. 60) maintained that doing research isa craft skill, in which the basic educational process is that oflearning by doing. After you have decided on an overall plan foryour project and on your research approach, you should carefullyconsider how you can acquire the skills required to carry out eachof the ‘craft’ elements. This takes time, and should be included asan element in each phase of the work. You should also considerthat some practice is required before you use the skill in yourproject, giving you the opportunity to gain some feedback andgiving you greater confidence when you use it ‘for serious’.
Exercise 1.4Think about your own research topic, and the skills you will need todevelop to carry it out. Then:1. Analyse the examples of plans of work in Boxes 1.13, 1.14 and 1.15to detect the steps taken to achieve the research aims. Compare thetypes of approach, and consider what you can learn from themrelated to your own project. Obviously the subject will be different,but you will see how the main stages of data collection, analysis andmaking conclusions are a common feature.2. Make a list of the likely skills that you will need to learn and practisebefore you can carry out the various stages in your research plan.The next steps: finding yourresearch problem areaThe aims of this section are for you:to review the contents of this chapter in relation to yourintended researchto explore your own subject for problem areasto assess the practicality and suitability of possible researchinto those areas, in order to narrow down your choice anddefine a research problemto decide what further information you requireto consult and discuss your ideas with others.Checklist of activities that will progress yourresearchStep 1: Use your assets to identify problemareasYour own academic, professional and personal experiences are valuableassets in your research. Make sure you make the best of them. On thebasis of what you already know about your subject, and your
understanding of the nature of research, examine your field of interest andidentify two or three problem areas that might be researchable. Problemareas might be found by detecting systems or organizations that do notseem to perform satisfactorily, either theoretically or practically. Larger-scale issues, for example energy conservation related to the environment,might interest you. Have you read of any widely held beliefs in yoursubject which you think are misleading or quite wrong, or is there asignificant lack of information about a topic you consider to be ofimportance?When you have selected the problem areas, explore the issues involvedby devising a number of questions that highlight the nature of the problemor reveal different aspects of it. This will help to lead you to more specificresearch problems that could be the basis of your own research project.Step 2: Be alertKeep your eyes and ears open for interesting research topics. At the endof virtually all research papers and reports, and even books, there is asection that outlines the need for further research, often quite preciselydefined. These could provide you with a good lead to your own researchproblem.Step 3: Choose a research problemConsider what further information you might need to obtain to clarify anddelineate the problem areas. Do not try to be too specific at first. It is agood idea to make a list of your key interests in your subject, so that youcan look at problems that contain some or all of your interests.Choosing a research problem to tackle for your project is a crucial stepthat will affect your life for months or even years. Make sure that you aresufficiently fascinated in the issues to motivate you through to the end.Keep the project manageable, however interesting and important thesubject is. Do this by limiting it to aspects within your own expertise. Leavethe other specialist areas for further research by others.As a guide to your investigations you may find it useful to answer thefollowing questions in each of your research problems:Has anyone else done research into the same or similar problems?Can you imagine how a methodical and scientific approach could beadopted to research into these particular problems?What issues would need to be explored?How important do you think that research into these problems is?
Do you think it might be possible to narrow them down (delineatethem) so as to make them a practical subject for your thesis? If so, canyou suggest some simple examples?Step 4: Break down the problemIn order to see what might be involved in tackling a research problem,break down the problem into practical components, i.e. sub-problems orsub-questions. Work out what you will actually need to do to answer these.Can you manage to do it in the time allotted? If not, narrow the scope ofthe problem.Step 5: Choose your research strategyThe three types of research strategy – exploration, testing out andproblem-solving – are related to the research objectives listed in step 1.Where does your research fit into these? A strategy is a plan. Relate thisalso to your mapping out of your individual research process, and perhapsadd a time and place element to get a framework of what you will actuallydo. The next chapter will explain in more detail the methods that you canuse to carry out this strategy.Step 6: Map your journeyFrom your understanding of the research process, can you map out anetwork or a chart that will describe your own journey through theprocess? The examples given are very general; you can make yours morespecific to your project. Look out for two essential aspects: thecharacteristic broadening and narrowing of the subject area as you distilyour research down to the essential elements; and the presence ofreiterative feedback loops – periodic checking back to theory andevidence.Step 7: ConsultConsult as many relevant people as possible to discuss your ideas –tutors, experts, fellow students. Mere verbal discussion about ideas isdifficult to pin down. In order to communicate your latest ideas as clearlyas possible, write them down. This gives the people you consultsomething concrete to comment on. Keep developing your ideas byredrafting your proposals in response to your discussions.Step 8: Discuss
Prepare a set of notes so that you can discuss these issues with your tutoror supervisor; it is best to give it to him/her to read before your discussion.This is an exploratory exercise, so do not try to formulate your detailedresearch problem at this stage. This will come later. Having studied thischapter of the book, at least you will know what sort of problems you arelooking for!Consolidation and assessmentFollowing the above steps should lead to the identification ofpromising problems and to a preliminary analysis of the issuesinvolved in each. This will help you to define the type ofbackground information you need to explore, and to delineatewhat issues might be of importance in narrowing down andclarifying a researchable problem to use as a basis for yourresearch.When you are prepared, you should arrange a tutorial with yourtutor or supervisor and hand over your notes for him/her to read.Your tutor or supervisor should discuss with you the potential forresearch in your suggested problem areas and selection ofpossible research problems, and will also suggest what you needto do next in your background investigations. You will now be ableto demonstrate an understanding of the basic characteristics ofacademic research, and you may want to discuss with your tutorsome of the issues about research raised in this chapter.Further readingA good place to start is to look at previously completed theses,dissertations, papers or research reports in your subject. This willnot provide you with instructions on how to proceed, but will giveyou plenty of food for thought, and help to stimulate your owncritical faculties about the content and quality of the workpresented. This will be important when it comes to reviewing yourown work later on.Most books on this subject cover the whole sequence of preparingand writing dissertations, theses, reports, etc., but hardly any
actually discuss why you should do research, and what theexaminers and other readers will be looking for. Despite this it isinteresting, if you have time, to compare the advice given at thisstage of the process. The approaches vary, depending on thelevel of research, and in some, the specific subject area cateredfor. Only look at the preliminary advice given in the first sections ofthe books and scan the contents page to see if there is anythingelse of interest further on.The following books are aimed at undergraduate, postgraduateand practitioner research and selective reading of the preliminarychapters will provide further guidance on research basics. Eachgives a slightly different view of the issues, so refer to as many aspossible. You can probably do this in the library without eventaking the books out on loan. When you locate them on theshelves, look at the contents list of promising books for relevantchapters.Blaxter, L., Hughes, C. and Tight, M. (2010) How toResearch, 4th edn. Buckingham: Open UniversityPress.The first chapter gives an entertaining review of what research isabout.Rudestam, K.E. and Newton, R. (2014) Surviving YourDissertation: A Comprehensive Guide to Content andProcess, 4th edn. Thousand Oaks, CA: Sage.Again, the first couple of chapters provide an introduction toresearch.David, M. and Sutton, C. (2010) Social Research: AnIntroduction, 2nd edn. London: Sage.A good chapter on ‘Getting Started’.Swetnam, D. (2000) Writing Your Dissertation: How toPlan, Prepare and Present Successful Work, 3rd edn.
Oxford: How to Books.Chapter 1 gives some simple advice on how to get started.Biggam, J. (2088) Succeeding with Your Master’sDissertation: A Step-by-Step Handbook, 4th edn.Buckingham: Open University Press.A useful, simple and easy-to-read book for anyone who has notdone a dissertation before.And some books with a more subject-oriented approach:Bell, J. (2018) Doing Your Research Project: A Guide forFirst-Time Researchers in Education, 7th edn.Buckingham: Open University Press.Brundage, A. (2017) Going to the Sources: A Guide toHistorical Research and Writing, 6th edn. Hoboken, NJ:Wiley.Cooper, D. and Schindler, P. (2013) Business ResearchMethods, 12th edn. Singapore: McGraw-Hill.Pennings, P. (2005) Doing Research in Political Science:An Introduction to Comparative Methods, 2nd edn.London: Sage.Polgar, S.T. (2013) Introduction to Research in the HealthSciences, 6th edn. Edinburgh: Churchill Livingstone.Saunders, N., Lewis, P. and Thornhill, A. (2015)Research Methods for Business Students, 7th edn.Harlow: Pearson Education.Wilson, E.B. (2003) An Introduction to Scientific Research.New York: Constable.
2 Research Theory and the Nature ofKnowledge: UnderstandingPhilosophies and ApproachesAimsTo provide an introduction to questions about the process of researchin order to make you think about what form of enquiry you will use toresearch into your own problem. This is done by:Exploring some aspects of the philosophy of research and thedebate about the nature of knowledgeExplaining in more detail the forms and characteristics ofscientific methodExploring alternative approaches to researchIntroductionThis chapter begins with a brief description of scientific enquiry,together with a short review of the debate in which criticalrationalism is challenged. Then the underlying philosophies in thisdebate, particularly related to the social sciences, are consideredin more detail.It has often been said that a study of the philosophy of the naturalor human sciences is irrelevant to researchers. As a commentaryon how theory after theory is erected, only to be torn down by thesubsequent one, it has little bearing on the day-to-day practice ofresearch, and only causes confusion. So why do I think that it isnecessary to know something about philosophy as a backgroundto research? Because everyone is a philosopher – everyone has aconcept of the world.
The alternative to philosophy is not no philosophy but badphilosophy. The ‘unphilosophical’ person has an unconsciousphilosophy, which they apply in their practice – whether ofscience or politics or daily life. (Collier, 1994, p. 16)Philosophy works by making arguments explicit. A sensitivitytowards philosophical issues is needed to enable you to evaluateresearch critically. It is necessary to discern the underlying, andperhaps contentious, assumptions upon which research reportsare based even when these are not explicit, and thus to be able tojudge the appropriateness of the methods that have beenemployed and the validity of the conclusions reached.Consequently, you will also have to consider these aspects inregard to your own research work.All philosophical positions and their attendant methodologies,explicitly or implicitly, hold a view about social reality. Thisview, in turn, will determine what can be regarded aslegitimate knowledge. Thus the ontological shapes theepistemological. (Williams and May, 2002, p. 69)Some simple examples will help to clarify this. Suppose you havebeen asked to carry out some research about children’splaygrounds in the city. You could collect very different types ofdata to analyse. For example, you could look at official statisticsabout how many playgrounds there are in relation to populationfigures, their sizes, facilities and locations, records of vandalismand child crime. Or you could arrange interviews with children andparents to find out what they felt about different playgrounds. Youcould also observe the playgrounds from above and plot children’splaying patterns in the form of geometrical shapes; or you couldobserve from nearby and record how the children used each pieceof equipment. You could also make measurements of how muchforce is exerted by children playing on the equipment, test itsstrength and measure rates of corrosion, in order to ensure that itis safely built. Each of these approaches involves basic theoreticalas well as methodological decisions.
So, for example, if you wanted to establish correlations between‘social facts’, you would favour collecting official statistics.However, if you were concerned with finding ‘social meanings’,then interview studies would be more appropriate. If you weremore interested in theories of interactionism or ethnomethodology,you would make a close observation of what people did, either asa whole or in detail. Finally, if you were interested in the technicalperformance of play equipment, then tests and experimentsaccording to scientific method would produce the relevantanswers.Accordingly, you will have to decide which philosophicalstandpoint(s) to adopt when carrying out your research. This is notto say that any approach is better or more true than the other.Your research approach will depend on the characteristics of yourresearch problem and your own convictions about the nature ofresearch. The decision will help you to determine the nature ofyour enquiry, the choice of appropriate research methods, and thecharacteristics of the outcomes that you can expect.Scientific enquiryBefore looking into the debate surrounding scientific enquiry, it isworth considering what comprises the conventional image ofscience. Hacking created a list of nine points, which I reproduce infull in Checklist 2.1. Although not all are held by any singlephilosopher or scientist, they present a useful summary of awidespread popular conception of science.Checklist 2.1 Points about scientific enquiry1. Realism. Science is an attempt to find out about one real world.Truths about the world are true regardless of what people think,and there is a unique best description of any chosen aspect ofthe world.2. Demarcation. There is a pretty sharp distinction betweenscientific theories and other kinds of beliefs.3. Science is cumulative. Although false starts are commonenough, science by and large builds on what is already known.Even Einstein’s theories are a development from Newton’s.
4. Observation–theory distinction. There is a fairly sharp contrastbetween reports of observations and statements of theory.5. Foundations. Observation and experiment provide foundationsfor and justification of hypotheses and theories.6. Theories have a deductive structure and tests of theoriesproceed by deducing observation reports from theoreticalhypotheses.7. Scientific concepts are rather precise, and the terms used inscience have fixed meanings.8. There is a context of justification and a context of discovery. Weshould distinguish (a) the psychological or social circumstancesin which a discovery is made from (b) the logical basis foraccepting the facts that have apparently been discovered.9. The unity of science. There should be just one science about theone real world. Less measurable sciences are reducible to moremeasurable ones. Sociology is reducible to psychology,psychology to biology, biology to chemistry, and chemistry tophysics. (Hacking, 1981, pp. 1–2)Although scientific enquiry has been carried out in a variety ofways through the centuries, it is possible to detect twoapproaches representing the extremes of method – induction anddeduction. At one extreme, induction aims to establish theoriespurely on the basis of observations, while, at the other, deductionaims to devise theories by intellectual means and then to testthem through observation. The hypothetico-deductive method, arelative newcomer to scientific thought, is a combination of both ofthese approaches.Pure inductionThe earliest and, even now, the commonest popular form ofscientific activity is that of induction. Every day, our experienceslead us to make conclusions, from which we tend to generalize.There are, however, certain shortcomings in this form of scientificprocess. Consider this handed-down version of Bertrand Russell’s(1912, p. 98) somewhat gruesome story of the inductivist turkey(actually a chicken in the original):Induction
This turkey found that, on the first morning in the turkey farm, hewas fed at 9.00 a.m. However, being a good inductivist, he did notjump to conclusions. Patiently, he waited many days, observingthat he was fed every day at 9.00 a.m., whether the sun shone orit rained, in windy weather and in calm. Eventually, his careful listof observation statements led him to conclude that ‘I am alwaysfed at 9.00 a.m.’ On Christmas Eve his inductive inference withtrue premises was shown to have led him to a patently falseconclusion, for on that day at 9.00 a.m., instead of being fed, histhroat was cut.This can be seen as an example of naive inductivism. Despite theobvious shortcomings of this method of scientific enquiry, thedevelopment of this approach in the seventeenth century by suchscientists as Galileo and Newton heralded the scientific revolution.The philosopher Francis Bacon summed this up by maintainingthat in order to understand nature, one should consult nature, andnot the writings of ancient philosophers such as Aristotle, or theBible.Many scientific advances have been achieved through inductivereasoning, through the formulation of generalizations and lawsderived inductively from experience and observation. However,three conditions must be satisfied for such generalizations to beconsidered legitimate by inductivists, as in Checklilst 2.2.Checklist 2.2 Conditions for inductive reasoning1. There must be a large number of observation statements.2. The observations must be repeated under a large range ofcircumstances and conditions.3. No observation statement must contradict the derivedgeneralization.In order to utilize conclusions drawn from inductive enquiry,predictions are made from the derived generalizations or laws,using logic and valid deductive reasoning, e.g. the next eclipse ofthe sun can be predicted from the laws of astronomy. Despite lawslike these having been and still being useful, induction’s merit was
disputed as long ago as the mid-eighteenth century by Hume. Hedemonstrated that the argument used to justify induction wascircular, using induction to defend induction. Consider thefollowing argument:The principle of induction worked successfully on occasion x1.The principle of induction worked successfully on occasion x2etc.The principle of induction always works.The concluding universal statement is inferred from the repeatedsuccessful applications of the principle of induction, and istherefore an inductive argument. This has traditionally been calledthe ‘problem of induction’.Two further serious problems for the naive inductivist remain. Thefirst is how large the number of observation statements must be;and the second is how large a range of circumstances andconditions must be repeated in order that true conclusions can bereached.It would be a brave (or perhaps foolhardy) person who would puthis/her hand into the fire more than once in order to establish thatfire burns. On the other hand, it would be rash to assert thatsmoking causes lung cancer on the evidence of only one case ofa heavy smoker dying of the disease. It seems clear that theappropriate number of observational statements required in orderto come to a conclusion by induction depends very much on theparticular observations made.The number of circumstances and conditions under whichobservations of a particular phenomenon are made can beinfinitely large. Obviously, some choice must be made in order toreject superfluous variations. But how is it possible to select whichof the circumstances and conditions are pertinent to render aninductive inference legitimate? To make valid decisions requires
some theoretical knowledge of the situation, that is, theory prior toobservation – the naive inductivist’s anathema!The doubts cast on the reliability of conclusions drawn frominduction, and the difficulty in fulfilling the conditions required tolead to proof, have led to a partial retreat in the claims of the naiveinductivist. While it is no longer claimed that 100% certainty canbe achieved in generalizations made through induction, it is heldthat repeated observations of a phenomenon under a variety ofconditions can lead to a conclusion which asserts the probabilityof the conclusion. The greater the number of observations underthe greater the number of conditions, the more likely is theprobability of the truth of the generalization. An even more limitedapproach leads to narrowing the predictions from those of generalstatements to those of particular cases; for example, engineeringscience can predict that a particular ship of a certain design willwithstand expected sea conditions and not sink, but not that allships are safe under any conditions. However, the basic circularproblem of induction remains.Briefly, there are responses to the problems of induction. Here arethree:1. The sceptical response made by Hume was: given thatscience is based on induction, which cannot be justified byappeal to logic or experience, acceptance of laws andtheories is little more than a psychological habit acquired as aresult of repeated observations.2. The reasonableness of induction: as all knowledge must bebased on observation, induction therefore is an ‘obvious’method of reaching conclusions from observations.3. Induction is non-scientific: science is not based on induction,and therefore induction does not need to be scientific. Thiswas said by Karl Popper, of whom more later.Deduction and the principle offalsification
The opposite of the ‘research then theory’ approach, typified byinduction, is the ‘theory then research’ approach, usingdeduction. Research is guided in this case by the theory whichprecedes it. Theories are speculative answers to perceivedproblems, and are tested by observation and experiment. Whilst itis possible to confirm the possible truth of a theory throughobservations that support it, theory can be falsified and totallyrejected by making observations that are inconsistent with itsstatement. In this way, science is seen to proceed by trial anderror: when one theory is rejected, another is proposed andtested, and thus the fittest theory survives. While it may not bepossible to prove that a particular theory is true, it can be shownto be the best available for the present, and better than any whichpreceded it.DeductionThe logical justification for this form of scientific enquiry is simplythis: while any number of observations cannot establish the truthof a theory, one observation conflicting with the theory statementis sufficient to prove its falsity. Take for example the theory that ‘allswans are white’. Whilst a million sightings of white swans cannotprove the truth of the theory, only one sighting of a non-whiteswan will be sufficient to reject it. Therein lies the strength of thisform of approach.HypothesisIn order for a theory to be tested, it must be expressed as astatement called a hypothesis. The essential nature of ahypothesis is that it must be falsifiable. This means that it must belogically possible to make true observational statements thatconflict with the hypothesis, and thus can falsify it.Hypothesis
Exercise 2.1Here are a number of statements. Examine each one and decide whichones are falsifiable, and are therefore possible to use as hypotheses, andwhich not. If not, can you say why not?1. It always rains on Saturdays.2. All squares have four sides of equal length.3. Heavy objects thrown from a height fall earthwards.4. Water boils when heated to 100 degrees Celsius.5. Either it is raining or it is not raining.6. A steel beam bends when it is loaded.7. Profits can be made in business enterprises.8. All circles are round.9. All round objects are circles.10. An object is either still or moving.These are very simple examples of statements, and it is easy todecide if they are hypotheses or not. In practical, and even more intheoretical, research it is generally more difficult to decide on the truthof hypotheses, as it is often not obvious how they can be tested. It isup to the researcher to demonstrate how effective tests can beformulated in order to attempt to falsify the hypothesis.FalsificationIt is widely accepted that science progresses by highlighting andinvestigating problems. It might be maintained, however, thattheories which respond to these problems are devised only as aresult of observation, and are therefore inherently inductive. Wecan show by example that this need not be the case.Observations often simply highlight problems in the explanatorypower of existing theories, and therefore give rise to speculativealternative theories, expressed as hypotheses. For example,during the Black Death, when many people were dying of theplague, the theory that the plague was spread by odours in the airemanating from the sick and dying was questioned when people
who had not been in the presence of afflicted people became ill.Alternative causes of the spread of the disease were sought. Aspeculation that there was some connection between theomnipresence of flea-infested rats and the spread of the diseaseled to a new theory. The previous theory was rejected.The process of falsification can lead to a less devastatingresult than an outright rejection of a theory, requiring a completelynew start. Popper maintained that much progress in science hasbeen promoted by the modification and increase in sophisticationof current theories. Take, for example, this very simplifiedillustration:FalsificationA theory states that water always boils at 100 degrees Celsius. Itwas observed that water boiled at a lower temperature than 100degrees Celsius at high altitude. This presented a problem, andgave rise to the speculation that differences in air pressure due toaltitude might have an influence on the boiling point of water, afactor not considered in the basic theory. The influence of altitudewas added to the theory, which now read: ‘water boils at 100degrees Celsius at sea level’. The basic theory is not totallyrejected but is refined in the alternative theory.In this case the new theory is more precise, and therefore morefalsifiable, than the original one, and therefore is to be preferred. Itis in this way that the sophisticated falsificationist sees the steadyprogress of science occurring, through the development ofprogressively more refined and falsifiable theories.Hypothetico-deductive or scientificmethodProgress in scientific thought of a much more dramatic nature isgained, on the one hand, by highly conjectural predictions thatturn out to be supported by observation, and on the other, byseemingly obvious and unadventurous conjectures being falsified.
This approach is commonly known as the hypothetico-deductive method and is generally synonymous with thescientific method.Hypothetico- deductive methodScientific methodA simple summary of the steps in scientific method is given inChecklist 2.3.Checklist 2.3 Steps in scientific method1. Identification or clarification of problems2. Formulation of tentative solutions or hypotheses3. Practical or theoretical testing of solutions or hypotheses4. Elimination or adjustment of unsuccessful solutionsThere are, however, theoretical limitations to the power offalsification. The most serious is that observation statements aretheory dependent and fallible. While theories can logically befalsified by true observation statements which contradict them,there is no proof that the observation statement is true. One needonly call to mind that before Copernicus, the interpretation ofobservation of the movement of the sun, stars and planets aroundthe earth resulted in the rejection of any theory that stated that theearth was not the centre of the universe. If all observationstatements are fallible, then two serious consequences can beimagined. Either a sound theory can be wrongly rejected by afalse observation statement, or a false theory is not challengedbecause a true observation statement is rejected as faulty. Hence:Conclusive falsifications are ruled out by the lack of aperfectly secure observational base on which they depend.(Chalmers, 1999, p. 60)
Additional problems are posed by the complexity of testingtheories in real life. Realistic scientific theories consist of acomplex of statements, each of which relies on assumptionsbased on previous theories. The methods of testing are likewisebased on assumptions and influenced by surrounding conditions.If the predictions of the theory are not borne out in the results ofthe tests, it could be the underlying premises which are at faultrather than the theory itself.There are many examples in the history of science where radicaltheories that had been falsified by current evidence, andaccording to the falsificationists should have been rejected, wereconserved for many years until advances in other areas of scienceoverturned the evidence. Examples of theories that survived inthis way are Newton’s gravitational theory, Bohr’s theory of theatom, Maxwell’s kinetic theory of gases, and Copernicus’s theoryof the solar system. Such valuable theories would have beenrejected prematurely if the falsificationist approach had beenslavishly followed.It has probably become quite evident to you that, despite itsshortcomings, you use inductive reasoning every day quitesuccessfully without even thinking about it, and also that a hugeamount of successful scientific research has been carried out usingthe hypothetico-deductive method. However, there is a wider debateabout scientific enquiry which examines the relationships betweenscience and the nature of knowledge.Not everyone agrees that knowledge, even scientific knowledge,is as firmly based on the relationship between reality, verifiablefacts and theoretical statements as is commonly accepted. Theactivities that surround scientific research are much morecomplicated than the simplistic examples and models used toillustrate the approaches to scientific enquiry. The study of thehistory of science has shown that merely studying therelationships between theories and individual observationstatements, or series of observation statements, does not takeinto account the complexity of major scientific theories, and the
often slow and tortuous development before they reached theirprecise form.In order to explore some major and different views on the natureof scientific enquiry, we can compare the critical rationalistapproach as put forward by Popper with the relativist approach ofKuhn, and finally with the ‘anti-method’ stance taken byFeyerabend. These three major approaches to scientific enquiryindicate that there is debate about the issues of theory evaluationand theory selection, and about ways of demarcating science fromnon-science.Critical rationalismThe critical rationalism approach, usually associated withPopper, maintains that rival theories can be judged againstspecific, unchanging, universal criteria, which are divorced from orset beyond the influences of time or society. This insistence onuniversal criteria leads the rationalist to believe that the better thetheory performs against these criteria, the closer it is to auniversal ‘truth’.Critical rationalismThe distinction between science and non-science is easily drawn.If the theory cannot be assessed in terms of universal criteria,then it is not scientific. For example, astrology is rejected by aninductivist rationalist, as it cannot be supported by facts gained byobservation. Likewise, the unfalsifiability of Marxist theory causesa rationalist to reject this theory as unscientific. Truth, which isundeniably a value worth striving for, is regarded as the mostimportant goal of science, and can be approached only throughrational means.Relativism
Opposed to the belief that certain universal rules based onrational criteria can be applied to judge the quality of theories, thestance of relativism implies that judgement is principallydependent on the values of the individuals or society and theperspectives from which they make their judgement. No universalcriteria can be ‘rationally’ applied, and an understanding ofdecisions made by individuals or organizations can only be gainedthrough a knowledge of the historical, psychological and socialbackgrounds of the participants.RelativismFigure 2.1 Observations often simply highlight problems inexisting theoriesKuhn claimed that a study of the history of scientific developmentreveals that scientific progress is not a smooth and steadyprocess. Once a branch of science has established itself as an
entity, e.g. as physics, biology or astronomy, then development ischaracterized by a cycle: normal science, leading to crisis, thenrevolution, which is then resolved in a new normal science.Knowledge has therefore a historical perspective: what isregarded as true at one time can be radically changed or evenreversed at a later date.Figure 2.2 Scientific progress is not a smooth and steady processNormal science is the period in which the established andgenerally accepted body of knowledge in the discipline isextended and improved, using established and successfultechniques. During this time there tends to be a conservativeapproach, aiming at refining and clarifying concepts and theories.Kuhn distinguished between two characteristics of normalscience: the generally accepted body of knowledge and the beliefin methodology, which he collectively called a paradigm. Theformer is a set of shared values, which depicts precepts andgeneralizations shared by specialists; the latter is the measure ofachievement, which represents the established methods and
techniques of problem-solving used as a model for those workingin the field. Normal scientific activity is carried on within the termsof the paradigm.ParadigmWhen, as a result of this scientific activity, evidence mounts thatanomalies cannot be solved within the terms of the paradigm,uncertainty sets in as the precepts of the paradigm are broughtinto question, and new ways of looking at phenomena areexplored. This crisis provokes a revolution in ways of thinking andresults in a new paradigm being evolved. The new paradigm isgenerally incommensurable with the old. The concepts in the newparadigm may be different or have completely different meaningsand entirely different problems might be addressed. This alsomeans that methods of testing will be quite dissimilar to thoseunder the old paradigm.For Popper scientific change is rational or at least rationallyreconstructible and falls within the realm of the logic of discovery.For Kuhn scientific change – from one ‘paradigm’ to another – is amystical conversion which is not and cannot be governed by rulesof reason and which falls totally within the realm of (social)psychology of discovery. Scientific change is a kind of religiouschange (Lakatos, 1980, p. 9).Figure 2.3 Theories, particularly when new and revolutionary,always contain numerous formal flaws and contradictions
Science is therefore non-cumulative because the shift from oneparadigm to another involves the jettisoning of much informationand of no longer valid questions. The often sudden switch when anew paradigm becomes established creates a new way of lookingat that branch of science, or even the whole world, e.g. themechanistic world, the world of relativity, or even more recentlythe quantum world (Kuhn, 2012).As a demonstration of this radical switch in features of twosuccessive paradigms, compare the lists of features of Newtonianand quantum physics in Box 2.1, as drawn up by Danah Zohar ina lecture she gave in April 1995 at Oxford Brookes University (seealso Zohar and Marshall, 1993).Box 2.1 Comparison between Newtonian and quantum researchNewtonian Quantum
Determinist IndeterminateCertainty UncertaintyContinuous (linear) Discontinuous (non-linear)Reductionist (non-creative) Emergent (creative)Atomistic (isolated) Holistic (contextual)Causal Acausal (non-local)Hierarchical Non-hierarchicalEither/or Both/andActuality PotentialitySubject/object split Observer participancyNon-evolutionary Evolutionary‘Anti-method’ stanceBut can the scientific view be taken as the only path toknowledge? Feyerabend warned about the dangers of allideologies, scientific included. Why science? Surely it is generallyaccepted that science has always been in the forefront of the fightagainst authoritarianism and superstition, has liberated ourintellectual thinking, and has provided us with the means tocombat ancient and rigid forms of thought? Do not science andenlightenment go hand in hand? Were Kropotkin, Ibsen, Lévi-Strauss, Marx and Engels all wrong when they rejected traditionalinstitutions and forms of belief, while embracing scientificthinking? ‘To these questions my answer is a firm Yes and No’,wrote Feyerabend (1981, p. 156)! He went on to explain hisanswer, making the following points:Any ideology which makes people question inherited beliefs andchallenge accepted truths contributes to the liberation of people,and is therefore welcome. Science in the seventeenth andeighteenth centuries was indeed such an ideology.
It does not follow that science has remained thus. On the contrary,science has become as rigid and oppressive as the ideologiesthat it once fought. At school and in universities, scientific ‘facts’are taught just like religious ‘facts’ were taught in the past.Criticism is suppressed. By the claim that science is abovecriticism because it has discovered the ‘truth’, freedom of thoughtis inhibited.It is generally claimed that science holds such an importantposition in our society because, first, science has discovered theproper method for achieving results, and secondly, the methodhas been proved to be effective by the mass of results which ithas produced.Feyerabend challenged these claims. Scientific method no longerexists, he claimed. The shortcomings of the inductionist methodhave long been revealed by Hume. Popper and Mill attempted toovercome this methodological problem in different ways. Theyboth agreed that theories cannot be proved, and that their qualitycannot be judged except in comparison with other theories. Thechosen theory might be deficient in many ways, but as long as it isbetter than rival theories, then it will take precedence. Thealternative rejected theories are not eliminated, as they provide astandard against which the accepted theory is judged, andhighlight the content of the preferred views.Knowledge so conceived is an ocean of alternativeschannelled and subdivided by an ocean of standards. It forcesour mind to make imaginative choices and thus makes it grow.It makes our mind capable of choosing, imagining, andcriticising. (Feyerabend 1981, p. 159)Even so, Popper’s criteria are clear, unambiguous and preciselyformulated. Theories are either falsifiable or not falsifiable.Theories that are not falsifiable have no place in science, and areeliminated. However, Feyerabend claimed that these criteriaapplied resolutely would eliminate science without replacing it byanything comparable; for theories, particularly when new andrevolutionary, are never formulated to reveal how they could be
falsified, and anyway contain numerous formal flaws andcontradictions. Popperian criteria are therefore useless as an aidto science!Feyerabend regarded Kuhn’s ideas as interesting, but hopelesslyvague, and defied anyone to prove that there ever was a period of‘normal science’ in history. He went on to contend that sciencedoes not deserve a special position because of its results; manyother activities regarded as being quite unscientific also produceresults. For example, oriental medicine and telepathy arephenomena that are known to produce results unexplained byscientific theories. They are dismissed because ideologicalpressures make us listen to science to the exclusion of everythingelse. Only when a phenomenon can be explained in purelyscientific terms is it accepted. But, maintained Feyerabend, historyshows that advances in knowledge have rarely relied purely onscientific foundations. Copernicus relied on the ideas of a crazyPythagorean, Philalaos, when he introduced a new view of theuniverse; and when the Chinese communists insisted thattraditional medicine replaced Western medicine in universities andhospitals, Chinese medical science advanced and Westernscience learned from it. One must therefore conclude that there isno argument which could support the exceptional role scienceplays today in society. Feyerabend concluded that:Science is just one of the many ideologies that propel societyand it should be treated as such … It is an intellectualdiscipline that can be examined and criticised by anyone whois interested … Ideologies are marvellous when used in thecompany of other ideologies. They become boring anddoctrinaire as soon as their merits lead to the removal of theiropponents. (1981, pp. 162–3)Approaches to social researchFor many researchers in the social sciences, the philosophicaldebates about the nature of scientific method are of littlerelevance. They are much more concerned with grasping
‘meanings’ and understanding complexes of meanings asopposed to discovering ‘truths’ or approximations to them.The types of research problems investigated in the field of socialsciences and the arts are extraordinarily varied and do not all fallcomfortably within the parameters of natural science. They extendfrom the analysis of precise problems, amenable to investigationusing ‘traditional’ scientific method within a well-establishedparadigm, to explorative situations where the variables areunknown and theoretical bases are yet to be formulated, to highlyvalue- and meaning-laden sociological studies for which the latestinterpretive and literary analytical methods are appropriate.It is not surprising that the various alternatives to the criticalrationalist approach do not have the authority and precision oncepossessed by scientific method. A common feature is the rejectionof a purely objective view of society and the belief that knowledgeis itself a social phenomenon. Despite this, social scientistsbelieve that theoretical views of society are still possible, asopposed to only practical ones.The biggest challenge faced by the social sciences is that theresearcher him/herself is part of a society and a culture andcannot be a ‘neutral observer’. Moreover, the subjects of studyalso have a voice and can express their individuality. Thus, ‘allknowledge of cultural reality, as may be seen, is alwaysknowledge from particular points of view’ (Weber, 1949, p. 81). Itis not possible to stand back and take a neutral point of view ofany aspect of society.The difficulty for the social scientist is to come to terms withthe fact that their subject-matter and those who study it live ina world constructed through meanings. (Hughes, 1997, p.137)How social scientists have grappled with this fact is best explainedby briefly describing the different stances taken by some of theleading exponents of developments in this field. Durkheimadvocated a natural scientific approach in his work, investigating
what he called ‘social facts’ – the positivist approach.Husserl, Schutz and members of the Chicago School saw themeaningfulness of the subject matter as paramount – theinterpretivist approach. Weber believed that both a positivistand an interpretivist viewpoint are required, verifying subjectiveinterpretations with the actual course of events – thereconciliatory approach. By adopting a systematicperspective, Bhaskar revealed the relationships between thenatural and social sciences as one of stratification. This could beused as the basis for a strong argument to defuse the ‘either/or’confrontation between positivists and interpretivists. Lévi-Strauss,ignoring this debate, looked to cultural symbols as the key tounderlying social strata. Subsequently, the certainty aboutknowledge and how we communicate it was undermined byFrench philosophers such as Derrida and Foucault, which leads tothe ever-questioning position of the postmodernist approach.Positivist approachInterpretivist approachReconciliatory approachThe positivist approachSocial science, understood here as the study of human society inthe widest sense, is a rich source of research problems. Thisimportant, and sometimes controversial, branch of science was,according to Beck (1979), first defined and named by AugusteComte, the nineteenth-century French philosopher. Comtemaintained that society could be analysed empirically just likeother subjects of scientific enquiry, and social laws and theoriescould be established on the basis of psychology and biology. Hebased his approach on the belief that all genuine knowledge isbased on information gained by experience through the senses,and can only be developed through further observation andexperiment. He turned his back on pure reason, seeing it as ameans of speculation and metaphysical exploration.
Hence, the positivist social scientist believes that themethodological procedures of natural science may appropriatelybe applied to the social sciences, and its results can be expressedas laws or empirical generalizations similar to those developed forthe natural sciences. Social scientists can thereby achieve a clear,well-founded analysis and interpretation of social phenomena,based on testable and verifiable data.Durkheim is famous for his enquiries into the division of labour,suicide, religion and education, as well as for his philosophicaldiscussions on the nature of sociology. Unlike Marx, who tendedto define the moral and social aspects of humanity in terms ofmaterial forces, Durkheim argued that society develops its ownsystem of phenomena that produce collectively shared norms andbeliefs. These ‘social facts’ as he called them, e.g. economicorganizations, laws, customs, criminality, exist in their own right,are external to us and are resistant to our will and constrain ourbehaviour.Box 2.2 The task of the sociologistThe task of the sociologist … is to describe the essentialcharacteristics of social facts, demonstrate how they come into being,enter into relationships with one another, act on each other, andfunction together to form social wholes. (Hughes, 1997, p. 25)In order to carry out their task, which is summarized in Box 2.2,social scientists must reject all popular conceptions of theprocesses in society and look at the social world as if for the firsttime. Having ‘discovered’ and defined social facts using scientificobservation techniques, the social scientist should seek thecauses of those facts amongst other social facts rather than inother scientific domains such as biology or psychology. By thusmaintaining sociology as an autonomous discipline, the socialscientist may use the knowledge gained to understand the originsof, and possibly suggest the cures for, various forms of social ills.In summary, this approach looks at society as the focus forresearch, and through understanding its internal laws andestablishing relevant facts we can in turn understand how and
why individuals behave as they do. However, not all philosophersagreed that human society was amenable to such a disembodiedanalysis.The interpretivist approachFrom the second half of the twentieth century, some of the mostprominent intellectuals refuted the claim that scientific methodcould adequately provide a real understanding of the complexinterrelationships in society and between individuals. From whatcan be collectively called the viewpoint of interpretivism, theprincipal objections were that positivist social science presented amisleading picture of the individual in society by ignoring theunique personal theoretical stances upon which each personbases his/her actions. Also, unlike the natural sciences, theresearcher is not observing phenomena from outside the system,but is inextricably bound into the human situation which he/she isstudying. In addition, by concentrating on the search for constantsin human behaviour, the researcher highlights the repetitive,predictable and invariant aspects of society and ignores what issubjective, individual and creative.InterpretivismAccording to Cohen et al. (2017), there are three main schools ofthought represented by opponents of positivism in the socialsciences: phenomenology as developed by Husserl and Schutz,ethnomethodology pioneered by Garfinkel, and symbolicinteractionism practised by members of the Chicago School. Theyall reject the assertion that human behaviour can be codified inlaws by identifying underlying regularities, and that society can bestudied from a detached, objective and impartial viewpoint by theresearcher.In order to compare the alternative bases for interpreting socialreality, Box 2.3 presents a useful table which highlights theessential differences between objectivist and interpretivistapproaches.
Box 2.3 Comparison between objectivist and interpretivist approachesin social scienceThe implication of the interpretivist approach to research is thatthe observers or ‘interpreters’ cannot be seen as disembodiedfrom the context of their investigations. They bring, perforce, theirown meaning and understanding to the investigation, and mustrecognize and acknowledge the perspective from which theymake their observations. There is a strong recognition of the factthat attempts to find understanding in research are mediated byour own historical and cultural milieu.In contrast to the positivistic approach, we must look at theindividuals in society, to understand their values and actions, inorder that we may understand the structures and workings ofsocial systems.The reconciliatory approachWeber disagreed with the pure interpretivists, maintaining that it isnecessary to verify the results of subjective interpretiveinvestigation by comparing them with the concrete course of
events. He makes a distinction between what one can perceive asfacts, i.e. those things that are, and what one can perceive asvalues, i.e. those things that may, or may not, be desirable. Adifferentiation must be maintained between facts and valuesbecause they are distinct kinds of phenomena. However, in orderto understand society, we have to take account of both of theseelements.How can this be done? Weber maintained that in order to describesocial practices adequately we must understand what meaningsthe practices have for the participants themselves. This requiresan understanding of the values involved, but without taking sidesor making value judgements. This understanding (often referred toas verstehen) is the subject matter of social science. It is thenpossible to investigate the social practices rationally through anassessment of the internal logic of the situation. In this way, onecan make a meaningful formulation of the elements, causes andeffects within complex social situations, taking into account thevalues inherent in them.It can be argued that it is impossible for the social scientist to takethis detached view of values, as he/she is a member of societyand culture, motivated by personal presuppositions and beliefs.Weber admits:There is no absolutely ‘objective’ analysis of culture … or of‘social phenomena’ independent of special and ‘one-sided’viewpoints … All knowledge of cultural reality, as may beseen, is always knowledge from particular points of view.(1949, pp. 72–81)Accordingly, any analysis of social phenomena is based on a‘view from somewhere’. This is inescapable and even to bedesired. In order to make a scientific valuation of a socialphenomenon, Weber made the following distinctions. There arethose things that have cultural significance for the researcher thatare reflected in the concepts used and the values incorporated inthe research: these are the ‘givens’ at the outset of the researchproject. And then there is the researcher’s responsibility to
determine the facts in a value-free manner once the topic andframework for the analysis have been formulated: this is the‘scientific’ element of the work.In addition to this dual procedure, Weber also offered anotherapproach to understanding ‘messy’ social phenomena. Heformulated a procedure that logically constructed rational,simplified models of social forms generated from one or twocentral values. These were called ‘ideal types’. For example, themodel of an ideal bureaucracy would be based solely on theprinciples of rational and calculative efficiency. Obviously, no suchbureaucracy actually exists, but by comparing the ideal with theexisting cases, one can begin to isolate those factors that giveeach bureaucracy its individual character.This reconciliatory approach, then, promotes a reciprocalmovement between the study of the individual, his/her beliefs andvalues, and the study of structures of society, in order to check theresults of the one against the other.Critical realismThe philosopher Roy Bhaskar has provided an alternative to thedichotomous argument of positivism versus interpretivism bytaking a more inclusive and systematic view of the relationshipsbetween the natural and social sciences. His approach, known ascritical realism, sees nature as stratified, with each layer usingthe previous one as a foundation and a basis for greatercomplexity. Thus physics is more basic than chemistry, which in itsturn is more basic than biology, which is more basic than thehuman sciences. The relationships between these domains, fromthe more basic to the more complex, are inclusive one-wayrelationships – the more complex emerging from the more basic.Thus, all humans are composed of chemical substances but notall chemical substances are human.Critical realism
This means that humans are governed by chemical, physical andbiological laws. However, whilst a human being is not able to goagainst the chemical, physical and biological laws, he/she can doall sorts of things that the chemicals of which he/she is madecould not do if they were following only their specific chemicallaws rather than the biological laws that govern organisms, or thesocial ‘laws’ that govern society. Hence, the study of humanscannot be reduced to the study of their biological, physical andchemical properties (Collier, 1994, pp. 107–9). Note how thischallenges point 9 on Hacking’s list of the popular conceptions ofscience given in Checklist 2.1.The complexities of the upper strata of this stratified worldobviously present problems when one wants to study themscientifically. Whilst the stratified world is an open system, it iseasier to isolate phenomena in the lower strata, e.g. to isolate achemical process that occurs within an organic process. However,it is not possible to isolate an organic process from the effects ofchemical processes, since it has its origin in them. Thus thehigher up the hierarchy we go, the more difficult it is to isolatephenomena.Bhaskar also has a profoundly integrationist view of therelationship between the individual and society, called by him thetransformation model of social activity. Rather than, on the onehand, studying society to understand individual actions, or on theother, studying individuals to understand the structures of society,or somewhere in between, checking the results of one studyagainst that of the other, Bhaskar argues that the reciprocalinteraction between individuals and society effects atransformation in both. Individuals and societies are mutuallyinterdependent: individual actions (praxis) are influenced by thesociety (structure) in which they are carried out, but these actionsin turn influence this same society (transformation).Collier (1994, p. 146) cites language as a good example of this.We must learn the rules (structure) of a language before we cancommunicate with it. We then use the language (praxis) tocommunicate, without consciously thinking of the rules or wantingto change them. But the existence of the language depends on
our using it. And through our use of it, we often change the ruleswithout necessarily intending to do so (transformation). You onlyhave to read a Victorian novel to see how much the rules ofEnglish have changed through the years.Hence, although we can conceptually distinguish betweenindividuals and society, the way they mutually act upon oneanother makes them separable only by analysis.Society is both the ever present condition (material cause)and the continually reproduced outcome of human agency.And praxis is both … conscious production and (normallyunconscious) reproduction of the conditions of … society. Onecould refer to the former as the duality of structure, and thelatter as the duality of praxis. (Bhaskar, 1989, pp. 34–5)This interrelational view of society and the individual, and thephilosophy of ‘stratification and emergence’, have been widelyinfluential in the human and borderline natural/human sciences.Researchers in the fields of sociology, economics,psychoanalysis, linguistics, history, geography, biology, ecologyand feminist theory have put the ideas of critical realism to gooduse.Structuralism, post-structuralism andpostmodernismBased principally on the view that all cultural phenomena areprimarily linguistic in character, the structuralist approachgained its label because of its assertion that subjectivity is formedby deep ‘structures’ that lie beneath the surface of social reality.Lévi-Strauss used a geological metaphor, stating that the overtaspects of cultural phenomena are formed by the complexlayering and folding of underlying strata. These can be revealedby semiotic analysis. ‘Cultural symbols and representations arethe surface structure and acquire the appearance of “reality”’(Seale, 2004, p. 44).
StructuralistPost-structuralist thought was developed by French philosopherssuch as Derrida and Foucault in the latter part of the twentiethcentury. Through the method of ‘deconstruction’ the claims toauthority made in texts and discourses were undermined. Thisdenial of the validity of structuralism’s method of binary oppositionled to maintaining that meanings and intellectual categories areshifting and unstable. According to Seale (2004, p. 44),postmodernist thought subsequently developed and became morewidely accepted through the appeal of its three basic principles:1. The decentred self – the belief that there are no humanuniversals that determine identity, but that the self is acreation of society.2. The rejection of claims to authority – the idea of progressthrough scientific objectivity and value neutrality is a fallacyand has resulted in a moral vacuum. Discourse must besubjected to critical analysis and traditions and values shouldbe constantly attacked.3. The commitment to instability in our practices ofunderstanding – as everything is put to question there can beno established way of thinking. Our understanding of theworld is subject to constant flux, all voices within a culturehave an equal right to be heard.Postmodernism is applied to a wide-ranging set of developmentsin critical theory, philosophy, architecture, art, literature andculture, which are generally characterized as either emergingfrom, in reaction to, or superseding, modernism. In sociology,postmodernism is described as being the result of economic,cultural and demographic changes; related terms in this contextinclude ‘postindustrial society’ and ‘late capitalism’, and it isattributed to factors such as the service economy, the importanceof the mass media and the rise of an increasingly interdependentworld economy.Exercise 2.2
1. From what you have read above, and any further reading you havedone, summarize, in one or two sentences for each, the essentialpoints put forward by Popper, Kuhn and Feyerabend in theirarguments about the nature of scientific enquiry.2. Of these writers, who do you think might take opposite views withregard to interpretivism as a valid approach to the gaining ofknowledge and understanding? Give reasons why you think so.3. In the description of a research project given in the following text,which aspects do you think demand a positivist, interpretivist or otherapproach? Answer by considering and commenting on each of thefour action points (a)–(d) given in the text. You might find it useful torefer to the comparative list given in Box 2.3 as a guide to youranswer. Is it necessary to combine the two approaches in this oneresearch project or do you regard them to be mutually exclusive?TextThis study investigates the relationship between drawings and productdesign. Interest by designers in how drawings may influence the design ofproducts has led to diverse ideas about how this may occur. Though anumber of authors have identified isolated instances where suchinfluences occur, there has been no overall investigation of theimplications of drawing in product design. To research this relationship, itis necessary to:1. investigate the relationship of drawings to communication and theprocess of abstraction2. assess how coded information is perceived in drawings as percepts,and how convincing these percepts may be3. research how drawings and the forms they assume may influencedesign as a method of wider communication4. empirically assess, through interviews with designers anddelineators, the extent of influence that certain types of drawingsmay have upon their designs.The study provides a comprehensive overview of the communicationsaspects of drawings and how these may influence product design. Itdemonstrates that specific drawing types may influence design throughthe internal dialogue a designer may have with him/herself whiledesigning, and through the force such drawings may have incommunicating the design intentions of one designer to another and, inparticular, the recollection of drawing images while a designer isconjecturing during the design process.
In order to relate this rather wide-ranging discussion of the nature ofknowledge to your own research work, it is a good idea to examinethe objectives of your research in relation to the various philosophicalstandpoints described above. This will help to clarify which type ofstrategy might be suitable for your research. Will a positivist approachbe appropriate, using induction, deduction and/or hypothetico-deductive methods? Or is your investigation more suited to aninterpretivist approach? Perhaps a combination of the two would bethe most productive. Alternatively, would a structuralist or postmodernapproach be more suitable? These decisions will affect the way inwhich you express your research problem and the methods by whichyou will search for answers, discussed in Chapter 7.ConclusionsAny discussion of the issues that surround the nature of scientificenquiry and the acquisition of knowledge must necessarily be verybrief to fit the scale of this introductory book. Acknowledging thefact that research can concern itself with a huge range of subjectsand disciplines, I have attempted to cover the relevant majorperspectives which, I hope, will raise questions about your ownresearch approach, whatever your type of research problem.The debate about knowledge and how we acquire it is acontinuous one, and one that is at the centre of any researcheffort. I believe it is important, throughout your research project,and probably in your subsequent career too, to gain a deepeningawareness of what this involves and how such factors can affect,not only our attitude to life, but also our everyday actions.The next steps: the theoretical basisof your research projectThe intention in this section is for you to examine your ownresearch problem area in order to discover which of thephilosophical positions discussed earlier have a relevance to the
type of approach you intend to take and how you will construct asuitable argument to take your research forward. The awarenessof your own personal attitude towards the problem, the level atwhich you wish to investigate it, and the perspectives from whichyou will approach it, will all help to determine the characteristics ofyour future work.The aims of this section are for you:to examine your research problem or problem area toinvestigate what role the various theoretical approaches willplay in your research strategyto explore your own attitudes to your research problem, theperspective from which you will view it and the level at whichyou intend to examine itto formulate an appropriate research strategy, and exploresuitable styles of expressing the problem.Checklist of activities that will progress yourresearchStep 1: Examine positionsWhatever you read will be based on some kind of philosophical position.Try to categorize this for each of the books, papers and other material youread. Becoming aware of these will help you to understand the argumentsbeing made and guide you to formulating your own philosophicalapproach.Step 2: How scientific are you?How much do you think your research will conform to Hacking’s list offeatures of scientific enquiry (Checklist 2.1)? Questioning yourself on howyou envisage the development of your project will highlight your ownassumptions about the nature and process of research.Think carefully about how closely your proposed research project fits intothe various philosophical approaches. Are the assumptions of scientificmethod acceptable in your research approach, or is a more relativist basisappropriate? Some of the questions you will have to ask are:
Is your problem about objects or about people, or the relationshipsbetween the two?Can it be analysed and explained in terms of forces or inner physicalprocesses, or rather in terms of meanings and subjective forces?Are notions of causation an important aspect, or are you seeking tofind explanations in order to reach an understanding of a situation?Will knowledge be gained through impartial observation and/orexperimentation, or will you have to immerse yourself in the situationand make subjective interpretations or value-laden observations?Based on the conclusions you have reached from these questions, whichtype(s) of approach do you regard as appropriate for your researchproject? Discuss the reasons for your decisions in the light of what youhave read about the issues surrounding the nature of knowledge and theapproaches to research. What questions are raised about the potentialvalue of your research which will need to be further investigated?Step 3: Induction or deduction?Depending on your subject and on your own position in the debate aboutresearch, will your main argument be based on inductive or deductivereasoning, or will you take the hypothetico-deductive approach? In muchqualitative social science research, concepts and theories are developedon the basis of observations and consultations – a classic inductive modeof operation. In economics, commerce, marketing, and many other appliedsubjects, there is plenty of scope for a deductive approach, e.g. doKeynesian theories really work? The natural sciences and manydisciplines based on scientific method commonly use hypothetico-deductive reasoning. The way you tackle your research problem and thekind of research methods you use will be intimately connected to yourmethod of thinking. Make sure you are clear about it.Step 4: ‘Bedding in’ your approachWhat I mean by ‘bedding in’ is that you provide a context for yourtheoretical approach by referring to sources and ways of thinking thatsupport it. It must be obvious to you by now that you are carrying forward,or challenging, the work of others. Your work will belong to a stream ofthought, a tradition or a particular perspective. The results of yourliterature review will enable you to trace this through by making referenceto those before you, and to their way of thinking. Your proposal will need tobriefly refer to these in the background section.
Consolidation and assessmentArrange a meeting with one of your tutors or your supervisor andexplain to him/her the results of your thinking which have beenprompted by the above tasks. You should be able to demonstratethat you have grasped the significance of the issues surroundingthe activities of research and the nature of knowledge, and howthese issues are relevant to your own research project. Yourpowers of argument will be put to the test!The outcome of the discussion should help you to assess theintellectual rigour on which your research project is based, andprepare you to investigate appropriate options for research designand to explore the question of quality in research. These subjectswill be covered in the next chapter.Further readingYou can go into much greater detail about the philosophy ofknowledge and research if you want to, but I suspect that you willnot have enough time to delve too deeply, unless your researchtopic focuses on some of these issues. However, it is advisable tohave a good general knowledge of the debate about thephilosophy of scientific knowledge and its detractors, in order toplace your research within the philosophical context. Whencompiling this chapter, I found the following books useful and wellworth a browse. The titles give an indication of the subject tackled.I have put the more approachable ones first.Two good introductory books to start with:Thompson, M. (2006) Philosophy, 4th edn. London:Hodder (Teach Yourself series).This is a simple introduction to philosophy which explains themain terminology and outlines the principal streams of thought.
Warburton, N. (2012) Philosophy: The Basics, 5th edn.London: Routledge.Another slim volume as a general introduction.The following concentrate on scientific approaches and dilemmas:Okasha, S. (2016) Philosophy of Science: A Very ShortIntroduction, 2nd edn. Oxford: Oxford University Press.A clear, non-technical introduction to the philosophy of science.Chalmers, A. (1999) What Is This Thing Called Science?3rd edn. Milton Keynes: Open University Press.This goes more into scientific method.The following concentrate on scientific approaches and dilemmas.The first three are general discussions about the nature ofscience, whilst the remainder are important texts that have greatlyinfluenced the debate about science: influential books for theenthusiast!Kuhn, T.S. (2012) The Structure of Scientific Revolutions,3rd edn. Chicago: Chicago University Press.Popper, K. (1992) The Logic of Scientific Discovery(Routledge Classics). London: Routledge.Feyerabend, P. (1993) Against Method: Outline of anAnarchistic Theory of Knowledge, 3rd edn. London:Verso.Medawar, P.B. (2009) Induction and Intuition in ScientificThought. Abingdon: Roultlege.Medawar, P.B. (1986) The Limits of Science. London:Methuen.
Bohm, D. and Peat, F. (2011) Science, Order andCreativity. London: Routledge.Hacking, I. (ed.) (1981) Scientific Revolutions. Oxford:Oxford University Press.See pp. 60–79, ‘The corroboration of theories’ by H. Putnam; andpp. 157–67, ‘How to defend society against science’ by P.Feyerabend.Lakatos, I. and Musgrave, A. (eds) (1970) Criticism andthe Growth of Knowledge. Cambridge: CambridgeUniversity Press.See pp. 91–196, ‘Methodology of scientific research programmes’by I. Lakatos.Archer, M., Bhaskar, R., Collier, A., Lawson, T. andNorrie, A. (1998) Critical Realism: Essential Readings.London: Routledge.And the following look at the foundations of the social sciences:Best, S. (2003) A Beginner’s Guide to Social Theory.London: Sage.Seale, C. (2012) Researching Society and Culture, 4thedn. London: Sage.This is the latest edition of this book. See particularly Part 1.Zohar, D. and Marshall, I. (1993) The Quantum Society.London: Bloomsbury.Skinner, Q. (ed.) (1990) The Return of Grand Theory in theHuman Sciences. Cambridge: Cambridge UniversityPress.
3 Using Language andUnderstanding ArgumentsAimsTo show how language is the basis for communicationTo investigate the role of concepts and theoryTo explain how to use argument to make your case convincingIntroductionThis chapter introduces a brief survey of the main components ofenquiry – concepts and theory. It aims to clarify the meaning andsignificance of terms constantly used in research. Using theseterms correctly will make it easier for you to communicate on amore technical level when discussing and analysing your researchproblem. This should help you to define your problem moreaccurately and help to indicate what research approach will beappropriate. It is worth noting that the precise meaning of manytechnical terms, such as ‘theory’, ‘law’ and ‘hypothesis’, oftendepends on the context in which they are used and also on theinterpretation of their meaning by the writer.This is followed by a review of how to use these concepts to forman argument that will provide the backbone of your research.Argument lies at the heart of communication about research. Youwill need to convince the reader of your writing at every stage ofyour research project: when you formulate your research problemand the reasons for doing the research, when you discuss therelevant background information, when choosing appropriateresearch methods for collecting and analysing data, and whencoming to conclusions at the end of the research. That is why it isnecessary to have a clear idea about the nature of argument andhow to judge its quality.
Concepts and theoryConceptsConcepts are general expressions of a particular phenomenon,e.g. cats, dogs, anger, achievement, alienation, speed,intelligence, socialism. If we examine these examples moreclosely, it is evident that each is a word that represents an idea.Everybody, everywhere makes use of concepts. Of course, withinthe same culture, many concepts are commonly shared, such asbaby, hate, justice; others, however, are only used andunderstood by small groups of people, specialists or members ofprofessions, for instance idioglossia, interstitial condensation,anticipatory socialization.ConceptsIt is through the use of concepts that we are able to impose somesort of coherent meaning on the world: it is through them that wemake sense of reality, and perceive order and coherence. We useconcepts to communicate our experience of the environmentaround us. Our perception of our surroundings is therefore highlydependent on the scale of our knowledge and our familiarity with awide range of concepts. The more we know and use, the moresense we can make of the data we pick up and the more certainwe will be of our perception (and cognitive) grasp of whatever is‘out there’. If it is the case that we see and understand the worldonly through the concepts that are available to us, then it followsthat people who have a different repertoire of concepts will beinclined to perceive the ‘same’ objective reality differently. Alawyer examining a case will utilize a very different range ofconcepts from, say, the simple and common knowledge notions ofthe lay person in that context; and a visitor to the modern urbancontext from a primitive rural culture would be as confused by thehustle and bustle of the modern city as would be the mythical manin the moon.
Researchers in natural sciences will try to define the meaning ofconcepts very precisely. This, in many instances, is possible withinthe terms of the research. Social scientists, on the other hand,often recognize that the concepts within their model may be basedon opinions, emotions, values, traditions, rules, etc. and cannot bepinned down in the same way. How some precision can be gainedin this context is discussed later in this chapter.StatementsThe use of concepts on their own in research is limited. If oneassumes that the aims of research are to provide a system ofclassification (a typology), offer explanations, make predictions,and acquire a sense of understanding, then it is evident that it isonly the first, providing a typology, which can be achieved usingconcepts alone. The remaining aims (offering explanations,making predictions and acquiring a sense of understanding) mustbe expressed in the form of statements – statements thatcontain concepts. It follows that, in most research, the meaningand value of concepts cannot be assessed apart from their use instatements. In other words, according to Reynolds, ‘the scientificvalue of concepts can only be judged in terms of the scientificutility of the statement containing them’ (2016, p. 45). However, itis possible to evaluate concepts by an assessment of the claritywith which they are described. Clarity can be measured by thedegree of agreement about a concept’s use and its meaningamong the users of the concept.StatementsMore exploration of types of statements is made later in thischapter in relation to their use in argument.SymbolsConcepts, which can perhaps be defined as a unit ofunderstanding, are expressed and communicated throughsymbols in the form of language, spoken or written, natural or
artificial (such as mathematics). To convey meanings accurately, itis obviously essential that the sender and receiver of the symbolsshould agree on their meaning.SymbolsSymbols, in the form of words in spoken or written language, arenormally referred to as terms, while the word ‘symbol’ is generallyreserved for a mathematical sign, e.g. £, +, %. It is possible toidentify two types of terms used in any language, natural orartificial. First, there are primitive terms. These cannot bedescribed by using other terms, but rely on a shared agreementas to their meaning. Secondly, there are derived terms. These canbe described by the use of primitive terms. Obviously, themeaning of primitive terms is more difficult to convey than that ofderived terms, as one can explain them only by indicatingexamples and non-examples of the term.Confusion can often occur when a term or symbol, however closelydefined, is used in a particular context, but has different meanings inother contexts. Take for example the term ‘architecture’, which hasone meaning in the built environment, but represents a completelydifferent concept in the context of information technology. As anotherexample, the symbol p will have different meanings for amathematician and a Greek scholar.This problem in research is common, particularly if the term or symbolis utilized in everyday language for other concepts. Often this addedmeaning, unintended by the writer, can dramatically change themeaning of a statement. To overcome this problem, researchers oftenuse abstract symbols, invented or seldom-used words, or even Latinor Greek phrases to label a concept. However, such symbols or termsare often criticized for being abstract, sterile or difficult to read.Unfortunately, as long as readers are careless about theirunderstanding of symbols or terms and insist on an ‘easy read’, thereseems to be no simple solution to this problem. Problems of a moreserious kind are caused when the writer does not define his/hersymbols or terms in the first place.
Figure 3.1 Confusion can often occur when a symbol has differentmeanings in other contextsAbstract and concrete conceptsConcepts have characteristics that can be defined as abstract orconcrete. According to Reynolds, abstract concepts are thoseconcepts that are completely independent of a specific time orplace. In other words, abstract concepts are not related to anyunique spatial (location) or temporal (historical time) setting. Theyare often developed by and closely associated with particulartypes of research, e.g. marginality (ethnographic research),rhetorical organization (discourse analysis). If a concept is specificto a particular time or place, then it is considered concrete(Reynolds, 2016, p. 49).
Exercise 3.1Consider the following concepts and decide whether each is concrete orabstract. If it is concrete, can you describe what it is specific to? Here aretwo examples:Temperature – abstractTemperature of the sun – concrete – specific to locationThe concepts are:1. three days2. 4 December 1967 to 6 December 19673. attitude4. what Mr Smith thinks of Mr Jones5. social system6. the Women’s Association7. group8. an afternoon in London.You will notice that in each case of a concrete concept, it is possible toidentify the abstract concepts contained therein; the concreteconcepts are instances of abstract concepts.Only one meaning of the word ‘abstract’ (independence from time andspace etc.) has been discussed so far. There is another meaning,which relates to the ‘depth’ of abstraction. Although it may be seenthat two ‘abstract’ concepts are independent from time and space, oneconcept may nevertheless be more abstract than the other. If oneconcept is included within the meaning of the other, the second ormore general concept is considered the more abstract. Reynolds(2016, p. 50) explains this by the following example. Consider theconcept ‘sentiment’, an emotional disposition directed towardsanother, and the concept ‘liking’, a positive feeling for another. Theconcept ‘sentiment’ may be considered as including ‘liking’ along with‘love’, ‘respect’ and ‘admiration’, but also ‘hate’, ‘loathing’ and others.In this example, ‘sentiment’ is the more abstract concept, for itencompasses the meaning of ‘liking’.
When concepts are used in research, it is important that their level ofabstraction is appropriate to the type of research undertaken.Concepts can be too abstract or too broad to be useful.Concept measurementOne of the most important features of a scientific statement in thenatural sciences is that it should have relevance to the real world:it should be possible to compare the statement to somephenomenon or phenomena. Normally, this is achieved byidentifying palpable events or occurrences that can be perceivedto demonstrate instances of the theoretical concepts. This meansthat it is essential that some of the theoretical concepts relate tosensory impressions in concrete situations. It is not obligatory,however, to be able to measure all of the concepts in a set ofstatements or theory. After all, it is not possible to actually see anatom or an electron. What can be seen are the effects that can beascribed to electrons or atoms. Theoretical concepts that cannotbe directly measured in a real-life setting are sometimes called‘hypothetical constructs’.In the social sciences, concept measurement is commonly adifficult issue. It is impossible to be so precise when dealing withmatters of belief, customs, behaviour and values. It may beobvious that there are differences in, say, levels of commitment,strength of belief, etc. A measurement of these levels will dependon the communication of the person (or persons, institutions, etc.)and the perception and personal judgement of the researcher. Nosimple reading of a thermometer dipped into the hot bathwaterhere! However, social data can be collected which are amenableto some forms of measurement, for example economic data,population statistics and suchlike.Operational definitionIn order to determine the existence of a theoretical concept in areal-life setting, one must devise special types of definitions toprovide the necessary operational instructions. These types ofinstructions are called operational definitions.
Operational definitionsReynolds defined these, as in Checklist 3.1, as procedures thatneed to be performed. Note that the stress is on activities to beundertaken, rather than the nature of the impressions.Checklist 3.1 Defining operational definitionsA set of procedures that describes the activities an observershould perform in order to receive sensory impressions (sounds,visual or tactile impressions, etc.) that indicate the existence ordegree of existence of a theoretical concept. (Reynolds, 2016, p.52)The significant feature of operational definitions is that they areabstract, that is, independent of time and space. This enables them tobe used in different real-life situations and at different times.For example, it may be possible to measure a group of people’sindividual state of fitness (the overall physical condition of the body) inany of three ways:1. Allow trained observers, e.g. sports teachers or trainers, toindicate their judgement on the basis of a series of performancetests.2. Collect measures of physiological activity, such as bloodpressure, breathing rate or pulse rate, at rest and during action.3. Give the individuals a questionnaire and study the pattern ofactivities that are regularly undertaken (a ‘yes’ answer to thequestion ‘Do you run more than 5 miles every week?’ may beconsidered as an indicator of greater fitness than ‘no’).Each of these procedures can be seen to relate to the theoreticalconcept of fitness and can be used to measure fitness in differentsituations. Each gives a description of what needs to be done(without going into technical details) in order to gain a sensoryimpression related to someone’s degree of fitness, i.e. ratings ofexpert observers, measurements related to physicalcharacteristics, or the individual’s responses on a questionnaire.
Consider how measuring a group of individuals’ levels of fitness inthis way is different to assessing how fit they generally feel, i.e.measuring not fitness but the feeling of fitness. The operationaldefinitions would be quite different and relate to more abstract andsubjective procedures of a qualitative nature rather thanquantitative measurement.In these cases it is important that the methods of assessment ormeasurement of these concepts are critically examined from thepoint of view of reliability. In the above example you could:1. question the subjects as to what they regarded fitness to be,and how near to that condition they felt they were2. observe their attitudes towards carrying out physical tasks,such as carrying heavy shopping bags or running for a bus3. analyse their conversations about the subject of fitness togauge their attitudes and feelings.These forms of assessment tend to rely heavily on individualinterpretation. So further attention should be given to the issue ofthe reliability of these measures. Subject error might occur if thesubject does not really understand the questions, or subject biasmight result if the subject tries falsely to impress the researcher asto how fit he/she feels. Observer error might also occur if theresearcher asks or observes some subjects at a particularlystressful time, perhaps after an exceptionally strenuous day atwork. Observer bias might result in uneven criteria being used,e.g. the conviction that overweight people must feel unfit.The relationship between theoretical concepts and the suitabilityof operational definitions for measuring the existence of thetheoretical concept is largely one of judgement. According toReynolds (2016, p. 55), a single operational definition can beevaluated by two criteria, as in Checklist 3.2.Checklist 3.2 Criteria for evaluating an operational definition1. Gauge its suitability as a measurement procedure.2. Assess its relation to a theoretical concept.
Levels of measurementIn 1946, S.S. Stephens suggested a hierarchy of levels ofmeasurement that has been widely adopted by statisticiansand researchers, not only for use in the measurement of conceptsand operational definitions, but also for measuring data. Hissystem comprises four levels of measurement, which he callsnominal, ordinal, interval and ratio levels of measurement. Basicto all measurement types is the essential nature of the concepts,operational definitions or data, which lend themselves tomeasurement at one level or another according to theircharacteristics and the manner in which they are considered andclassified for analysis. In many textbooks, the term measurementseems to be used interchangeably with the term quantification.Levels of MeasurementQuantificationThe nominal level of measurement is very basic andunrefined. Its simple function is to divide the data into separatecategories that can then be compared with each other. By firstgiving names to or labelling the parts or states of concepts or bynaming discrete units of data, we are then able to measure theconcept or data at the simplest level. For example, manytheoretical concepts are conceived on a nominal level ofmeasurement. For example, buildings may be classified into manytypes, e.g. commercial, industrial, educational, religious. Manyoperational definitions are on a nominal level, e.g. sex (male orfemale), marital status (single, married, separated, divorced orwidowed). This applies in the same way for some types of data,e.g. dividing a group of children into boys and girls, or into fair-haired, brown-haired or black-haired children, and so on.Nominal level of measurementIn effect, different states of a concept or different categories ofdata which are quantified on a nominal scale can only be labelled,
and it is not possible to make statements about the differencesbetween the states or categories, except to say that they arerecognized as being different.We can represent nominal data by certain graphic and statisticaldevices. Bar graphs, for example, can be appropriately used torepresent the comparative measurement of nominal data i.e.numbers of cases in each category. By measuring this type ofdata, using statistical techniques, it is possible to locate the mode,find a percentage relationship of one subgroup to another or ofone subgroup to the total group, and compute the chi-square. (Wewill discuss the mode and chi-square later (in Chapter 7); they arementioned here merely to indicate that nominal data may beprocessed statistically.)The ordinal level of measurement is used if a concept isconsidered to have a number of states, or the data have a numberof values that can be rank ordered, and it is assumed that somemeaning is conveyed by the relative order of the states. Using theordinal level of measurement implies that an entity beingmeasured is quantified in terms of being more than or less than,or of a greater or lesser order than, a comparative entity, oftenexpressed by the symbols < or >.Ordinal level of measurementFor anyone studying at school or at university, the most familiarordinal measures are the grades used to rate academicperformance. An A always means more than a B, and a B alwaysmeans more than a C, but the difference between A and B maynot always be the same as the difference between B and C interms of academic achievement. Similarly, we measure level ofeducation grossly on an ordinal scale by saying individuals areunschooled, or have an elementary school, a secondary school, acollege or a university education. Likewise, we measure membersof the workforce on an ordinal scale by calling them unskilled,semi-skilled or skilled.
Most of the theoretical concepts in the social sciences seem to beat an ordinal level of measurement. In summary, ‘ordinal level ofmeasurement’ applies to concepts that vary in such a way thatdifferent states of the concept can be rank ordered with respect tosome characteristic.The ordinal scale of measurement expands the range of statisticaltechniques that can be applied to data. Using the ordinal scale,we can find the mode and the median, determine the percentageor percentile rank, and test by the chi-square. We can alsoindicate relationships by means of rank correlation.The interval level of measurement has two essentialcharacteristics: it has equal units of measurement; and its zeropoint, if present, is arbitrary. Temperature scales are one of themost familiar types of interval scales. In each of the Fahrenheitand Celsius scales, the gradation between each degree is equalto all the others, and the zero point has been establishedarbitrarily. The Fahrenheit scale clearly shows how arbitrary is thesetting of the zero point. At first, the zero point was taken byGabriel Fahrenheit to be the coldest temperature observed inIceland. Later he made the lowest temperature obtainable with amixture of salt and ice, and took this to be 0 degrees. Among themeasurements of the whole range of possible temperatures,taking this point was evidently a purely arbitrary decision. It placedthe freezing point of water at 32 degrees, and the boiling point at212 degrees above zero.Interval level of measurementFigure 3.2 For anyone studying at university, the most familiarordinal measures are the grades used to rate academicperformance
Although equal-interval theoretical concepts like temperatureabound in the physical sciences, they are harder to find in thesocial sciences. Though abstract concepts are rarely inherentlyinterval based, operational measures employed to quantify themoften use measurement at an interval level. For example, attitudesare frequently measured on a scale like this:Unfavourable -4 -3 -2 -1 0 +1 +2 +3 +4 FavourableIf it is assumed that the difference between +2 and +4 is the sameas the difference between say 0 and -2, then this can be seen asan attempt to apply an interval level of measurement to thismeasurement procedure. This is quite a big assumption to make!The tendency for some social scientists to assume the affirmativeis probably because some of the most useful summary measuresand statistical tests require measurement on an interval level, e.g.
for determining the mode, mean, standard deviation, t-test, F-testand product moment correlation.Questions are also frequently raised about the unrealisticpreciseness of the responses. Are the meanings intended by theresearcher’s questions equivalent to those understood by therespondent? Is the formulaic choice of answers given compatiblewith what the respondent wishes to reply? I am sure youremember your reaction to attitude quizzes, where the answer ‘itall depends’ seems more appropriate to a question than any of themultiple-choice answers offered.The ratio level of measurement has a true zero, that is, thepoint where the measurement is truly equal to nought – the totalabsence of the quantity being measured. We are all familiar withconcepts in physical science that are both theoretically andoperationally conceptualized at a ratio level of measurement.Time, distance, velocity (a combination of time and distance),mass and weight are all concepts that have a zero state in aninterval scale, both theoretically and operationally. So, there is noambiguity in the statements ‘twice as far’, ‘twice as fast’ and ‘twiceas heavy’. Compared with this, other statements that use this levelof measurement inappropriately are meaningless, e.g. ‘twice asclever’, ‘twice as prejudiced’ or ‘twice the prestige’, since there isno way of knowing where zero clever, zero prejudice or zeroprestige are.Ratio level of measurementA characteristic difference between the ratio scale and all otherscales is that the ratio scale can express values in terms ofmultiples of fractional parts, and the ratios are true ratios. A metrerule can do that: a metre is a multiple (by 100) of a centimetredistance, a millimetre is a tenth (a fractional part) of a centimetre.The ratios are 1:100 and 1:10. Of all levels of measurement, theratio scale is amenable to the greatest range of statistical tests. Itcan be used for determining the geometric mean, the harmonicmean, the percentage variation and all other statisticaldeterminations.
In summary, the simple tests for various kinds of concept and datameasurement are given in Checklist 3.3.Checklist 3.3 Simple tests for levels of measurementIf you can say that:one object is different from another, you have a nominal scaleone object is bigger, better or more of anything than another, youhave an ordinal scaleone object is so many units (degrees, inches) more thananother, you have an interval scaleone object is so many times as big or bright or tall or heavy asanother, you have a ratio scale.Exercise 3.2What level of measurement is appropriate for the following concepts andoperational definitions?1. Reading interest: poetry, fiction, biography2. Sex: girls, boys3. Voltage: 25 V, 75 V, 2000 V4. Intelligence quotient (IQ): 90, 102, 125, 1305. Exam grade: C-, C+, B, B+, A6. Comfort level: very uncomfortable, uncomfortable, comfortable, verycomfortable7. Atmospheric pressure: 0.5 bar, 1.3 bar, 2.7 bar, 10 bar8. Wealth: destitute, poor, middle income, high income, wealthy9. Annual income: less than £1000, £1000–£4999, £5000–£9999,£10,000 and over10. Annual income: £3000, £7300, £9000, £17,00011. Dew point (point at which condensation occurs): 2 °C, 12 °C, 17 °C,24 °C12. Titles: knight, baronet, lord, viscount, earl, duke13. Trades: plumber, electrician, mason, carpenter14. Playbricks: red cube, yellow pyramid, red cylinder, blue sphere,yellow cube, red sphereIt will be seen that one of the principal advantages of usingquantification is that it permits the formulation of more precise
statements regarding the degree of association, or correlation,between two or more concepts. Exploration of the relationshipsbetween concepts is often one of the important activities inresearch. Different levels of measurement can be used tocategorize the different types of relationship between concepts. Anominal level of association is equivalent to the statement that twoconcepts are or are not related. An ordinal level of association isequivalent to the statement that concepts are positively, not ornegatively correlated.When data are analysed, such statements of association aregenerally related to operational definitions. Levels ofmeasurement, if correctly applied, are particularly useful whendiscussing the interrelations among more than two variables.We have discussed at some length the nature and characteristics ofconcepts, and how they can be measured. In your research, you willbe using many concepts and probably measuring some of them too.However, what will certainly be of most interest to you, and toeveryone else, is not just the meaning of the concepts, or theirmeasurement, but the nature and extent of the relationships betweenthem, which if explained will lead towards a greater understanding ofthe subject under scrutiny. This is where theory is important.TheoryCohen et al. (2017) maintained that, for professional scientists,science is seen as a way of comprehending the world; as ameans of explanation and understanding, of prediction andcontrol. For them, the ultimate aim of science is theory. Theoriesare expressed in the form of statements, i.e. theoreticalstatements.Theory
Kerlinger (1999, p. 27) defined theory as ‘a set of interrelatedconstructs (concepts), definitions and propositions that presents asystematic view of phenomena by specifying relations amongvariables, with a purpose of explaining and predictingphenomena’. Theory effectively combines diverse and isolatedpieces of empirical data to create an intelligible conceptual modelthat is capable of being more generally applied. Mouly (1978, p.42) expressed it like this: ‘if nothing else, a theory is aconvenience – a necessity, really – organizing a whole slough ofunsorted facts, laws, concepts, constructs, principles, into ameaningful and manageable form. It constitutes an attempt tomake sense out of what we know concerning a givenphenomenon.’Additionally, however, theory provides a useful platform fromwhich to launch a quest for information and discoveries, and animpetus for research. New hypotheses can be suggested and newquestions revealed. Theory formulation leads to the identificationof important areas that require further research, points out whereinformation is missing, and makes it possible for a researcher topropose the existence of hitherto unidentified phenomena.Figure 3.3 Theory formulation makes it possible for a researcherto propose the existence of hitherto unidentified phenomena
Theory varies greatly in status and quality, depending on the typeof discipline or subject area considered. In well-establishednatural sciences, for example in chemistry, theories tend to behighly developed and refined, while in other fields, for example inthe social sciences, they tend to be relatively undeveloped, lesswidely accepted and of more uneven quality. This is due to thepioneering quality of much theoretical work carried out incomparatively new disciplines. In the social sciences, Silverman(2012, p. 26) sees theories as living entities, developed andmodified by good research. However, because they instruct us tolook at phenomena in particular ways, theories, and the conceptson which they are based, are self-confirming, meaning that theycan never be disproved, as in natural science, but only found tobe more or less useful. The nature of a theory is therefore stronglyinfluenced by the level of maturity of the particular specialization.The early stages of a science must be dominated by empiricalwork, that is, the accumulation and classification of data. Only as
a discipline matures can an adequate body of theory bedeveloped.What criteria are there for judging the quality of a theory? Shownin Checklist 3.4 is what Mouly identified as the maincharacteristics of a good theory. You will notice that theory in thiscase is seen from a positivistic point of view, so his quality criteriacannot be generally applied to all theory in all forms of research.Checklist 3.4 Main characteristics of a good theory1. A theoretical system must permit deductions that can be testedempirically; that is, it must provide the means for its confirmationor rejection. One can test the validity of a theory only through thevalidity of the propositions (hypotheses) that can be derived fromit. If repeated attempts to disconfirm its various hypotheses fail,then greater confidence can be placed in its validity. This can goon indefinitely, until possibly some hypothesis proves untenable.This would constitute indirect evidence of the inadequacy of thetheory and could lead to its rejection (or more commonly to itsreplacement by a more adequate theory that can incorporate theexception).2. Theory must be compatible with both the observation andpreviously validated theories. It must be grounded in empiricaldata that have been verified and must rest on sound postulatesand hypotheses. The better the theory, the more adequately itcan explain the phenomenon under consideration, and the morefact it can incorporate into a meaningful structure of ever-greatergeneralizability.3. Theories must be stated in simple terms; the theory is best thatexplains the most in the simplest way. This is the law ofparsimony. A theory must explain the data adequately and yetmust not be so comprehensive as to be unwieldy. On the otherhand, it must not overlook the variables simply because they aredifficult to explain (Mouly, 1978).ParsimonyThis sounds all very well for the natural sciences, but some ofthese conditions cannot be achieved in the social sciences. Inpoint 2, it may be impossible to ground the theory on empirical
data that have been verified in the sense of measurement andrepeated observations. Following this, it may be difficult thereforeto objectively test the validity of its various hypotheses asdemanded in point 1.What is important to stress, though, is the relationship betweendeveloping theory and previously validated theory as mentioned inpoint 2. The theoretical background to one’s enquiries willdetermine how one looks at the world. As Quine (1969) argued,our experience of the world of facts does not impose any singletheory on us. Theories are underdetermined by facts, and ourfactual knowledge of the external world is capable of supportingmany different interpretations of it. The answer to the question‘what exists?’ can only receive the answer ‘what exists is whattheory posits’. Since there are different theories, these will positdifferent things. There will always be more than one logicallyequivalent theory consistent with the evidence we have. This isnot because the evidence may be insufficient, but because thesame facts can be accommodated in different ways by alterationsin the configuration of the theory (Hughes and Sharrock, 1997, pp.88–91).One philosopher of science expressed it this way: ‘it is generallyagreed … that the idea of a descriptive vocabulary which isapplicable to observations, but which is entirely innocent oftheoretical influences, is unrealizable’ (Harré, 1972, p. 25).Therefore, without a theory, one could argue that phenomenacannot be understood, and research cannot be carried out withouta theoretical underpinning: ‘models, concepts and theories areself-confirming in the sense that they instruct us to look atphenomena in particular ways. This means that they can never bedisproved but only found to be more or less useful’ (Silverman,2012, p. 29).It follows then that all theories must, by their very nature, beprovisional. However sophisticated and elegant a theory is, itcannot be all-encompassing or final. The fact that it is a theory, anabstraction from real life, means that it must always be subject topossible change or development, and in extreme cases, evenreplacement.
Figure 3.4 All theories must, by definition, be provisionalTheoretical statementsReynolds (2016, Chapter 5) claimed that scientific knowledge isbasically a collection of abstract theoretical statements. Heidentified three different conceptions of how sets of statementsshould be organized in order to constitute a ‘theory’. The first,which plays a very important part in science, is the view thatknowledge is essentially a set of laws – statements that can beconsidered, for the time being anyway, to be the ‘real truth’. Thesecond conception of what constitutes a theory is axiomatic inform, i.e. based on self-evident truths. The third, referred to as thecausal process form, sees theory as made up of an interrelatedset of definitions and statements. The following summary of hisdescriptions of theoretical statements will indicate some of theirmain characteristics.Set of laws theory
A theoretical statement is simply the declaration of a theory,making an assertion of how things are or how they are related orreact together. It will be useful at this stage to introduce a list ofterms used for theoretical statements which will provide a kind ofhierarchy, with at the lowest extreme those statements which areconjectural and untested, and at the highest extreme thosestatements which are generally recognized to be true.In the early stages of a research project, tentative theoreticalstatements are often made which are products of the researchproblem. Preliminary investigation into the problem might suggestreasons behind it, or relationships between factors creating orinfluencing it. Sometimes, especially in the social sciences, owingto the qualitative nature of the data, it is not possible to formulatea theoretical statement which can be rigorously tested, in thesense that clear evidence for support or rejection will result.Nonetheless, a statement can indicate a clear direction and scopefor the research work needing to be carried out. In this case,theoretical statements can be formulated as researchquestions or propositions.Research questionsPropositionsIf, however, the reasons or relationships are expressed in atheoretical statement which has not yet been tested against datacollected in a concrete situation, but which it is possible to test,the statement is called a hypothesis. The source of the hypothesismay be a variation of a law or a development of axiomatictheory, or may be initiated by a scientist’s informed guess orintuition; but its essential feature is that it can be compared withexperimental or observational data collected in the real world. Itshould be possible to devise tests for the hypothesis which willprovide clear evidence to support it, or lead to its rejection.Axiomatic theory
If research reveals a similar pattern of events in a number ofempirical studies, the pattern is often summarized as anempirical generalization – a generalization based on severalempirical studies. Obviously, because they summarize patterns inempirical research, all the concepts in an empirical generalizationmust be directly measurable.Empirical generalizationLawsScientists’ confidence in laws is considerably greater than it is inempirical generalizations. As already noted, laws are statementsthat describe relationships. They are so well supported byevidence, and confidence in their reliability is so strong, that theyare considered to express the ‘truth’ in the cases to which theyapply.Paradigm is a term used not to describe a particular type oftheoretical statement, but rather to indicate the overall effect of theacceptance of a particular general theoretical approach, oftenexpressed as a law or a theory (e.g. Newton’s laws of mechanicsand of gravitation, Einstein’s theory of relativity), and the influenceit has on the scientists’ view of the world. As a result, actualscientific practice, following from the acceptance of these scientificlaws and theories, including application and instrumentation,provides models from which spring particular coherent traditionsof scientific research. The basic innovative work has beencompleted, and there is a consensus among the scientific bodythat these theories represent a firm foundation for furtherinvestigation. This ‘normal’ scientific work is concerned withresearch that searches for evidence in the real world for theworkings of the theory, enumerates and measures the theorymore precisely, and explores the validity of the theory in relatedfields.A summary of this discussion is given in Checklist 3.5.
Checklist 3.5 Statements, hypotheses, empirical generalizations, lawsand paradigmsAt its most tentative, a theoretical statement can be formulatedas a research question or proposition in order to direct theresearch efforts.If there is as yet no empirical evidence for or against atheoretical statement, but it can be tested empirically, it is calleda hypothesis.If there is moderate support, it is called an empiricalgeneralization.If the support is overwhelming, it is called a law. Be warned thatsince scientists have different standards for evaluatingtheoretical statements, one person’s law may be anotherperson’s hypothesis.When a system of laws is commonly accepted, it leads to asanctioning of particular ways of thinking and methods ofinvestigation: this is called a paradigm.Figure 3.5 Relational statements form the bedrock of scientificknowledge
Language and argumentThe medium of words, spoken or written, is by far the mostcommon method of communication among humans. Allphilosophical arguments are based on language and theinterpretation of meanings, as is research of almost every kind.Most of our thinking is done in words. It is only when we expressan idea in a sentence that we can appreciate what it actually is,how it can connect with other ideas, whether it is true or has avalue when compared with other ideas. And, most of the problemswe face are expressed in words and the information we use tocombat these problems comes to us in words.Argument lies at the heart of research. Evidence about a situationis amassed in the form of data that is analysed and interpretedusing argument to convince the reader of the reliability of theconclusions. We cannot develop critical skills unless we have
developed the ability to handle language: investigating anddefining meanings, appreciating the effects of grammatical forms,and understanding the thread of an argument through anextended piece of text. For example, doing a literature review is alarge exercise in analytical reading. All the books, articles,websites, etc. that you read will contain arguments in order to tryto convince you, the reader, of the truth of the message orinformation imparted. Not all of these arguments will be sufficientlyrigorous to justify the claims made, and it is up to you to detect theshortcomings and inconsistencies. It is logic that lies at the heartof all arguments.The logic described in this chapter is what is called traditionallogic, originating with Aristotle in the fourth century BC rather thanmodern or mathematical logic, particularly associated with Russelland Frege. The former is expressed in words as opposed to thelatter that is expressed in mathematical symbols – too complex toexplain here.Statements: existent and relationalAccording to Reynolds (2016, pp. 67–76), statements can beclassified into two groups: those that state that a concept exists,and those that describe a relationship between two concepts.Existence statements state that a concept exists, and theyprovide a typology or a description. Existence statements claimthat instances of concept exist in the real world. For example,here are some statements that make existence claims:Existence statementsThat object is a cow.That cow is brown and black.That object is a human.That human has a high level of intelligence.That (event) is a traffic jam.
Each of these statements follows the same basic pattern: itprovides a concept, identifies it by a term, and applies it to a thingor an event. The above are examples of existence statements intheir simplest form. They can, however, be more complicatedwithout losing their basic form. For example:If1. there are 40 or more individuals in group X2. each individual plays an orchestral musical instrument, and3. each individual cooperates in X to perform pieces from thesymphonic repertoire then the group X is a symphonyorchestra.Existence statements can be ‘right’ or ‘wrong’ depending on thecircumstances. Take for example the rather abstract statement ‘Itis 5 o’clock here.’ This can be seen to be correct anywhere once aday. If, however, we state in a more concrete fashion that ‘It is 5o’clock on 15 November in London’, this can only be correct onceand in one place. It can thus be seen that the level of abstractionof a statement has a powerful influence on its potential forcorrectness, i.e. the more abstract a statement, the more capacityit has to be right, and conversely, the more concrete a statement,the more capacity it has for being wrong.Relational statements impart information about a relationshipbetween two concepts. By referring to the instance of oneconcept, they state that another concept exists and is linked to thefirst. For example:Relational statementsIf a person is an acrobat, then he will be agile.Relational statements form the bedrock of scientific knowledge.Existence statements can only provide a typology or aclassification of objects or phenomena by applying definitions tothe world around us; we must rely on relational statements to
explain, predict and provide us with a sense of understanding ofour surroundings.There are two broad classifications of relational statements. Thefirst describes an association between two concepts, and thesecond describes a causal relationship between two concepts.The above relational statement can be taken as an example of anassociational statement:If a person is an acrobat, then he will be agile.By slightly changing the wording of this sentence we cantransform it into a causal statement:Becoming an acrobat will make a person be more agile.That is, becoming an acrobat will cause that person to be moreagile.With associational statements, when degrees of associationat the interval or ratio level of measurement are used, the word‘correlation’ is often employed to refer to the degree ofassociation.Associational statementsThere are three types of correlation between two concepts:positive, e.g. strong people are muscular (and vice versa), i.e.high value in one concept associated with high value in secondconcept, or low value associated with low valuenegative, e.g. grass at low altitudes grows longer, i.e. low valuein one concept associated with high value in second conceptnone, e.g. men and women have equal rights in a democracy,i.e. no information about associated high or low values in eitherconcept.
The degree of association is often measurable and is usuallyexpressed by +1.0 as maximum positive correlation, 0.0 as nocorrelation and -1.0 as maximum negative correlation.In causal statements, which describe what is sometimescalled a ‘cause and effect’ relationship, the concept or variablethat is the cause is referred to as the ‘independent variable’(because it varies independently), and the variable that is affectedis referred to as the ‘dependent variable’ (because it is dependenton the independent variable). An example is ‘smoking a lot makesone ill’, where ‘smoking a lot’ is the independent variable and ‘ill’ isthe dependent variable.Causal statementsIt may be possible to examine causality between two or moreconcepts if quantification is possible. Three ways of measuring thevariation of the dependent variable should be considered:the direct variation effect of each variablethe interaction of the variations of variablesvariations caused by mistakes and errors of measurements.Causal statements can be deterministic, meaning that undercertain conditions an event will inevitably follow, e.g. ‘if you dropan apple, it will fall’. However, it is not always possible to be socertain of an outcome, so a probabilistic statement might be moresuitable, e.g. ‘if parents are intelligent, their children are likely tobe intelligent too’. A quantification of the order of probability maybe possible, e.g. ‘with a probability of 0.7’.The classification of statements described is summarized inDiagram 3.1.Diagram 3.1 Statements
Statements can be made on three levels of abstraction:1. Theoretical statements. These are abstract statements thatare based on theoretical concepts, e.g. ‘bodily comfortdepends on environmental conditions’.2. Operational statements. These are less abstract in that theyare based on the definitions of theoretical concepts which arecapable of measurement, e.g. ‘the rate of heartbeat relates tothe surrounding still air temperature and the level of activity’.3. Concrete statements. These are based on specific findings,i.e. the measurements themselves, e.g. ‘the heart beats at102 beats per minute at a surrounding air temperature of 32degrees Celsius at an energy consumption level of 42kilocalories per hour’.Levels of abstractionDiscourse and argument
Statements on their own provide information on discrete units.When they are strung together to form a larger structure, they areoften referred to as a discourse. There are many different formsof discourse, depending on the use to which language is put. Insome discourses, statements are not merely presented for ourinformation (or misinformation, as the case may be); they areconnected in a specific, logical way. Some of the statements areoffered as reasons for others. This kind of discourse is termed anargument. It is a discourse that not only makes assertions but alsoasserts that some of those assertions are reasons for others.DiscourseIn the case of ordinary speech the term ‘argument’ is often usedwhen referring to a dispute, or a situation in which people whohold different views on some controversial subject try to bring theother person around to their way of thinking. But as a technicalterm in logic, argument is a special kind of discourse, in which aclaim is made that one or more particular statements should beaccepted as true, or probably true, on the grounds that certainother statements are true. Put another way: by the process ofreasoning, using the operation of logic, a conclusion is inferredfrom the statements given. An argument can be seen as theverbal record of this reasoning. You can check for the minimalingredients of an argument using Checklist 3.6.Checklist 3.6 Minimal ingredients of an argument1. At least one statement that is reasoned for (this is the conclusionof the argument)2. At least one statement that is alleged to support it (this is thepremise of the argument)3. Some signal or suggestion that an argument is under way(where this is a word or phrase, we shall call it the logicalindicator)There is a great variety of English expressions, some ratherroundabout, which can serve as logical indicators, so you have tobe on the lookout for them. Beardsley (1975, p. 14) has given a
useful list of the commonest ones, though you can probably thinkof a few more yourself. His suggestion that each of the words orphrases in Checklist 3.7 usually shows that the statement thatfollows is a conclusion can be easily tested by trying them out invarious sentences yourself. Such expressions are calledconclusion indicators.Checklist 3.7 Conclusion indicatorstherefore…hence …thus …so …implies that …entails that …which shows that …proves that …indicates that …consequently …allows us to conclude that …we may deduce that …points to the conclusion that …suggests very strongly that …leads me to believe that …bears out my point that …from which follows that ….Each of the words or phrases in Checklist 3.8 usually shows thatthe statement that follows is a reason. Again, you can test theseout by looking for them in existing text, or devising sentencescontaining these to form arguments. Such expressions are calledpremise indicators.Checklist 3.8 Premise indicatorsfor …since …because …for the reason that …in view of the fact that …on the correct supposition that …assuming, as we may, that …
may be inferred from the fact that …may be deduced from …as shown by …as indicated by …as is substantiated by ….Even with the use of this list of logical indicators, it is sometimesdifficult to determine whether a discourse is an argument: thereare cases, especially in a complex piece of text, when it isimpossible to be sure whether the minimal ingredients of anargument are indeed present. Usually, the presence of a logicalindicator is quite decisive – but even here it is vital to establishthat the logical indicator is being used in a logical sense, ratherthan in some other sense.Exercise 3.3Can you recognize an argument when you see one?Which of the following sentences are statements and which arearguments? If statements, of what kind are they, and if associational, whattype of correlation do they state? If arguments, what are the premise orpremises, the conclusion, and the logical indicators (premise andconclusion indicators) in each?1. Since Parliament increased the amount of taxation on petrol, theamount of traffic on the roads has decreased appreciably.2. The criminal is not to blame for what he did. His sense of socialresponsibility was diminished by a deprived upbringing.3. If it is true that the outcome of the vote was the result of bribes, thenthe election should be held again.4. Why do I insist that London is still the best place in the world to live?Look at the unequalled variety of drama, music, art and educationalopportunity there.5. The train drivers have gone on strike because the railway companyis not offering them a sufficient pay rise.6. An informal atmosphere in the office environment is associated withlow stress levels in the workforce.7. The railway company could not be offering a sufficient pay rise to thetrain drivers, as they are threatening to go on strike.8. In Western wedding celebrations the age of the couple and theextent of the wedding celebrations are associated: the greater the
age of the couple, the smaller the celebrations tend to be, while thegreater celebrations are enjoyed by the younger couples.9. I do not regard Parliament as trustworthy, my reason being that itsmembers have misled us too often.10. The ban on fishing lobsters has been continued, so it follows thatlobsters will soon be unobtainable.11. Cod will soon be scarce. Halibut will be scarce too.Using argumentArgument is one of the basic elements in research. The quality ofthe argument used in the introduction of the research problem; theexamination and analysis of the problem; the presentation of thefindings; and the analysis and conclusions strongly determines thequality of the overall research project. It is important forresearchers, therefore, to have a clear understanding of thevarious types of argument, the ingredients of sound argument andof the logic contained therein, and the pitfalls that may beencountered by the unwary.In the following we will concern ourselves primarily with theanalysis of arguments, with their logical soundness orunsoundness, i.e. whether their reasoning is good, or how good itis. It can be said that when the reasoning is very good, anargument can be tantamount to a proof. It is not possible,however, to achieve this high standard in all circumstances. On apractical level, we must often deal with important issues at a lowerlevel of reasoning. In this situation it is essential to be able toestimate the level of soundness of the arguments.Once we have recognized the existence of an argument, the nextstep is to examine what it is arguing for, and how the argument isbuilt up. The first thing to establish is the point of the argument,that is, the conclusion.The simplest kind of argument consists of just one premise and aconclusion that is claimed to follow from it, or to be implied by it.An example in which each is stated in a separate sentence is thefollowing:
The most prominent political issues are those about which thepress writes the most. Therefore, journalists have a greatinfluence in the selection of political issues around whichpublic debate revolves.When analysing an argument, it is the logical indicators that wemust look to first for clues to the logical structure. You will find thatthe main problem in formulating an argument is making theelements of verbal texture – the syntax, the order of words andtopics, the connectives – clearly reveal the logical relationships ofthe argument. Beardsley (1975, p. 19) suggested that there weretwo fundamental rules to keep in mind, as in Checklist 3.9.Checklist 3.9 Rules for logical statements1. The rule of grouping. As far as possible, reasons for the sameconclusion should be kept together, and their similar logicalstatus called to the reader’s attention.2. The rule of direction. When there is a series of assertions, eachbeing a reason for the next one, the argument should move in asingle direction, so that the order of the words helps to remind usof the order of the thought.Exercise 3.4What is the conclusion of the following arguments? If the conclusion is notexplicitly stated, put it in your own words.1. We must accept that the chemist made a mistake in giving Mrs Smithtranquillizing pills instead of the contraceptive pills which sherequested. But how can she claim damages against him, when anyinconveniences she suffered are far outweighed by the birth of a fine,healthy eighth child?2. People are always complaining about the amount of traffic in cities.They do not realize what enormous costs would be involved inreducing traffic. We would need to introduce barriers to throughtraffic, build ring roads, improve public transport, restrict deliveries toshops, build new car parks, and employ an army of traffic wardens toenforce the new legislation. The price of everything would go up tocover the expense of reducing urban traffic.
3. Bicycles make far less noise than cars; they produce no carbonmonoxide; they can easily get through city traffic without violatingtraffic regulations; they can be parked anywhere; they are just as fastas cars in congested city centres; they require no petrol; and theycarry as many passengers (namely, one) as most commuter cars.Even if they have disadvantages, these facts ought to persuade theauthorities to provide more special facilities and designated routesfor the cyclist in town centres.4. In the long run the greatest medical danger lies in reducing infantmortality amongst poor people. Saving the lives of an increasingnumber of babies will cause the population of the underprivileged toexpand so quickly that widespread hunger and overcrowding arebound to result.5. It is difficult to disagree with the opinions of environmental expertsthat, because of the difficulty of disposing of radioactive waste, thedangers of transporting the waste, and the problems of processingradioactive substances, nuclear power is not the answer to ourenergy problems; it is to be noted that even politicians are beginningto take note of these views.It is worth noting in the above arguments how the rules of groupingand direction, if followed, are a great help in making the argumentreadily understandable and easy to read. When working out anargument yourself, ideas never come in an orderly fashion. Flashes ofinsight, repetitive mulling over the ingredients of the argument, andconstant shuffling of its components belong to the developmentalprocess of an argument. However, when the time comes tocommunicate it to others, it is well worth spending the time to organizeits presentation in the simplest and most lucid way possible.Deductive and inductive argumentsAs discussed previously in this chapter, arguments aretraditionally divided into two different types, deductive andinductive. Deductive and inductive arguments can be seen asseeking the truth from opposite directions. Through deductiveargument we infer the particular from the general, while throughinductive argument we infer general truths from the particular. This
is illustrated more easily by quoting a well-known classicalexample of deductive argument:All humans are mortal. – first premiseSocrates is human. – second premiseTherefore Socrates is mortal. – conclusionThe particular conclusion (Socrates is mortal) is inferred (validly)from premises, the first of which (all humans are mortal) is ageneral or universal proposition, while the second one is aparticular proposition (about Socrates).Compare this with another classical illustration of inductiveargument:Socrates is human and is mortal. – first pair of premisesXanthippe is human and is mortal. – second pair of premisesSappho is human and is mortal. – third pair of premisesTherefore probably all humans are mortal. – conclusionHere, a general or universal conclusion (probably all humans aremortal) is inferred from premises all of which are particularpropositions (e.g. Sappho is human and mortal). Note theimportant reservation about the strength of proof of the argument– ‘probably’. While this is a simply explained method ofdistinguishing between deduction and induction, it is not adequatein all cases.Although in every argument there is the claim that its premisesprovide some grounds for the truth of its conclusion, only adeductive argument involves the claim that its premises provideconclusive grounds. We use the technical terms ‘valid’ and‘invalid’ in place of ‘correct’ and ‘incorrect’.
A deductive argument is said to be valid when its premises, if true,do provide conclusive grounds for its conclusion; and, the otherway round, when premises and conclusions are logically related insuch a way that it follows that the premises could not possibly betrue if the conclusion was not true. Deductive argument must beeither valid or invalid. Determining whether it is one or the othermay be easy in short, succinct arguments, but often presentsmore of a problem when the argument is extended and complex.In comparison to a deductive argument, an inductive argumentcontains the claim that its premises only provide some support forthe conclusion, rather than furnishing conclusive grounds for itstruth. This results in the fact that inductive arguments can beneither ‘valid’ nor ‘invalid’ in the ‘black and white’ sense used fordeductive arguments. They can, however, be evaluated on thestrength of the support they provide for their conclusions by theirpremises. The stronger the support, the more likely that theconclusions will tend to be true.Exercise 3.5Using the above definitions, determine which of the following argumentsare deductive and which are inductive.1. I read about a boy who played professional soccer from the age of16. He must have been a good junior player.2. The car was priced at £5000, and I paid £500 in deposit; according tomy calculation I still owe £4500.3. The Minister of Industry looked pale as he emerged from the meetingwith the General Workers’ Union; it looks as if they gave him anultimatum.4. She will probably die soon; she is definitely getting weaker all thetime.5. There are traces of non-permitted chemicals in his blood; he hasobviously been taking stimulants.6. A car is a vehicle, so a car’s wheel is a vehicle’s wheel.7. There are more people in the world than there are hairs on any oneperson’s head; it follows that there must be at least two people withthe same number of hairs on their heads.8. Reptiles are cold-blooded; so lizards must be cold-blooded.9. This vase is ancient, has been found in a tomb by the Nile, iscovered in hieroglyphics; it must be an Ancient Egyptian artefact.
10. The referee said he infringed the rules; so he did infringe the rules.Deductive and inductive arguments can be seen as seeking thetruth from opposite directions. Through deductive argument weinfer the particular from the general, while through inductiveargument we infer general truths from the particular.Logic in argumentWe should be able to detect a correct logical structure in anargument to determine whether the argument is valid or invalid.So what are the characteristics of logic, which governs thestructure of argument? Hodges (2001) defined logic as the studyof consistent sets of beliefs. He added that some people preferredto define logic as the study of the validity or the correctness ofarguments. That there is no conflict in these views can bedemonstrated in the following discussion.The principal concern of the discipline of logic is consistency.However, there are many types of consistency, and logic isconcerned only with one of them – the compatibility of beliefs.This is very different to the type of consistency that a man displaysif he supports the same football team throughout his life, or whena racehorse wins every race he runs. The type of consistency wewill be considering is not to do with loyalty, faithfulness ormaintenance of standards, but is rather to do with how a set ofbeliefs can be shown to be consistent with each other. Hodgessums it up in this way:Consistencya set of beliefs is called consistent if these beliefs could all betrue together in some possible situation. The set of beliefs iscalled inconsistent if there is no possible situation in which allthe beliefs are true. (Hodges 2001, p. 1)
A good way to test whether you can distinguish between aconsistent and an inconsistent set of beliefs is to try the nextexercise.Exercise 3.6Which of the following sets of beliefs do you think are consistent?1. I’ve never sat on a horse in my life. But if I mounted one now, itwould only take me two minutes to become an expert horseman andbe able to win a show jumping competition.2. I knew I would never get ill. But somehow I just caught a disease.3. There is no financial crisis in Britain today – it is just a rumour that isput about by politicians who want to increase taxation in order to payfor the rapidly increasing national debt.4. Peter joined the sports club three years ago, and has become itsmost generous member. Last year he paid for the travel expenses ofall the teams.5. So many people travel abroad for their holidays. The English seasideresorts are not as popular as before.6. I think that killing animals is immoral and should be stopped. Weshould therefore not export live animals for slaughter.When we were discussing the difference between deductive andinductive arguments, we briefly mentioned the difference betweenthe concepts of validity and truth. These concepts, logical truthand validity, have been two of the most important notions in thehistory of reason, so we will examine them more closely. In fact,the elements of these notions must be clearly understood beforethere can be any worthwhile discussion of the processes ofargument.You can find two major types of true statements in discourse. Onekind is empirical truth, which means to say, derived fromexperience. This is the type of statement that people make everyday when they say that something is true or that it is a fact. Thiskind of true statement is also at the basis of all scientificobservations and historical assertions, for example, ‘the bat is atype of flying rodent’, ‘steam is emitted when water is boiled’, and‘Henry VIII had six wives’. However, in this section we will
consider the other type of true statement in more detail. Thissecond type is based on logical truth.Empirical truthLogical truthLogically true statements can be subdivided into three varieties.The first type is called trivial, because the truth revealed is soobvious that nothing new is learned from it: for example, ‘Today isSunday, therefore it is not Saturday’, or ‘If she is a daughter, thenshe must be a female’. These are known as tautologies: it ismanifest that these ‘truths’ are an inherent part of the semantics oflanguage, and do not really increase our knowledge about theterms. However, in more complicated or abstract statements, theymight in fact be very useful in making evident relations betweenwords which were not previously obvious.In arithmetic, statements are made about the relationshipsbetween numbers and arithmetic concepts. These mathematicalpropositions define the rules of arithmetic and are said to be trueby necessity. For example, it would be difficult to imagine that 2 +2 did not equal 4, and it would be senseless to assert otherwise.Figure 3.6 True by necessity
There is a third variety of true statements which logiciansfrequently speak of, namely those true by definition. There are allsorts of definitions, but it is not the truth of the definitions that thelogicians have in mind, it is more the logical form of the argument.It must be remembered that logic is concerned not with the truthor falsity of premises or conclusions, but rather with thecorrectness of arguments. Therefore we do not say that inductivearguments are true, incorrect or untrue; we say that they containvalid deductions, correct inductions or, at the other extreme,assorted fallacies.It follows, therefore, that the validity of deductive arguments isdetermined by their logical form, not by the content of thestatements they comprise. As mentioned previously, when thepremise or premises of such an argument are related to theconclusion in such a way that the conclusion must be true if thepremises are true, then the argument is said to be valid. Anyargument where this is not the case is called invalid (i.e. if there is
any possibility that the premises could be true and the conclusionfalse). See Salmon (2012, pp. 11–13 and p. 80) for a moredetailed explanation.ValiditySalmon points out that each of the following three combinations ispossible for valid deductive arguments:1. True premises and a true conclusion – for example:All diamonds are hard. TrueSome diamonds are gems. TrueTherefore some gems are hard. True2. Some or all premises false and a true conclusion – forexample:All cats have wings. FalseAll birds are cats. FalseTherefore all birds have wings. True3. Some or all premises false and a false conclusion – forexample:All cats have wings. FalseAll cats are dogs. FalseTherefore all dogs have wings. FalseAlthough some of the premises in the above arguments are false,if they were true the conclusions would have to be true. However,this cannot work the other way round: a deductive argumentcannot be valid if it has true premises but a false conclusion.Figure 3.7 The simplest kind of argument consists of just onepremise and conclusion
Fallacies in argumentLogic is the protection against trickery and sloppy thinking. Logicdeals with arguments that are based on reason. Mistakes arepossible and even frequent in applying forms of logical argument.These mistakes are termed fallacies. However, not all mistakes inargument are genuine mistakes; there are innumerable examplesof the calculated use of quasi-reasoning in order to convince orconvert the unwary.The recognition of fallacies is not new, many of them having beennoted as early as Aristotle. You can probably devise an argumentyourself which is entirely logical, whose validity is clearlydemonstrated by the conclusion being derived from its premises,and which carefully follows all the rules of syllogism, but which isbased on premises that are phoney, tricks and delusions. Thereare brilliant deceptions for getting people to accept all sorts offalse premises as true, and these tricks of argument are so
common that even when people realize that they are beinghoodwinked they tend to let it pass.The word ‘fallacy’ is often used in two ways: sometimes it is usedto describe any kind of attitude that is fraudulent or deceitful, andat other times it is used, in a narrower sense, to indicate adefective manner of reasoning or a wily or cunning form ofpersuasion. In the following analysis of fallacies, it is the secondmeaning which is taken, i.e. where an argument purports to abideby the rules of sound argument but in fact fails to do so. The mostimportant division is between the formal fallacies and the informalones.Formal fallacies have some error in the structure of thelogic. Although they often resemble valid forms of argument,the logical route only takes us from A to B by way of disjointedor missing paths. In brief, the fallacy occurs because the chainof reasoning itself is defective.Informal fallacies, on the other hand, often use validreasoning on terms that are not of sufficient quality to meritsuch treatment. They can be linguistic, and allow ambiguitiesof language to admit error, and leave out something needed tosustain the argument; permit irrelevant factors to weigh on theconclusion; or allow unwarranted presumptions to alter theconclusion which is reached.Formal fallaciesInformal fallaciesThe following exercise tests your ability to spot the types of fallacyaccording to the above two categories.Exercise 3.7Each of the following texts contains a fallacy. Can you spot what it is, andfit it into one of the above categories, and explain the nature of the fallacy?When you look up the answers in the back of the book, the category will
be given and you can check to see if you discerned the characteristics ofthe various types of fallacy.1. If I dance too much, I’ll be tired. Since I have not danced too much, Iwill not be tired.2. The duchess has a fine ship, but she has barnacles on her bottom.3. The ship of government, like any other ship, works best when thereis a strong captain in charge of it. This is why government bydictatorship is more effective.4. Look, you’re a lecturer. Your university decided to increase yourworking hours because they knew it will be good for lecturers. It musttherefore be good for you.5. Talk of the Loch Ness monster is nonsense. We know that it does notexist because every single attempt to find it has failed utterly.6. I’m not going to get a job. There will be all that extra responsibility,not to mention the loss of my freedom. Think of the costs of travellingto work and buying suitable clothing. Then there are the increasedworries about status.7. All musicians are really sensitive people. It happens that some reallysensitive people are not properly appreciated. So some musiciansare not properly appreciated.8. The small, domesticated carnivorous quadruped positioned itself insedentary mode in superior relationship to the coarse-textured rush-woven horizontal surface fabric.9. If the Americans wanted good trade relations, they would encouragethe production of specialist goods in other countries. Since they dosupport this type of production, we know that they do want goodtrade relations.10. I don’t think we should employ Mr Smith. I am told that he is a poorgolf player. Careless people are bad at golf, so I don’t think it is agood omen.Obviously, it is important to recognize fallacies when you read or listento people’s arguments. It is just as important to avoid fallacies in yourown writing (or to use them to best effect if you intend to deceive!).The above examples are only a small selection of types of fallacy:whole books are written on the subject, so if you would like to readmore about fallacy in all its guises, I can recommend the books byPirie (1985) as a light-hearted account (from which I adapted some ofthe above examples), and Fearnside and Holther (1959) for a moreserious and technical approach.
Classification and analogy in argumentThe idea of class plays a very fundamental role in much of ourthinking, deductive and inductive. We instinctively categorizemuch of what we experience in order to make sense of it, i.e. thisis dangerous, that is not. There are two ways of forming a class:by collection or by division. By collection we class things thatshare a common property (or properties), e.g. ‘having four legs’.Through division, we can further divide this collection intosubclasses, e.g. ‘brown quadrupeds’, ‘red quadrupeds’, ‘greyquadrupeds’, etc. Division not only creates new classes out of anold one, but also relates the new classes to the old one, i.e. theyare subordinated to it. Division is only one way of relating classesto each other. We can also relate classes to each other bysubsumption and by coordination.Subsumption goes in the opposite direction from division. We startwith a class, say a discarded tin can, and ask: of what larger classshall we consider this to be a subclass? We have variousalternatives to choose from: metal objects, refuse, usedpackaging, cylindrical objects, etc. When we divide a class intosubclasses, the subclasses between themselves have parallelstatus: they are coordinate classes. When we subsume one classunder another, the subsumed class is at least implicitly contrastedwith the other subclasses that could also be subsumedcoordinately with it. Take, for example, the simple divisions inDiagram 3.2. The classes subsumed under ‘male’, i.e. ‘breedingbull’ and ‘beef steer’, can be contrasted with those subsumedunder ‘female’, i.e. ‘dairy cow’ and ‘beef heifer’.The diagram represents a classification. The classes named in thediagram are called categories, which are ordered in various levels,or ranks, from the most general to the most specific. Within eachrank, the coordinate categories are distinguished by reference to aparticular property, which is the basis of division, i.e. in thediagram at the first rank the basis of division is sex, at the secondrank the basis of division is function.The rule that guards against confusion is this: in each rank thecategories are to be distinguished according to only one basis of
division. For example, in ‘black; white; green; red; blue; textured’the last category obviously does not refer to categories of colour.Other serious faults can be found in classifications. Sometimesthe categories in a given rank will fail to be exclusive of eachother, so that the same thing can belong in both of two or moreoverlapping categories, e.g. if the third-rank categories in Diagram3.2 had been, say, Angus, Holstein, Hereford, Jersey (which couldbe either male or female). Sometimes a given rank will fail to beexhaustive, so that we find that we have no category for some ofthe things we want to classify, for example, what happens if one ofthe females we want to classify using Diagram 3.2 is neither adairy cow nor a beef heifer?Diagram 3.2 Classification treeExercise 3.8Test your skill in recognizing which of the following class divisions: (a) aretechnically correct; (b) use more than one basis of division; (c) produceoverlapping categories; (d) leave gaps1. Books: paperback books and hardback books, ebooks.2. Bread rolls: soft bread rolls and hard bread rolls.3. Pies: one-crust pies, two-crust pies, mud pies.4. School buildings: primary school buildings and secondary schoolbuildings.5. Prime ministers: distinguished prime ministers, mundane primeministers, insignificant prime ministers.6. Days: sunny days, rainy days, foggy days, cold days, snowy days.7. Pens: fountain pens, ball-point pens, felt-tip pens.8. Sources of energy: atomic fusion or fission, fossil fuels, water.9. Planets: those that orbit between earth and sun and those that orbitbeyond the earth.
10. Cars: diesel cars, petrol cars, four-door cars, electric cars, two-doorcars.There are many ways of classifying objects and events. Depending onthe situation, one way may be more useful or revealing than another.As our knowledge of a class increases we may repeatedly revise ourclassification scheme in order to add new categories or to reject somebases of division in favour of others. It is well worth asking oneselfbefore setting up a classification: what purpose will the classificationserve, and what relevant new information will the classificationprovide?Sometimes it is very useful to compare two things of widelydifferent categories. We call this activity drawing an analogy.This type of comparison can enable us to see an object or an ideain a new light, and hint at new hypotheses or generalizations thatwe otherwise might not have contemplated. Analogies are widelyused in inductive argument, and this is how they work.AnalogyWhen we compare objects of two categories, we can sometimesobserve that objects of one category are similar in certainrespects to objects of the other category. If we know that objectsof the first category have certain other characteristics, but do notknow whether these are present in the second category, we couldconclude by analogy that, since objects of the two categories arealike in some respects, they are alike in other respects as well.Therefore, objects of the second category also have the additionalcharacteristics observed in the first category.The carrying out of medical experiments on animals is a commonmanifestation of this argument, where it is claimed that similaritiesbetween animals and humans will allow the effects of drugs onhumans to be predicted by their effect on animals. Researchersdo not claim that animals are identical to humans (they could be
put into the same class if they were), but hold that the differences,for the purposes of the tests, are not significant.Like other kinds of inductive arguments, analogical argumentsmay be strong or weak. The strength of an analogy dependsprincipally upon the similarities between the two types of objectsbeing compared. Any two kinds of objects can be similar in manyways and dissimilar in many others. Salmon pointed out that:The crucial question for analogical arguments is this: are theobjects which are being compared similar in ways that arerelevant to the argument … The more relevant similarities thatare present, the stronger the argument, and vice versa. (1984,p. 108)It is not always easy, in a particular situation, to determineprecisely what should be regarded as relevant or irrelevant forsimilarities or differences between objects of the classes. Thequestion that must be posed is: what kinds of similarities ordifferences are likely to have a significant effect on the outcome ofthe phenomenon being investigated? For example, when testingdrugs on rabbits, the fact that humans do not have long ears isunlikely to be a relevant difference. The fact that both aremammals will be a relevant similarity.Analogical arguments appear not only in scientific discourse. Theyare commonly used in philosophical literature (e.g. in Plato’sRepublic, where he compares the state and the individual). It mustbe remembered that in dealing with an argument from analogy, aswith other forms of inductive argument, there is no way of provingthat the argument is correct in all instances, or even exactly towhat extent it is reliable.Figure 3.8 Analogical arguments may be strong or weak
ConclusionsAny discussion of the issues that surround the nature of scientificenquiry and the acquisition of knowledge must necessarily be verybrief to fit the scale of this introductory book. Acknowledging thefact that research can concern itself with a huge range of subjectsand disciplines, I have attempted to cover the relevant majorperspectives which, I hope, will raise questions about your ownresearch approach, whatever your type of research problem.The debate about knowledge and how we acquire it is acontinuous one, and one that is at the centre of any researcheffort. I believe it is important, throughout your research project,and probably in your subsequent career too, to gain a deepeningawareness of what this involves and how such factors can affectnot only our attitude to life, but also our everyday actions.
As discourse is at the heart of intellectual life, a sound techniquein the use of language and argument will be a valuable asset toenable you to put your points across strongly and effectively. Youwill also be in a good position to analyse the work of others, to testthe validity of their arguments, and to avoid the pitfalls presentedby fallacies.The next steps: which type ofresearch for your topic?The aims of this section are for you:to examine the important concepts which are used in yoursubject, and the appropriate dimensions and indicatorsto investigate which relevant existing theories, hypotheses orarguments are related to your problem fieldto determine the appropriate type of research approach foryour topicto explore the type of appropriate research activities for yourdata collection and processing.Checklist of activities that will progress yourresearchStep 1: List conceptsConcepts are the foundations of thinking and communication. List anddefine the main concepts in your area of study. The most important ofthese should appear in your thesis title, though you might list severalmore.Step 2: Make statementsYou can devise statements using these concepts to describe the focus ofyour study.Step 3: Operationalize
Abstract concepts are useful to express complex phenomena and ideas.However, they are difficult to study without defining operational definitionsfor them. Do this for the main concepts that you will be investigating inorder to give you an indication of what sort of data you will need to collectand analyse. This will give you an indication of the likely scale of yourproject. At this stage you can reduce the scope of the research if it lookslike becoming too much to manage, or goes outside your fields ofexpertise.Step 4: QuantifyThe possible levels of measurement for the different concepts willdetermine the appropriate types of analysis. Check these for your mainconcepts, keeping in mind their operational definitions. This will indicatewhat analytical methods might be suitable.Step 5: TheorizeYour background reading should have given you a good indication of thetheory surrounding your chosen topic. Briefly describe the most importantlaws, theories and hypotheses relevant to your subject area, and theirrelationship to your research problem.There might be arguments about what theory is relevant or most helpful infinding solutions to the problems you have identified. Trace the theoreticalstrands in what you have read, investigate any controversies, and askyourself where you stand in the debate. This will be an essential aspect ofthe discussion of the background to your study when you write yourresearch proposal.Step 6: CategorizeLook at your area of study and, using what you know about the variousresearch approaches, select those most appropriate. Be prepared todiscuss why.It might be obvious which category of research your work will fall into, butit is also likely that sections of your research will be of different types. Thisis no bad thing, as looking at a problem from different perspectivesprovides a more holistic view. Consider, in the light of your investigationsin response to the above points, what types of research you will be doing.Step 7: Check the scope
The types of research will indicate the appropriate research methodologyto use. You can now trace out the sort of research activities you will haveto undertake. This will help to answer the question, ‘What will I actually doin my project?’ Try not to become too detailed at this stage, but get a feelfor the likely scale of the work needed (i.e. don’t devise the questions to aquestionnaire, but do consider which and how many people you mightneed to question, what sort of topics you need to cover, how you mightanalyse the results). Again, this exercise will give you an indication of thelikely scale of the work so you can check as to its practicability.Step 8: RedraftYou should now be in a position to explain the main concepts that lie at thefoundation of your research, where they came from in your backgroundreading, how they form your research problem and how they will beinvestigated. You will be able to define the type(s) of research that you areproposing, the general nature of your methodology and the expectedscope of the project.Draw up another draft (or the first one if you have not got that far yet) ofyour proposal. Try to incorporate all the results of your investigations so farin a logical sequence. You should explain the main issues and theoriessurrounding your subject and show how your research project emergesfrom these. The best way is to demonstrate that there is a gap inknowledge that you can help to fill; or that there is a theory or belief thatyou disagree with and wish to contest; or that there is a debate in whichyou want to participate, to add something new or to find a new synthesis.Once you have explained this, you can then describe the main conceptsthat are important in your investigations, the type of research theseinvestigations will fall into, and the sort of research methods that will beappropriate.Consolidation and assessmentWhen you are prepared, arrange a tutorial with your tutor orsupervisor. Explain the main concepts that underpin your area ofstudy and how these are used to describe the focus of your study.Demonstrate an understanding of the relevant theories andconcepts etc. and how they relate to the subject and problemwhich you have chosen.Discuss how the main concepts can be investigated through theiroperational definitions. Give some indication of the dimensions
and indicators that could be used to examine and measure theconcepts. Explain how this will affect the nature of your researchwork. This is to do with practical matters, so try to bring thediscussion down to what you will actually be doing to researchthese concepts. Do this by explaining the relevance of yourproposed research approach(es) and the type of researchactivities that are associated with them.Although you are likely to be still at a very exploratory stage inyour research, this exercise should encourage you to consciouslyexamine the sorts of issues and terminology that feature in yourchosen field of study. Knowledge about these is often taken forgranted by specialists in the field, because they are so much apart of their everyday thinking. You should therefore acquire agood understanding of the concepts and theories etc. that featureas a main part of your subject.Knowledge about different types of research will put your ownresearch efforts into context, within the range of options open toyou. You will appreciate that the type(s) of research you undertakewill have an important influence on the research activities that youwill need to pursue.Further readingThe foundation stones of research, i.e. concepts and theories, arediscussed in much more detail in specialized books on thephilosophy of research. You can quickly get into deep water on thesubject of thinking and argument. I would recommend Brink-Budgen or Bonnett to start with, and perhaps follow up thereferences in there if you want to find out more on specific issues.The others I have listed require you to have either a specialinterest or to have chosen a topic that focuses on aspects of thesesubjects.Brink-Budgen, R. (2010) Critical Thinking for Students:Learn the Skills of Critical Assessment and EffectiveArgument, 4th edn. Oxford: How To Books.
Bonnett, A. (2001) How to Argue. Harlow: PearsonEducation.If you really want to get into more depth about logic and argumentthese are reasonably approachable books, listed in order ofeasiest first:Hodges, W. (2001) Logic: An Introduction to ElementaryLogic, 2nd edn. London: Penguin.Salmon, M.H. (2012) Introduction to Logic and CriticalThinking, 6th edn. Boston: Centage Learning.Gensler, H.J. (2010) The A to Z of Logic. Lanham, MD:Scarecrow Press.Fisher, A. (2004) The Logic of Real Arguments, 2nd edn.Cambridge: Cambridge University Press.Pearson, R.A. (1993) Academic Vocabulary and Argument:An Introductory Guide, 2nd edn. Sheffield: PAVIC,Sheffield Hallam University.Kuhn, D. (1991) The Skills of Argument. Cambridge:Cambridge University Press.These are two amusing books about fallacy that might interestyou:Pirie, M. (2015) How to Win Every Argument: The Use andAbuse of Logic. 2nd edn. London: The Continuum.Well written and entertaining.Thouless, R.H. (1974) Straight and Crooked Thinking, rev.edn. London: Pan Books.
Old, but still entertaining and thought-provoking.For a classic book that concentrates more on concepts andtheory, I can recommend the following:Reynolds, P.D. (2016) A Primer in Theory Construction.Basingstoke: Routledge.See Chapters 3 and 5 initially. The rest makes for stimulatingreading too.
4 Finding, Organizing and RetrievingInformationAimsTo explain where you can find the necessary information toinform your projectTo show you how to get the best out of the Internet and yourlibraryTo help you develop the skills of readingTo present efficient techniques for recording, storing andretrieving your own collected informationIntroductionOne of the most important tasks that you have to undertake in theearly stages of research is a review of the literature relating toyour research problem. Careful consideration at this stage iscrucial, as the way you organize your note-taking and informationstorage will have repercussions in terms of both the managementof your time and the quality of your results and written reports inthe subsequent stages of the research process.According to Wiersma and Jurs (2008), the process should focuson the three questions posed in Checklist 4.1. Notice thatanswering these questions will require you to use methods forsearching, analysis and drawing conclusions, as well as perhapsdissemination.Checklist 4.1 Questions on reviewing the literatureWhere is the literature to be found?What should be done with the information?What use is to be made of the information?
Finding the literatureThis involves searching for publications and other informationrelevant to the research problem. This should start at thebeginning of the research process and continue throughout theproject. Reviewing the literature includes some specific activitiesthat take place in a sequence. Because of the amount ofinformation available, you should follow a systematic process. Inaddition, it is advisable that, at each stage of the review, you do ascomplete and as accurate a job as possible. This will minimize theneed to go back to the source of the information for a misseddetail, which will save you later time and effort. Students, in thelast stages of writing their thesis, sometimes spend hours ofvaluable time frantically trying to retrace the details (e.g. pagenumbers, author’s name or journal title) of an important referencethat they had collected a year or two previously.Reviewing the literature is essential, not only in providing acontext for the research subject and specifically limiting andidentifying the research problem, but also in providing you withimportant information for subsequent parts of the researchinvestigation, such as alternative theoretical standpoints andsuitable research methods.Sources of informationThe Internet is now the first choice of everyone who wants toknow something about anything. It enables a connection to thewidest range of information and literature if you know how tooptimize its features to your best advantage. On its own you candownload plenty of useful material on any subject, for which youmay have to pay; but in combination with your library that hassubscribed to numerous academic journals, you can have full andfree access to the latest, most authoritative literature on yoursubject of choice.However, your university, college or specialist professional libraryis likely to be the most valuable starting point for seekinginformation. Providing helpful advice to its users, it provides
access to a wide range of material, including books containingbackground information, reports of research studies, periodicals,technical reports and academic theses, available either in thelibrary itself or from other libraries.The InternetWith thousands of pages being added every day, the World WideWeb (WWW) is the biggest single source of information in theworld. However, the content is of extremely variable quality, andthe biggest challenge when using it is to track down good-qualitymaterial. You can easily waste hours trawling through rubbish insearch of the goodies. Careful use of search terms helps toeliminate the trash. Usually, the more precise your searchparameters, the more manageable the search results will be. Notall information on the Web is free.The strengths of the medium are its international content, itsusually good currency and its sheer mass of data. The mass ofinformation can result in many false leads, causing time wastingand frustration, and the authority of the information is oftenquestionable and should be checked with other sources ifaccuracy is vital. The problems of tracking down relevant andreliable information are partially solved by the use of good searchengines, the latest one designed particularly for academicsearches being Google Scholar. Access is convenient from acomputer or other Internet access device, and data can be easilysaved and copied (depending on copyright restrictions and theissue of plagiarism). Some information is charged for, but chargesare defined before access is gained and therefore costs arecontrollable.More data are continuously being posted on the Internet(according to one estimate there are currently over 4.66 billionpages on the World Wide Web, or if you include also the non-indexed and Deep Web pages the number goes up to 180quadrillion pages!). This includes research information that is notavailable elsewhere, e.g. in Web pages linked to researchprojects. You should consult your academic library for informationabout free access to information services which make easy
access available to a wide range of databases of journal articles,abstracts, conference literature, etc.The most reliable and comprehensive source of academic papersand other publications is Google Scholar. This searches onlyacademic literature in response to the key words that you select. Ifyour university or organization library has an Athens account, agood move is to log into it before you make your searches onGoogle Scholar. This will provide you with access to free full textdownload facilities for each item if available through Athens.Alternatively, your library might be signed up to Shibboleth, inwhich case you need to go to ‘library links’ in ‘settings’ on theGoogle Scholar search page, and type in your university in thebox provided before starting your searches.Evaluating Web sourcesAnyone can add pages to the World Wide Web, so how can youjudge if the information you have found is reliable? Checklist 4.2gives seven different tests you can make to judge the quality ofthe contents.Checklist 4.2 Tests to evaluate websites1. Is the website accurate? Does it say what sources the data arebased on? Compare the data with other sources. If they divergegreatly, is there some explanation for this?2. What authority is it based on? Find out who authored the pages– are they recognized experts or part of a reputableorganization? Check if other publications are cited or if theyprovide a bibliography of other articles, reports or books. Youmay need to track down the ‘home page’ to get to the details.Also see if there is a postal address and phone number toindicate the ‘reality’ of the organization. Web addresses that endin ‘ac’ (meaning academic) are likely to be university or collegeaddresses and therefore point to some intellectual credibility –no guarantee of quality but nevertheless a useful indicator.3. Is the information biased? Many pressure groups andcommercial organizations use the Web to promote their ideasand products, and present information in a one-sided way. Canyou detect a vested interest in the subject on the part of theauthor? Find out more about the authors, e.g. does the
information about animal experiments come from anantivisection organization, a cosmetics company, or anindependent research institute?4. How detailed is the information? Is the information so generalthat it is of little use, or so detailed and specialized that it isdifficult to understand? Investigate whether it is only fragmentaryand misses out important issues in the subject, and whether theevidence is backed up by relevant data. There may be usefullinks to further information, other websites or printedpublications.5. Is it out of date? Pages stay on the Web until they are removed.Some have obviously been forgotten and are hopelessly out ofdate. Try to find a date of posting or updating (perhaps on theView/Page Info option on your Web browser). Note that someupdates might not update all the contents. Check any linksprovided to see if they work.6. Have you cross-checked? Compare the contents with othersources of information such as books, articles, official statisticsand other websites. Does the information tally with or contradictthese? If the latter, can you see why?7. For dynamically created pages, real-time information anddatabases on the Deep or Invisible Web you need to accessspecialized and limited-area search engines. The Deep Web iscurrently estimated to be over 100 times larger than the surfaceWeb and, according to experts, over 95% of this is available tothe general public. The following site provides a long list of thesegateway sites for the Deep Web and lots of useful information onhow to access them and warnings about the dangers of movinginto this territory:https://www.deepweb-sites.comBibliographyHave you tried pre-evaluated ‘subject gateways’? The informationon these sites has been vetted by experts in the relevant subjects,so can be relied upon to be of high quality. Here are some youcould try:Aerade: Cranfield Universityhttp://aerade.cranfield.ac.ukDMOZ: The Open Directory Projecthttp://dmoz-odp.org
ELDIS: Sharing the best in global developmentresearch for policy and practicewww.eldis.orgSciCentral: Gateway to the best science newssourceswww.scicentral.comScout Report Archiveshttps://scout.wisc.edu/archivesTechXtrahttps://www.researchinformation.info/product/techxtraVifamath: The Virtual Library of Mathematicshttp://vifamath.deWorldWideScience.org: One-stop searching ofworldwide science sourceshttp://worldwidescience.orgWWW Virtual Libraryhttp://vlib.orgLibrariesThere are different, but similar, systems of cataloguing thecontents of libraries in different institutions. Therefore an essentialearly task is to find and understand the system your libraryutilizes, and where materials in the library are located. This shouldbe done as soon as possible to allow you to find your way throughthe shelves and computer terminals of the library to obtain thematerials required. Librarians are mobilized at the beginning ofeach academic year to help you and other new students tounderstand the way the library works and how to look for requiredmaterials. If you are a new student you are strongly advised toattend not only a library tour, usually arranged by the library staffat the beginning of term, but also information sources seminars,which are often organized specifically for research students.There are essentially two main types of information source youneed to be aware of, as shown in Checklist 4.3.
Checklist 4.3 Main types of information source1. Bibliographic databases. These are the indexes, catalogues andother bibliographic tools that enable you to trace the secondarysources you require.2. Secondary sources. These consist of the ‘literature’ of yoursubject, i.e. the books, journal articles, reports and papers youneed to find for the information you require for your research.Your selection of these will depend on your subject area.This next section concentrates on bibliographic aids and indexes,and describes the various types you are likely to find useful.Library catalogues are usually restricted to the bibliographicaldetails of items in the library’s collection, and their numbering onthe library’s shelves. Most library catalogues are now offereddigitally, and are often referred to as online public accesscatalogues (OPACs). There is usually access to documentreferences by author, title and subject, and most libraries operateon a similar basis. However, there may be significant differences.It is important to learn the best way to operate the catalogue, andlibrary staff will be happy to show you. It is also possible to searchthe OPACs of other libraries worldwide through the Internet. Youmay want to do this if there is a library specializing in your subject,e.g. German literature, or if there is a major copyright library whichreceives copies of all new books published in a particular country,e.g. the British Library. Inter-library loans allow you to get copiesat a reasonable cost of any publications that are not available inyour local library.Library cataloguesBibliographic databases are online and can be accessedthrough your library, often even from remote locations. There are alarge number of databases that specialize in specific areas ofknowledge, e.g. science, medicine, humanities. Once you havelocated a suitable database, by alphabetical list or subject search,then you can search for the titles of literature on the desired topicby key word searching. Your next job is to locate the specific
article, book or other publication either in the library collection oronline.Bibliographic databasesIt is worth noting that bibliographic aids and indexes, even thosedevoted to a specific subject, are rarely anything likecomprehensive. For example, the Architectural Periodicals Indexcontains references to articles from many, but not all, architecturaljournals. It is limited to those purchased by its producer, the libraryof the Royal Institute of British Architects. This means you willneed to do an additional organized literature search through thesources most likely to cover your research subject. See whetherthere are any courses of instruction, organized by your subjectlibrarian, to offer guidance on these.Directories of organizations, although not bibliographical sources,are nevertheless useful in tracing details of organizations inspecific fields, and of key personnel in those fields.E-books are becoming increasingly available and many are free todownload. You will be able to access them electronically throughyour library. They are presented as collections, such as historicalor business, as well as more general subjects, and there areongoing projects by Google and Internet Archive to scan millionsof books from great libraries in the US and UK.E-journals are also accessible through your library electronicsources. The journals are listed alphabetically or as collections. Tofind a paper, you will have to know what you are looking for, i.e.the title of the paper, date and journal details such as issue andvolume number. For less precise searches, you can select asubject area for relevant journals to be listed.E-newspapers from around the world are easily accessed, andarticles can be found by key word searches, or papers ofparticular dates inspected. Again, your library will be able to giveyou access to these.
Dictionaries and encyclopaedias can be accessed electronically –an e-reference desk – for reference information as well as maps,statistics, museums, thesauri, translators, local information,government, societies and much more!Once you have obtained copies of books or papers that you finduseful for your research, don’t ignore the lists of references at theend of the publications. These may lead you to more relevantliterature.Finally, people, although they are non-bibliographical sources, canfrequently advise on where to look! Librarians are people – usethem! In particular, make sure that your subject librarian knowsabout what topic you are researching.Exercise 4.1This is a practical task, not a written exercise. Go to your university or anyother library and familiarize yourself with the sources of informationmentioned above. In a university or college library, inform yourself aboutthe library tours arranged by the library staff for new students. Find outwhere the sources of information are and how you can use them. Try outboth written and computer-based indexes, for example, by searchingunder a key word. You can compare the type of information that eachsource produces. This exercise will take some time, so do not interruptyour study of this chapter of the book for too long while you complete yourexploration of all the library facilities. You will find that informationtechnology is a fast-developing field, so you should keep up to date withthe latest facilities offered.Do not overlook the existence of specialist libraries. Manyprofessional bodies, associations and other organizations possesstheir own collection of specialist literature. A visit to the librariesappropriate to your subject may be essential, as the informationcontained therein may be unobtainable elsewhere. You can alsogain access to other non-specialist but certainly ‘special’ libraries,such as the Bodleian in Oxford and the British Library in London,as well as the national libraries in other countries, e.g. theBibliothèque Nationale de France or the Library of Congress inWashington DC, where valuable historical publications andmanuscripts can be found.
Private collections and historical records, which are locatedthroughout the country, may be of interest to you, particularly ifyou are doing historical research.Search techniques for online catalogues,databases and the InternetChecklist 4.4 provides basic hints on how to make effectivesearches for information. In order to find what you want, you mustfirst be clear what you are looking for.Checklist 4.4 How to make effective searches for informationSearches rely on single words or a combination of severalwords. Every subject will contain several crucial terms thatdistinguish it from other subjects, so the trick is to select these. Ifyou are unfamiliar with the subject, look it up in a specialistdictionary or an encyclopaedia to see what terms are used sothat you can build up your own list of key words. Remember alsothe use of different words in different parts of the world, e.g.car/automobile, lift/elevator, pavement/sidewalk, and alternativespellings, e.g. organisation/organization.Databases usually provide the option of a free text search, or akey word, subject or index search. The former looks for yourchosen word(s) in all the text contained in the database, whilstthe latter only searches the key words, subject lists or index listsprovided by the authors with their books or papers. These listsfocus on the subject matter of the publication, and so give amore reliable guide to the contents. Many databases include athesaurus – a list of indexing terms that will help you to use thestandard terms. Sometimes the number of articles to which theterm is assigned is given.It is usually possible to narrow your search by indicating placeand time, where these are relevant. The publication date is abasic piece of bibliographic data.Adding a * to words or parts of words automatically widens thesearch parameters in the form of wildcards. For example,‘automa*’ will find all the words starting with that stem, e.g.automation, automatic, automaton. Inserting the * symbol into aword takes care of different spelling versions e.g. ‘labo*r’ will findlabour and labor.Boolean logic is a fancy word for the technique of usingconnecting words such as ‘and’, ‘or’ and ‘not’. These refine the
search by defining more closely what you want to include or not.For example:Schools and finance: this narrows down your search byonly selecting records that contain both terms.(Lifts or elevators): this widens your search by selectingrecords that contain either or both terms. Note that ‘or’statements must be in brackets.Nurseries and playschools not schools: this eliminates theterms that you do not want to consider. However, be carefulthat you do not eliminate useful records.It is best if you keep the search terms simple and search severaltimes using different variations and terms.The commonest way of searching for information on the Web isto use a search engine such as Google, Google Scholar, Lycos,Bing, Ecosia, DuckDuckGo.You should carefully read the instructions on the search methodsrecommended. The commonest way is to use concepts in theform of key words, together with the use of Boolean logic.See this ‘cheat list’ for more help on formulating search terms:www.googleguide.com/print/adv_op_ref.pdfExercise 4.2This is another practical task to practise your use of the Internet.1. Open up Google Scholar, then using Boolean logic, set up a searchof concepts related to your research topic. Note the number ofpublications selected, then try to narrow down the choice if thenumber is unmanageably large, or widen it if there are noneselected, or only very few. Check through the results to see if there isanything potentially useful.2. Try the same process with other search engines for information onthe general internet. Notice that you have to be much more selectiveto discount the frivolous, out-of-date, opinionated, unattributed sitesin order to find those that can be trusted and have verfied qualities.The best way to learn how to get the most out of the Internet is to useit. Trying things out yourself is the only way to explore what potentialthis medium has to help you in your project. It is useful, however, to
consult a guide on doing internet research that provides moreinformation than I have been able to include here about the featuresof the Web and the tools for exploring it. You can get plenty of onlineadvice by Googling ‘Guides to internet research’.Dealing with informationAfter you locate the references and the data in your literaturesearch, you might wonder what to do with them. In fact, if thereare many references, you might get worried and say: ‘How am Igoing to read all these reports?’ The first step that you should takeis to determine whether the data included in a document, book orother source are appropriate to your research problem. If not, thesource should be discarded. If you decide that the source containsinformation relevant to your research problem, then you shouldfollow a systematic and consistent approach to dealing with thisinformation, in order to make it easy to retrieve when you requireit. Although several approaches for storing and retrievinginformation are suggested, you will have to decide on a systemthat suits you best, or even develop your own methods. This mustwork well for you, but it must also conform to the basic guidelinesof good practice. How you do this is dealt with in detail in Chapter5.Activities relating to dealing with your information include readingand note-taking, discussed below.ReadingSince you are likely to be dealing with masses of written sources,you have to acquire the appropriate technique of reading theircontents in a limited time. That means, as Newman (1989, p. 12)puts it, acquiring the technique of learning how to ‘tear the heartout of books’ or how to ‘gut’ them efficiently and quickly. This willenable you to work your way through the large masses ofinformation.According to Fairbairn and Winch (2011), reading involves (1)finding meaning in written texts, (2) literal, inferential and
evaluative comprehension of texts, and (3) reading in differentways. In addition, Krantz and Kimmelman (1992) state that yourgoals should be those listed in Checklist 4.5. These indicate verywell that this type of reading is not meant to be a passive activity –you must actively engage in not only the subject matter, but alsothe way it is presented.Checklist 4.5 Goals of reading1. To review the text.2. To use context clues and a dictionary to understand new words.3. To identify and mark important ideas in a chapter.4. To recognize how authors organize and develop ideas.5. To identify new words and phrases that describe the methods orpatterns of organizing and developing ideas.6. To apply comprehension skills to vocabulary and text material.Reading text quickly is not easy if you do not have the appropriatetechnique. Reading systematically to fully understand a textbookdemands a lot of effort and may be time-consuming and not reallynecessary. You can possibly find shortcuts to locate the part orpassage of the text that interests you and this will enable you todevelop ways of using your reading time more effectively. Thereare, in fact, several available techniques contained in manypublications on increasing reading and comprehension speed(Fairbairn and Winch, 2011; Freeman and Meed, 1993; Krantzand Kimmelman, 1992; Cottrell, 2019; Smith and Smith, 1990).Figure 4.1 Reading text quickly is not easy if you do not have theappropriate technique
Each of the above-mentioned authors suggests techniques ofreading, but do not forget that you might develop your owntechnique of reading that might work better for you in terms oftime, effort and comprehension. Freeman and Meed (1993, pp.31–41) suggest the techniques in Checklist 4.6, which indicate thedifferent levels of engagement that you could apply to the text.The order of this list suggests that your reading will normallyprogress from a quick overview to test the relevance of thedocument, to progressively deeper involvement as you home inon the really interesting parts relevant to your research. Leavenumber five for your spare time!Checklist 4.6 Reading techniques1. Skimming. This involves looking quickly through the book andreading only things like contents, headings, introductions andconclusions. This is a quick and efficient way of familiarizingyourself with a publication and is useful if you wish to check
whether a written report is relevant, or you wish to find particularinformation or ideas quickly.2. Scanning. This is a very rapid search for some important point. Itmay be a page number, a title or a key word. The essential thingis that you deliberately ignore everything except the one item forwhich you are scanning. You use scanning when you look up anumber in the telephone directory.3. Reading to understand. This involves detailed study of a chapter,passage or article in order to absorb all the major facts andideas. You may read it more than once, and take notes tosummarize what you have read.4. Word-by-word reading. Very occasionally, you actually need toread every word extremely carefully; for example, when readingan exam question or following a set of instructions.5. Reading for pleasure. This is the reading you do to relax andenjoy, as with a novel.Smith and Smith (1990, pp. 54–89), in an alternative approach,suggest some techniques for reading skills which include: (1)efficient scanning with the use of key words and key terms, (2)use of titles to anticipate the contents of a reading passage, and(3) finding out the content of a text by looking at the first sentenceof each paragraph.With respect to an adequate reading speed, Newman (1989, pp.12–14) argues that reading techniques fall into a basic patterninvolving four stages, as shown in Checklist 4.7.Checklist 4.7 Stages in readingStage 1: Try to gain some quick impression of what the book isabout; what question or questions the author is trying to answer;how the book is structured; and whether, in fact, the questionstackled and the answers put forward are relevant to your needs.You can do this by glancing over the cover or jacket, the preface(if any), the list of contents, and the index. Try then to gain anoverall impression of the book and its structure.Stage 2: If you decide that the book is relevant to your researchsubject, then you must formulate the question or questions thatyou anticipate will be answered in the book. This enables you tolocate the required information and will save you time and effort,as you cannot afford to go on reading aimlessly through the book.
In addition, at this stage you must adopt an active and analyticalattitude.Stage 3: After formulating the main question or questions that youanticipate the book will answer, you must review the book to lookfor answers to your questions. This involves locating the parts ofthe book where your questions are dealt with. You must then lookfor the answers or conclusions that the author has drawn, andalso at how the author arrived at them. You will also look atarguments and evidence put forward to support the viewsexpressed and you will make an attempt to assess the validity ofthe evidence and the structure of the argument which utilizes suchevidence. There are, however, cases where conclusions areunsupported, arguments or evidence are non-existent, orsometimes there is no conclusion at all.Stage 4: Supposing that you have extracted the relevantinformation from the written report, you must now record your datain note form, so that later you can retrieve it and use it easily atthe appropriate stage. The different ways of doing this areexplained later in this chapter.Exercise 4.3Put your reading skills to the test by reading an article in different ways.Select, from the information you have collected about your subject, anyarticle or research paper that you have not yet read but that you thinkmight contain useful information. Now read it in the following ways:1. First skim the text and write down very briefly the main points of theinformation offered. Remember to first check the contents orabstract, the main headings, and perhaps part of the last paragraphabout conclusions. Do not write down more than twenty words, butmake sure that they are quite precise and organized in a coherentway so that you will easily understand your notes at a later date.2. Next select two main concepts that you are interested in and thatseem to be featured prominently in the article (perhaps they arementioned in the title). Scan the text and note what is written aboutthese two concepts.3. Now take just one section headed by a subtitle and read it carefully,making detailed notes on all the major issues and facts. You shouldhave about ten short notes to cover the information properly.4. Finally, read the last paragraph(s) very carefully indeed. What are themain conclusions and why are they important, or perhaps they arenot? What effect do they have on your thinking about your subject?
You could express this as personal notes to remind you of thesignificance of this article in some aspect of your work.5. The other type of reading – reading for pleasure – you can practisewithout needing it to be the subject of an exercise!It is good to be aware of how you are going to read a text before youbegin. Your reading method should be decided by the type and detailof information which you require from the text. A conscious decisionbefore reading or scanning can save you much time by avoidinginappropriate techniques, resulting in too much or too little informationbeing extracted.Note-takingYou should take notes only of the information that you judgerelevant to your research subject. It is generally agreed that,except for quotations, when you take notes you should use yourown active vocabulary and not the author’s. Newman (1989, p.15) says that it is important to use your active vocabulary in note-taking, because it helps to ensure that you have a reasonably fullgrasp of the meaning of the information concerned; it might assistin laying down a better memory trace in the brain; and finally,when you come to re-read and synthesize the notes, yourpersonal active vocabulary is more easily comprehensible andrecognizable.Reasons for taking notesYou might sometimes wonder why you take notes. Freeman andMeed (1993, p. 43) lay down a number of reasons why. Youshould remember that note-taking is not restricted to makingrecords from written reports, but may also be applied to lectures,conferences, interviews and so on. Reasons for taking notesinclude those in Checklist 4.8.
Checklist 4.8 Reasons for taking notesTo help you remember something. You can’t hope to retain awhole lecture, book or discussion permanently in your memory,so instead you make notes of the most important items and usethe notes for revision and reference.To keep a permanent record of something. If you attend a lectureor visit somewhere as part of your course, your notes may beyour own record of what took place.To help in your planning. Notes can be a good way of starting offa project or a piece of writing; you can note down the mainthings you need to do, the books you need to read, and so on.To reorder material. Making notes, which can be reshuffled,provides one of the most useful opportunities for rearrangingmaterial in whichever form is most convenient to you.To help you understand what you are learning. Writing thingsdown yourself forces you to think them through properly and isone of the best ways of remembering them.To help you to concentrate. If you are listening to someonetalking, your mind may easily wander; making notes helps tokeep you active and involved.To show other people. You may want other learners to benefitfrom the notes you have made.When making notes, you will have to make judgements all thetime about what information will, or might, be of use to you. Thesejudgements will depend on your own knowledge of your subjectand the nature of your research problem. During yourinvestigations, suggests van Dalen (1979, p. 92), you may want tocopy many specific facts from references, such as dates, places,names, statistics, formulas and definitions; or summarizearguments, questions, explanations, illustrations or descriptions.In addition you may find it useful to write comments about yourreactions to the reference materials and state relationships,conclusions or interpretations that come to mind during thecontemplative phases of your work. You may also want to jotdown items that require further checking.When writing a report you may want to draw upon your notes for avariety of reasons, for example, to support a particular position orto illustrate a point of view. You may want to make comparisons,
weave a web of logical evidence, or buttress arguments bypassages from recognized authorities. A body of notes, collectedwith discrimination, could provide the building blocks for all ofthese, assuming that the range of your notes is sufficiently wide.However, a haphazard collection of notes will leave you with animpoverished resource, and may cause your investigation tocollapse (van Dalen, 1979, pp. 92–3).Note-taking techniquesSmith and Smith (1990, pp. 93–103) put forward what they call‘golden rules’ to help you take successful notes, including thoseshown in Checklist 4.9.Although these refer to writing notes on paper, you can apply thesame principles when you use a word processor or a dedicatedreference management software such as EndNote.Checklist 4.9 Golden rules for taking notes1. Clarify your purpose.2. If you are taking paper-based notes, write all your notes on thesame-sized paper or cards. If you are using referencemanagement software, make sure you understand the system ofrecording notes.3. When you begin, set out your notes properly. In making a fullbibliographic record of the source of the notes you are writing,include author (or speaker); date of publishing or of event (e.g.lecture, interview); title of book, article, conference, Web page,lecture, etc.; where published or held; detailed page numbersreferring to the individual points, opinions and data which arenoted; and usefully, the library catalogue number or otherinformation to enable you to locate the book, article, Web page,etc. quickly at a later date.4. Use the title of the chapter or lecture, or suitable key words tohelp you summarize the main ideas of the text.5. Keep your own ideas, comments and criticisms separate fromthose in the text.6. When you finish, sum up what you have written.
Do not forget that you take notes to help in your writing at a laterstage of your research project. To assist this, your notes should bebrief and clear; if they are too long, you will find it tedious to gothrough them, either to search for a specific point, or to refreshyour memory. For the same reason they should be easy to readand understand; if you cannot read through them to refresh yourmemory, they will fail in their purpose. You should organize themto suit the way you learn and your reasons for learning. You willfind that if they are not relevant to you, you will never look at themagain (Freeman and Meed, 1993, p. 44).Wiersma and Jurs (2008) argue that, when taking notes onresearch reports in particular, you should use some form ofabstracting which involves a summary in note form, containing theitems of information shown in Checklist 4.10. This is a usefulchecklist that, when applied on all occasions, will ensure that youdo not forget any aspect of the report and can then readily makecomparisons between various research projects.Checklist 4.10 Information you should extract from research reportsBibliography entry: an accurate and complete bibliographic entryheads the notes.Problem: this is a statement of the research problem of thereport being reviewed; it may include statements of hypotheses.Subjects: the individuals involved in research studies are oftencalled ‘subjects’ of the research.Procedures: this section describes how the research wasconducted. It includes such items as the measurements usedand the analyses performed. This section may also be called‘methodology’.Results and conclusions: this section identifies the relevantresults and conclusions of the study. A distinction may be madebetween results and conclusions – results being whateveroccurred, such as certain statistics, and conclusions being whatthe researcher has made of the results. In long reports withmany results and conclusions, it is best to number them.For further details on the type of abstract or summary recommendedby Wiersma and Jurs (2008), including some samples of abstracts,you may refer to the original.
There are many different ways of taking notes and you have tofind a system that works for you. However, there are someimportant points that you should bear in mind. Never make noteson the text as this would spoil the source (and invoke the rage ofthe librarian if it is a library copy!). Keep your own separatenotebook or your laptop in an accessible place, for you neverknow when you may want to record something; notes on backs ofenvelopes always get lost.When you locate a passage that you think is relevant to yourresearch project, read it first before taking any notes, so that youcan be really selective. The notes, written in your own words, donot have to be in the form of fully articulated sentences, but makethem clear enough for you to understand them when you usethem at a later stage, when you might have forgotten their context.Carefully copy direct quotations from the text when you think youmight wish to use them later, but keep these fairly rare andprecisely referenced.Do be ready to learn from the writing of others on how to improveyour own writing. You may come across examples of excellentstyle in sentence structure, vocabulary or metaphor that can bebeneficial for your own writing (Fairbairn and Winch, 2011).Figure 4.2 There are many different ways of taking notes
Collecting and ordering your notesThere are two basic formats for storing and retrieving notes:computer basedpaper basedThe computer-based formats rely on various database programs.These are set up to deal with lots of bits of information so that youcan easily store and retrieve them. The most common non-specialist database programs come with standard programpackages, such as Microsoft Access. Standard databaseprograms have a limit on the length and format of fields used forstoring your information, so do set up a typical record of the noteson one book to see if you can include all the information you want
and in the way you want it. You can even devise a simple systemusing your normal word processing package.However, it is best to use more specialized programs calledreference management software, which are aimed at exactly thisjob of finding references, recording notes and doing all the citationand referencing. The current main ones are EndNote, RefWorks,Zotero and Mendeley, though there are many more, someavailable for free (seehttp://en.wikipedia.org/wiki/Comparison_of_reference_management_software for a list of different softwares compared). Check ifany are available on your university or college network. Importantissues to watch out for is the database connectivity, which allowsyou to import data directly through the reference managementsoftware, and also with which word processing programs they canintegrate.The basic requirements of the system using a computer formatare similar to those of a paper-based one. Notes should be shortand on a single topic, they should be thoroughly referenced, andthey should be stored under allotted headings. The majoradvantages of a computer-based system are that you have muchmore powerful search facilities; your notes are easily retrieved,copied, revised and edited; and you do not need to rewrite yourreference information (lots of complicated formatting andpunctuation), which is generated automatically according to whatyou have cited in your text. You can also store all your notes on aremovable medium such as a rewritable CD, memory stick orflash drive, or on the cloud. The main disadvantage is that, inorder to avoid copying out, you need to have your own computerwith you wherever you need to make the notes – not such aproblem if you own a laptop, although some of these programshave apps for use with smartphone or tablet. You will also have tospend time learning how to use the program.Further information on which applications are suitable and how touse them can probably be obtained from your university computercentre or library. Do use a specialist bibliographic databaseprogram if it is available to you.
The paper-based format needs no electronic equipment, though afew accessories make life easier. The principle behind this systemis to write your notes on sheets of paper or card, and then orderthe sheets in such a way that you can find the notes when youneed them later. You will need to think of the manner in which youwant to accumulate your notes. Many different formats aresuggested, but you can design your own setup which might bemore suitable for you than another. However, the most importantfeature is that each card, piece of paper or other form of recordmust include only one idea or one fact or one item or onequestion. That is because at this stage you do not yet know inwhich part of the structure of your research project this one pieceof information will fit, or if it will be used at all.You can use A4 sheets of paper and store them in ring folders, oralternatively use index cards (the larger sizes are more useful)and store these in boxes designed for the purpose. The idea is tostore your notes under certain headings so that you can find whatyou have collected on those subjects. The headings can bevarious, depending on your subject and how you will beapproaching it. You will have to work out the best method yourself.The kinds of headings commonly used are:key wordsauthor namespublication titlesdates (particularly useful in historical studies)subjects – or aspects of the main subject.An effective note-taking system preserves the most significantideas in a form that facilitates shifting, comparing, grouping andordering items (van Dalen, 1979, p. 88). Leedy and Ormrod(2015) make useful suggestions on the use of coloured duplicatednote cards and bibliography cards, with separate copies filedunder different headings, producing parallel files of, for example,author, place, date, subject.This way allows you to find the material whether, for example, youare looking for the writings of a particular author, or informationabout a particular subject or a particular date. It is best to keep the
notes short, that is, concentrated on one topic or aspect of thesubject. Start another page or card when the topic of your noteschanges. This is because you may want to search out all yournotes on one topic, so you can pull out all the pages or cardsunder that topic heading.The greatest advantage of this format is that you can make andtake your notes anywhere without needing any equipment apartfrom paper or card and a pen. The main disadvantage is that youwill need to rewrite the material from your handwritten notes onthe word processor when you use them.There are several bits of information that you must record for eachand every one of the notes that you take, irrespective of theformat you use, as shown in Checklist 4.11.Checklist 4.11 Essential information to record on every noteThe author(s) of the text – surname and first names. Perhapsthe name on the book is the editor of the book, who hascompiled a series of chapters or papers by various authors. Inthis case you will also need the name(s) of the author(s) of therelevant text.The title of the book – including a subtitle. If it is a journal ornewspaper, you will have to record the full name of the journal ornewspaper.If it is an article or paper in an edited book or journal ornewspaper, with different authors for different chapters orpapers, then the title of the relevant chapter or paper is alsorequired.If it is a website, the URL (Web address) and the date youaccessed it.The date of publication (in a book, look on the reverse of the titlepage for this).The place it was published (ditto).The name of the publisher (ditto).The page number or numbers where the information you havemade notes from appears.Also useful is a reference to where you found the information,e.g. which library and the book code number, so that you caneasily track it down again.You might also use material from lectures or conferences. In thiscase, give full details including the speaker, title of talk,
conference title, venue and date.This information, attached to every note, will enable you to fullyreference it, and to find the original information again if you needto.Obviously, you can devise your own system that uses the bestfeatures of each format. Whichever format you want to use, youmust decide exactly how you will do it, and test it out beforeputting it into general use.There are several advantages and disadvantages in relying on acomputer system compared with using a paper-based system, asindicated in Box 4.1.Box 4.1 Advantages and disadvantages of using a computer-basednote systemThe advantagesYou can search databases through the system to find usefulliterature and download all the citation and referencing details.You need write the notes only once. You can copy the text whenyou want to use it and edit it without writing it all out again. Thispromotes accuracy as well as saving time. (There are dangers inusing this facility; see the disadvantages below.)You can easily search through your notes for key words, authors,concepts, etc.You can repeat bibliographic information for each note at thetouch of a button.You can link notes to outlines, subjects, authors, dates and othercategories of your choice.You can keep all your notes in highly portable form on removableelectronic media or the cloud.You can easily devise bibliographic lists with the copy facility, sothat you need write the bibliographic details only once. Thismeans that you can be sure that the information is correct,however many times you need to copy the references.The disadvantages
It takes time to learn how to use the database programeffectively.You need to have frequent (perhaps even constant) access to acomputer, although some programs have apps for use onsmartphones and tablets.You are limited to where you can take notes, i.e. to where youcan take or use a computer. Possessing a laptop computer helpsin this respect. Using a smartphone or tablet may also bepossible.You can be tempted to miss out on valuable redrafting stages asyou electronically copy and order your notes for use in essays,papers, etc. There is a danger that you may consider the sets ofideas and quotations you have listed in some sort of order fromyour notes, and printed out, to be a finished piece of writing. Infact, considerable further effort will be necessary to turn what isbasically a list of notes into a clear and concise argument.Having to physically rewrite text always prompts you to thinkanalytically about what you are writing.Finally, do not forget that you should decide which form of note-taking you should adopt and which suits your research projectbest. Notes are a means to an end, not an end in themselves.However they are recorded and stored, the essential thing is thatthey are useful and contain the necessary information, and can betraced when required (Wiersma and Jurs, 2008).Exercise 4.4This exercise will enable you to practise some of the skills involved innote-taking which, of course, also involves an analytical approach to boththe form and the content of what is read. It uses material that you want toreview anyway, so will not take any time away from your research, but willgive you some step-by-step guidance on the procedure of note-taking fromthe literature.Select a short article or paper reporting on some research in your area ofinterest.1. Read through this text quite quickly so that you get an overallunderstanding of what it is about.2. You should now have a good understanding of the contents of thearticle. The next task is to analyse and record what you have read.You should use one of the note-taking techniques (e.g. filing cards,
sheets of paper or computer database) which have been describedearlier in this chapter. Use the following questions to act as a guideto your note-taking, making a brief note in response to eachquestion.1. When and where was the research carried out?2. What were the main aims of this research project?3. How were the main concepts defined?4. What were the main methods of collecting data. Describe themain features of these methods and any interesting pointsabout how difficulties in collecting the required information wereto be overcome.5. What sort of data were gathered and how were they organized?6. What were the main methods used to analyse the data?7. What were the results of analysis?8. What conclusions were drawn in the report?Do not forget to put no more than one subject item on eachcard/sheet/record. Remember to put with each note: the nameof the author(s), the date of the publication(s), the source of theinformation (i.e. title of book, article, etc., which can beabbreviated), and the page number(s) where the informationappears.3. Now you have a comprehensive set of notes on the contents of thearticle. To retrieve the notes at a later date, it is important that eachnote is filed under one, or several, key words or subject headings.Devise a series of key words that point to the main features ofinterest for you in this research. These could be subject, concepts,methods, results, etc. Some headings will have several (or even all)of the note letters listed beneath; others might have only one. Ideally,if you wanted to keep a collection of notes under each key word orheading, you would need to make multiple copies of many of thenotes, or at least work out some system of cross-referencing. Thisduplication can be avoided if you use a database program, whereyou can give key words to each note for easy retrieval by key word.4. Test your retrieval system by picking one key word and seeing howeasily you can retrieve the notes under this word. The processshould be equally easy after your have sorted the notes fromnumerous sources.This might be a rather long exercise, but it should have given you anidea of what is involved in understanding, analysing and recording awritten text. Note-taking is a meticulous process if done correctly.
Remember, if you devise and maintain your own efficient system ofrecording and ordering notes, it will save you hours (possibly days) oftime and frustration later in your studies.Personalized bibliographiesAs well as compiling a collection of notes, you will need to make aseparate list of all the references you have used, and ofreferences that you think may be useful – a personalizedbibliography. The style of the information in your bibliography mustcomply with one of the established systems. The ‘Harvard system’(author, date) is the most commonly used nowadays, particularlyin the sciences and social sciences. If you use a referencemanagement program, the list is automatically produced as youcompile your notes. Details of the style are discussed in Chapter5, but you can easily look at the references in this book for amodel of how to do it. Again, the reference management programwill do this all for you automatically, in any of the standardreference styles.Check that you have the information in Checklist 4.12 for each ofyour references.Checklist 4.12 Information needed for each reference1. The author or authors2. The date of publication3. The title of the book, paper or article, or anything else such aslecture, tape, film4. If it is a published work, the place of publication and thepublisher5. If it is a paper in a journal or an article in a newspaper, the title ofthe journal or newspaper, the volume and issue numbers, thepage numbers of the paper or article6. If it is a chapter in an edited book, the title of the edited book, thename of the editor and the page numbers of the chapterDon’t forget to include any relevant publications, dissertations,reports, etc. that you yourself have written in the past.
It is very easy now to compile huge lists of publications by savingthe results of electronic searches in the library onto electronicmedia. These lists can be convenient for trawling through at yourleisure away from the library, but do be selective and discardinappropriate references. When you compile a bibliography, orbetter said, a list of references for a paper or a dissertation, makesure that only the relevant references are included, i.e. only thepublications referred to in the text. Padding out the list to showhow well-read you are does not impress the reader! Yourreference software will do this job for you at a press of a button,as long as you have followed the procedure required to tag thecitations.ConclusionsHow you take notes and how you organize them matters. Thequality of your research depends on it. Many sections of yourdissertation, thesis or research report will be directly based onanalysis of the information that you have collected over manymonths from a wide variety of sources. The richer your collectionof notes and the better your system of filing and retrieval, themore resources you will have at your disposal for your analyticalreview.It is very difficult to change or improve your system at a later date,so time spent now on devising a system tailored to your needs iswell spent.The next steps: devising your owninformation system and writing yourliterature reviewThe aims of this section are for you:to find relevant literature for your own projectto devise your standardized note-taking, referencing, storageand retrieval system
to check that it works.Checklist of activities that will progress yourresearchStep 1: Use your libraryMake the most of your library. Become adept at using all the differentinformation sourcing techniques available. You will need to enrol ontraining sessions to learn how to do electronic searches and to learn aboutall the different databases and other sources of information. Find out aboutWeb-based information services (e.g. Web of Science, Janet, Library HubDiscover, Gabriel, Zetoc). You should also be able to get a list of subject-related information sources and databases from your subject librarian.Step 2: Get onlineYou will not always need to go to the library for this information. There area lot of Web-based databases and library catalogues that you can accessfrom a computer anywhere. You might need to have a password to accesssome of them: consult your university/college library to obtain these.Step 3: Be alertedYou can sign up for alerts to new information from some databaseorganizations. This saves you constantly making new searches.Step 4: Track forwardTrack forward to further sources by consulting citation indexes. In whichother works has the paper you are reading been referred to?Step 5: Track backAlso check the references and bibliographies of anything you read forrelevant past literature.Step 6: Keep track
The Web is a seductive and sometimes frustrating source of information.Avoid getting sidetracked into irrelevant issues and poor quality sites oflittle academic credibility. It is a good idea to limit the time for each sessionso you don’t get lost. Earmark useful sites to visit in more detail duringfuture sessions.Step 7: ReadBecome conscious of the level of reading you need to use in eachcircumstance. Are you scanning for leads to information or are youactually analysing the text in detail? The type of notes you make will bedifferent in each case.Step 8: Take noteSpend some time analysing your note-taking procedures. Differentiatebetween making a summary and making a commentary: both are valuablebut should not be confused. Also differentiate between opinions, factualinformation, theory, methods, results, interpretations, examples, as well assorting into subjects and concepts.Step 9: Get sortedYour system of recording notes, storing and retrieving them is of vitalimportance to your project. Now is the time to sort it out! Although I havegiven a balanced review of paper-based and computer-based systems, Iwould strongly recommend you using a specialized computer bibliographicdatabase program such as EndNote to store your notes. They providemore flexibility and useful compatibility with Web-based bibliographicdatabases.Using a set of notes you have made from your reading of your subjectliterature, decide on the system you will use to collect, reference, storeand retrieve all the information that you will collect over the next year ortwo (or three or four!).Think carefully of the conditions under which you will be making yournotes or collecting information, e.g. in the university library, at home, in adistant library, at an interview. This will affect the medium in which youmake your initial notes: it can be cards, paper sheets, photocopies, taperecordings, etc. Will you then transcribe all the notes onto one medium,e.g. cards or a computer database, or will you keep them on a range ofmedia? Decide on a standard form of referencing for each snippet ofinformation. Consider how you will cross-reference your notes with your
bibliography list. Under how many headings or key words can youorganize your information?Step 10: Try it outHow you will be able to retrieve your information will depend on how it isreferenced and stored. Try to evolve a system that reduces the amount ofrewriting of text and references. This will reduce the number of errors.Computers can help here.Whatever system you decide on, check, by making a test on a smallsample of your notes, that you can do the following:Trace the exact location of each source from which you made thenotes, including where you found it, library sort code if relevant, fullpublication details, and page number.Easily pick out all your notes on a particular subject, concept orkeyword.Trace all notes on the writings of a particular author.Distinguish between quotations, summaries and commentaries.Step 11: Back up!Make regular copies or backups of your notes in case they get lost ordamaged. This is particularly important with electronic media as they areespecially vulnerable.Consolidation and assessmentWhen you are prepared, you should arrange a tutorial with yourtutor or supervisor. Discuss with him/her your decisions onrelevant information sources and search methods.You should be able to demonstrate to him/her, with actualexamples of your notes, how you propose to manage therecording, referencing, storage and retrieval of your collectedinformation.This exercise should result in you having an appropriatelytargeted method of information collection, and in the creation ofyour own well-considered system of recording, storing andretrieving information. This will be in place from an early stage in
your research programme, in order to enhance your efficiency andresult in a significant saving of time later in your studies.Further readingHere are some short descriptions of what you may find useful inthe following books, starting with searching and organising theliterature:Fink, A. (2014) Conducting Research Literature Reviews,4th edn. Thousand Oaks, CA: Sage.A guide through the multidimensional sea of academic literaturewith techniques on how to deal with the information when youhave found it.On the Oxford Brookes University library catalogue there were2,350 references to information guides! Many of these werebibliographies devoted to a particular subject; you could search onyour university or college catalogue for one relevant to your topic.Here are three examples, the first very general, the next less so,and the third quite narrowly focused:Hurt, C.D. (1998) Information Sources in Science andTechnology, 3rd edn. Englewood Cliffs, NJ: LibrariesUnlimited.Amico, E.B. (ed.) (1998) Reader’s Guide to Women’sStudies. Chicago, IL: Fitzroy Dearborn.United Kingdom Sports Council (1998) Women andSport: From Brighton to Windhoek. Facing the Challenge.London: UK Sports Council.And here are some of the multitude of books to help you navigatethe Internet. The Open Web is becoming central to studentresearch practice, not least because of its accessibility. Theseclear texts describe search strategies and outline the critical skills
necessary to deal with such diverse and disorganized materials.Always try to get the latest edition, as they get out of date fast!The first one, as the name suggests, is specifically aimed atstudents and I think it is really useful, though a bit out of date:Winship, I. and McNab, A. (2002) The Student’s Guide tothe Internet. London: Library Association.Ó Dochartaigh, N. (2007) Internet Research Skills: How toDo Your Literature Search and Find Research InformationOnline, 2nd edn. London: Sage.Ó Dochartaigh, N. (2002) The Internet ResearchHandbook: A Practical Guide for Students and Researchersin the Social Sciences. London: Sage.Sherman, C. (2001) The Invisible Web: UncoveringInformation Sources Search Engines Can’t See. Medford,NJ: CyberAge.And now, some books about note-taking and organizing yourinformation. Most books about how to do research will have asection on this, but you may want to compare advice andapproaches with those given in this chapter. The first two areexamples of these:Blaxter, L., Hughes, C. and Tight, M. (2010) How toResearch, 4th edn. Buckingham: Open UniversityPress.Chapter 4 deals with reading and note-taking quitecomprehensively, though there is nothing on computer-basedbibliographic databases.Leedy, D. and Ormrod, J. (2015) Practical Research:Planning and Design, 11th edn. Harlow: PearsonEducation.
Chapter 3 gives advice on note-taking and writing your literaturereview.Here are some books more narrowly dedicated to taking notesand getting organized. The two Fry books complement each otherwell:McWhorter, K. and Sember, B. (2017) College Readingand Study Skills, 13th edn. Harlow: Pearson.Fry, R. (2011) Getting Organized, 4th edn. Boston:Centage Learning.Fry, R. (2012) Improve Your Writing, 6th edn. Boston:Centage Learning.Below is information about bibliographic database programs, thefive reckoned to be the best according to Scribendi Inc:www.scribendi.com/advice/reference_management_software_solutions.en.htmlCheck what is available on your college/university network beforeyou make an expensive decision to buy a program yourself ordownload one of the free ones. Remember, your university willprovide you with training and support on their preferred program.EndNote. Great if you are collaborating on a research paper orreport as it lets you share with up to 14 colleagues anywherethroughout the world. www.endnote.comRefWorks. Highly rated reference management softwareprogram. You have to pay for this one, but check first whetheryour college/university has subscribed to it.www.refworks.com/refworks2/default.aspx?r=authentication::initZotero. A free and easy-to-use tool to help you collect,organize, cite and share research. www.zotero.orgMendeley. Good for collaborative work, and particularly usefulin technical and scientific fields. www.mendeley.com
If you are interested in speed reading, here are some books. Butbeware, these skills take time and practice to perfect. Have youreally got the time now?Butler, D. (2017) Speed Reading with the Right Brain:Learn to Read Ideas Instead of Just Words. CreateSpaceIndependent Publishing Platform.Buzan, Tony (2009) The Speed Reading Book. London:BBC.Collins, S. (2018) Speed Reading: How to Absorb Tons ofText Ultra Fast and Remember the Most Important of It.CreateSpace Independent Publishing Platform.
5 Doing Your Literature Review,Forming Original Ideas andDefining Your Research TopicAimsTo provide practical advice on how to carry out your literaturereviewTo indicate how you can formulate original ideas for yourresearch topicTo suggest different ways of expressing your research intentionsIntroductionBecause a huge amount of relevant literature may be available,reviewing this literature ought to follow a systematic, meticulouslydesigned process, characterized by perceptive reading and anattention to detail, particularly that which relates to your researchproblem. Your reading will help you to become aware of howothers have dealt with topics in your research subject, of whatknowledge they have acquired, and of how to collect dataappropriate to your research problem. It will also help you todefine precisely what you intend to research in your own project.Doing a literature reviewWhat is a literature review, and why do you need to do one? Lookat it this way. When you have finished your thesis and come tosome really useful conclusions, will you not want other peopleinterested in your subject to read what you have written? Yourthesis should take its place in the collection of other writings, all ofwhich contribute to the sum of knowledge about that subject. So,in order to understand the present ‘state of the art’ you too need to
read what other people have written and make some kind of anassessment of where your research will fit into that body of work.Swales and Feak (2000, p. 115) explain that literature reviews fallinto two basic types: a survey article (an expert’s general review ofcurrent literature on a particular topic), and a review that formspart of a research paper, proposal, thesis or dissertation. We areobviously concerned with the latter, and in particular with theproposal literature review.A literature review is an essential part of any dissertation or thesisand is also required for most academic research papers andresearch reports. This type of review forms an importantintroduction to the research project, and apart from gauging thelevel and extent of knowledge in a subject, it provides afoundation for the formulation of your own research project. Byidentifying a gap in knowledge, such as an unresolved clash ofbeliefs, a lack of understanding of a situation, or revelation oftechnical or social problems that need to be resolved, claims thatcan be challenged, and advances in techniques and methods thatcan be developed, it provides a rationale for the definition of yourmain research question and underpins the argument about whyyour project is worth doing. It therefore forms a distinctlyrecognizable section near the beginning of your research proposalas well as your dissertation or thesis and leads on to the morespecific and practical description of the research activities.Not all of the material you will read will be relevant to your interests,and where it is relevant, not all of the ideas put forward will beagreeable to you. Doing a literature review means not only trackingdown all the relevant information but also taking a critical position onthe ideas contained therein, and thus producing an argument thatleads to the exposition of your research project.Critical reading skills
Critical reading is a skill that needs to be developed. Anyone whohas studied their language and its literature and has analysedimportant literary works will have an idea of some of the issuesthat need to be investigated. Critical reading of academic textsfaces similar challenges, as discussed below.First, before looking at the different aspects of criticism, what dowe actually mean by ‘criticism’? Perhaps a better word in thiscontext would be ‘analysis’, because the point of the exercise isnot only to denigrate or find fault with the style of writing or ideas,but also to present a critique, a scrutiny, an analysis, or anexamination of them. Providing a description is not enough; yourtask is to give your own personal and professional appraisal of thecontent and quality of the text in question. In order to be able to dothis, you will have to look at the text from different perspectives toreveal a multidimensional view of the work. So, what are theseperspectives?The structure of the argument. You can analyse this by firstdetecting the conclusion-type words or so-called conclusionindicators (e.g. ‘therefore’, ‘it follows that’, ‘as a result’, ‘we canconclude that’, and other such similar words and phrases) in orderto pinpoint the conclusion(s). The main conclusions shouldnormally appear towards the end of the work, though there maybe intermediate conclusions scattered throughout. There are thethree aspects that need to be examined when assessing thequality of the conclusions:1. What evidence is given to support the conclusions?2. Is the evidence credible, i.e. does it come from reliablesources?3. Is the logic of the argument sound, i.e. what are the steps inthe argument that lead from the evidence to the conclusions?You need to do this kind of analysis coolly, like a judge appraisingthe argument of a lawyer making a case. A useful technique is toextract the evidence and conclusions in the form of short phrases,and arrange them in the sequence of the argument. This lets youexamine the logic form without being distracted by thesurrounding text. This process sounds a lot simpler than it often is
– arguments can be quite convoluted or incomplete – but, just likea judge, you should make a note of your summing up of thisaspect of your critique. Chapter 3 goes into much more detailabout the characteristics of argument.The assumptions upon which the writings and arguments arebased. All writing is rooted in theory and based on values, andmust be appraised in relation to them. Sometimes they are quiteclearly stated at the beginning of the text; sometimes they areobscured or not mentioned. You will need to have someknowledge of the different theoretical positions in your subject inorder to be able to detect them and know what they imply. Somecommon examples are: a feminist approach in social science, aKeynesian approach in economics, a modernist approach inarchitecture, and a Freudian approach in psychology. In eachsubject there are competing theoretical standpoints with their ownvalues. Only by being aware of them can you make your ownconsidered evaluation of the literature.The wider context of the work. Intellectual work is carried out in acomplex arena where power, politics, fashion, economics,competing orthodoxies and many other factors play influentialroles. These can be determining factors in the formulation ofviews and need to be exposed in order to understand the forcesbehind them. For example, the forces behind the industrialrevolution were formative in the thinking of the day, just as thoseof the electronic revolution are today.Comparison with other work. There are no absolute values towhich you can appeal in order to make assessments. There areno clear rules about what is right and wrong. ‘Research writing isa contested terrain, within which alternative views and positionsmay be taken up’ (Blaxter, Hughes and Tight, 2010, p. 127).Critical reading can, however, be used to make comparisonsbetween texts in order to highlight the different approaches, levelsof thoroughness, contradictions, strength of arguments,implications of theoretical stances and accepted values, and typesof conclusions. This will enable you to group together or divide thevarious strands in the literature to help you map out the largerpicture that forms the background to your project.
‘The literature’ – what does that mean? In every subject there is avast number of books, papers, reports, documents, Web content,etc., new and old, that can be consulted; so where do you beginand, just as important, where do you stop! The main objective is tofind the relevant sources that will provide the answers to thequestions that you pose, which means that you have to firstdecide what you want to do the literature review for. In mostreviews, there are several outcomes that are aimed at, so it isgood to be clear about the different strategies you need to employto deliver the goods. Here are some of the main ones in Checklist5.1.Checklist 5.1 Different strategies you can use for your literature reviewCritical review – concentrate on the content and quality of eachpublicationState of the art – discover the latest findings and theoriesDifferent and competing theories – search out alternative ideasMethodology review – compare how the research was practicallycarried outCross-disciplinary exploration – explore the differentperspectives according to disciplineHistorical overview – how the subject got to where it isDifferent strategies explained inmore detailIn order to provide you with more practical details of how thesevarious strategies can be followed, here is an explanation of stepsto take in each.Critical appraisalThis is an exercise in the detailed examination of each publicationto gauge its quality. You need to ascertain whether it is a crediblepiece of work the content of which can be relied upon. With somepublications, this work has already been done for you. The editorsof academic journals subject the papers submitted to a peer
review process before they can be published, and many editedbooks and conference papers are treated in the same way. Butmost other sources of literature, particularly those found on theWeb, have not undergone any previous critical scrutiny. So whatdo you look for?In historical documents there are the questions of authenticity andbias to be answered. Is the document genuine? Is it obtained froma reliable source? For what purpose has it been written – is therea particular function that it has been aimed at, e.g. to support aparticular cause? Is it a contemporary account or has it beenwritten long after the event? Is the use of language the same astoday, or are the meanings different? Some documents may usearchaic symbols that are difficult to interpret.Newspaper and magazine articles and other publications from thepopular press are rarely scrutinized for accuracy and do notgenerally identify the sources of their information. Personalopinions and straw polls are often presented as facts, so cautionmust be observed when you want to refer to this kind of source.This does not mean that these reports are not useful. They maygenuinely reflect the spirit of the times and provide near-to-the-event accounts of happenings.For many Web-based materials it can be difficult to locate thesource, authorship and date. There is therefore little to go on fortesting the reliability of the material. Only use information that canbe authenticated in some way. It could be issued by a reputableorganization, a well-known expert or a government body. Ofcourse, many of the reputable books and papers are obtainablethrough the Web, so these pose no problem in authentication.Despite the approval of peer reviewers, or the respected status of theauthors, not all publications are of the same quality, so the criticalreading skills as mentioned above should be applied.State of the art
The latest publications will present the latest findings. Mostsubjects are being investigated by many researchers, so newknowledge is constantly being developed. In order to avoidrepeating work, you need to be aware of the current standing ofknowledge in your particular field of interest. You might find thatthere are review papers covering the latest advances in yourchosen subject. These provide a potted overview of the latestresearch by reviewing and summarizing all the latest publications.They can save you a lot of time as the hard work has been donefor you! Usefully, many research papers and theses mention at theend the areas where further research is required.As one of the main objectives of doing research is to develop newknowledge (a mandatory requirement of a PhD) you need toidentify those areas where there is a lack of knowledge orunderstanding. This will then provide you with indications of fruitfulsubjects for investigation in order to develop viable researchquestions which will be the foundations of your research project.Different and competing theoriesIn order to determine the current position of your subject, youshould discover not only what the orthodoxies are, i.e. the normalaccepted ways of thinking and the accepted assumptions – thepresent paradigm, but also those ideas which challenge the statusquo. The ‘mavericks’ often reveal that false assumptions havebeen made undermining the veracity of current thinking, or thatcertain influential factors have been ignored, or that the acceptedrules cannot be applied to certain situations.There sometimes is no firmly established orthodoxy but competingtheories, depending on how the evidence is interpreted. Forexample, global warming has its accepters and deniers. If you aregoing to research in such a situation, which side will you take andfor what reasons? You will need to make an assessment of thevarious theories and come to a reasoned conclusion.Methodology review
Every good research paper will explain how the research wascarried out, as well as what the results of the project were. This isreally useful for you to discover ways to collect data and toanalyse using methods that are suitable for the types of problemtackled. However, you should also make your own personalassessment of the suitability of the methods used in that research,and perhaps make comparisons with other ways that it has beendone in other research projects. Research methods are the toolsof research; using the right tools for the job is essential forassuring the quality of the outcome.Cross-disciplinary explorationNot all research projects can be neatly categorized by subjectdiscipline. Many, particularly in the social sciences andhumanities, cross disciplinary boundaries in order to provide afuller and probably more accurate account of the situation studied.You can see how aspects as diverse as economics, politics,psychology, society, traditions, beliefs, culture, etc. can influenceevents. In the sciences, the study of the interplay betweendifferent disciplines and scales has been a fruitful approach torevealing new understanding of phenomena and the developmentof new techniques and materials. Doing cross-disciplinaryresearch is demanding, especially if you are not a member of ateam. You need to have an understanding of the differentdisciplines in order to cope with the methods and terminologyinvolved. However, there is an argument that maintains that a newoutsider’s view of a subject can often reveal aspects ignored bythe cognoscenti due to their blinkered orthodox approach.You will need to make an assessment of the validity of the cross-cutting research, assess its advantages and disadvantages, andwhether it has led to a better result as consequence of thecombination of approaches.Historical overviewEvery discipline has a history. Often, how a discipline hasdeveloped is an important factor in where it is now. The advances
made in a subject are often determined by historical chance, thepersonal, social, political and economic factors that resonatedown to the present day. These developments alone can be afruitful subject of research. If you are planning to do a historicalstudy, obviously this kind of research documentation is anessential starting point for your project. Even if that is not yourobjective, the way that knowledge has been developed and themomentum that has been generated could be an important factorin the way that the state of the art has been achieved, and hasimplications for further study. Institutions have been established,heroes have been declared, research organizations have beenformed, funding sources have been created, all based on pastevents. The implications could be an important influence on yourown project.Doing the reviewThe literature review for a paper, dissertation or thesis will need tobe carried out normally using four of these strategies. They areshown in Checklist 5.2, arranged from the general to theparticular, their relative importance depending on the nature ofyour subject. This way of organizing your review provides a goodfoundation on which to base your research and place it in contextto your choice of subject.Checklist 5.2 Directions of investigation for literary reviews1. Research theory and philosophy: to establish the intellectualcontext(s) of research related to your subject.2. History of developments in your subject: to trace the backgroundto present thinking.3. Latest research and developments in your subject: to informabout the current issues being investigated and the latestthinking and practice, to discuss the conflicting arguments, andto detect a gap in knowledge.4. Terminology and research methods: to explore technical wordsthat are used and their meanings, and the practical techniquesthat have been used, particularly those that might be relevant toyour project.
You will find that doing the literature review is not a linear process.Reading, writing and reviewing form an iterative process: the moreyou read and learn, the more you will be able to understand theareas of consensus, conflict and uncertainty. Whilst it is importantthat you give the impression that you are widely read in thesubject, you must be selective in the references you use. You donot have much space to write the review. If the amount of relevantliterature is huge, you might have to restrict the scope of yourproject by narrowing down the issues addressed. It is easy to seethat, in many topics, economics, politics, philosophy, socialconditions, legislation, governance, etc. all play a part. But youcannot cope with all them! Pare your topic down to the core ofyour interest, and stay as much as possible within your area ofexpertise. On the other hand, if there seems to be a paucity ofliterature, this could be because your topic has not beenresearched much or that you have not spent enough timesearching! If there really has not been much research, then youhave an excellent reason for doing some.Style and content of the reviewOnce you have decided on the content of your review and thesequence required to support your argument, you will have toconsider the way that you present the review as a clearintroduction to your proposal.Assume that your reader knows nothing about your subject.University research committees, which vet student researchproposals at MPhil and PhD levels, are made up of academicsfrom a variety of disciplines. For funded research, however, it islikely that there will be experts in your subject area on the vettingcommittee, so you will have to make sure that the scope of theliterature is complete, i.e. that no influential texts have beenignored. It is best to start your proposal with a short but clearstatement of the focus of the proposed study. This is not part ofthe literature review, but sets out the subject of study. The reviewshould begin with a general outline of the features of the relevantliterature.
For example, accounts of polar exploration expeditions are not anew phenomenon. The tendency in early descriptions was to laudthe heroism of inspirational figures that had pioneered new routes,such as Cook or Amundsen. The more recent approach has beento analyse the physical and mental achievements of the explorersin relation to human strength and endurance. Most recently,exploration history has been examined in a context of social andpolitical power struggles, technological advances and scientificresearch.This gives you a good excuse to introduce literature that supportsyour introductory statements. You have a choice on how to refer tothe literature at different levels of detail. Hyland (1999, p. 350)usefully identified four levels of citation in decreasing order ofdetail, as listed in Checklist 5.3 (the citations are for illustrationonly). I have added a citation in its different forms to illustrate howthese are applied.Checklist 5.3 Levels of citation1 Block and indented quotationsThis is used for longer quotations – more than one phrase orsentence. Note that you need to include a page number for quotationsso that the reader can pinpoint the source within the cited publication.Ranger insisted on the need for good-quality design:Buildings will not function properly if insufficient consideration isgiven to solve complicated design problems. Poor design resultsin inconvenient planning, poor quality of spaces, highermaintenance costs and frequent breakdown of services. We oweit to clients to insist on the inclusion of sufficient resources todevote to the design activities in development projects. (Ranger,2017, p. 36)2 Within-sentence quotationsUsed for short quotations that sit comfortably within your text.
According to Ranger, ‘Buildings will not function properly if insufficientconsideration is given to solve complicated design problems’ (2017, p.36).3 Paraphrase/summaryRanger (2017) argued that neglecting the resolution of design issuesresults in poor quality buildings.4 Generalization (combining severalsources)There is wide agreement that lack of sufficient time and resources toresolve design problems in buildings results in poor qualitydevelopments (Hendry, 2018; Ranger, 2017; Smith, 2002; Waldren,1999).In the literature review for a proposal you will not have enoughspace to use block quotations, and only if the quotation isabsolutely central to your argument should you use a within-sentence quotation. As the aim of the literature review is to covera lot of ground, use should be made of the fourth option tocombine several sources, leaving the third option for significantviews of influential writers.With regard to the citation system, the examples above use theHarvard (author–date) system, but I recommend using the BritishStandard (numeric) system, as described in Chapter 8, if youneed to include a large number of references as this enables youto do so without laborious lists of names within your text.When making citations, take care to avoid ambiguity. It should beclear whether the author cited actually made the discovery orformulated the idea, commented on it, or compiled informationabout it. For example:Much has been written about the effects of poverty (Kraus,2015).
or:The relationship between poverty and health has beenextensively studied in recent years (Kraus, 2015).Does this mean that Kraus has written or studied a lot about ithimself, or has he made an analysis of the writings, or has hecompiled a list of relevant literature? More precision would clearup these ambiguities.Another issue to watch is the accuracy of your summaries of theideas and the fairness of the evaluation of the writings of others.You need to make the judgement as to the amount of informationgiven to provide an accurate impression of the original, and toparaphrase it in such a way as not to mislead or misrepresent thework.Checklist 5.4 gives a handy checklist of points for you to reviewthe content and form of your literature review, based on acompilation of comments from academics, as listed by Swalesand Feak (2000, p. 149). These repeatedly reiterate theconnection that should be made between your review of theliterature and your own research subject and how you will goabout researching it. No one wants to read literature reviews forgeneral interest – they should be aimed at a particular purpose.Checklist 5.4 Points to observe for your literature reviewMake sure that your review is not just a list of previous researchpapers or other literature, devoid of any assessment of theirrelative importance and their interconnections. Make anoverview of the literature to produce a guide to the rich interplayand major steps in the development of research in your subject.Check that the important issues of your research problem areintroduced through the analysis of the literature. A simplechronological account of previous research will not give asufficient thrust to the argument of why your research problem issignificant and how it continues the research effort.Ensure that the general theoretical background is intimatelyconnected to your examination of the more detailed writings
about ideas and specific research that led up to your ownresearch project. The theory should help the reader understandthe attitudes behind the reviewed literature and your ownphilosophical stance.Make links across discipline boundaries when doing aninterdisciplinary review, rather than keeping each separate andexamined in turn. Many research subjects cannot behermetically sealed within one discipline, so the connections arethere to be exposed. You might even be able to suggest somenew links that need to be investigated.Ensure that you have included some account of how theprevious research was done, so that you have a precedent foryour own approach to methodology.How many references should you have? This depends on thesubject, but as the literature review part of a research proposalhas to be very compact owing to limitation of space, you areunlikely to be able to cite more than 20–30 authors, and five to tenmight even be sufficient in a narrowly defined field. The importantthing is to select those that are really significant for your work. Asusual, it is a good idea to look at previous proposals in yoursubject area to see what has been successful before. Someexamples are also given below and in Chapter 8.Forming original ideas and definingyour research projectWhy do you need original ideas, and where do they come from?The point of doing any research is to produce new knowledge. It isthis quality of ‘newness’ that requires originality. Finding a topicthat no one has investigated, or has not analysed in the way youpropose, demands an element of originality that requires a level ofimagination which goes beyond considering what is alreadyknown. However, there is no need to engage in ‘blue sky thinking’,divorced from all context. Your literature review will provide youwith the context, and your analytical and logical skills will be usedto extract the research topic that you discover to be necessary toinvestigate.
In order to structure your argument for your research work, it isgood to take a systematic approach to arranging your notes fromthe literature into a logical sequence that leads to the statement ofthe main research problem from which you can derive your mainresearch question. Try Exercise 5.1 to set up a convincing casefor the proposed research project.Exercise 5.1Here is a useful technique you can use to structure your review in a waythat suits your subject and your argument, as used by Swales and Feak(2000, p. 143) in one of their exercises. You will find that it requires severalattempts to achieve a sequence that produces a good argument, backedup by relevant literature, which leads logically to the research problem.1. Select the notes you made from the literature that you think are mostrelevant to your project. Select only those notes that summarize thepositions, ideas or evaluations of the various writers. Also review anycomments you have made about the texts when you first read them.This is where brevity in the notes will be seen as a virtue.Alternatively, if the notes are too long, summarize each set of notesinto one paragraph that states the essential ingredients. Print theseout so that you can scan them easily by eye. You may have perhapsbetween 10 and 20 of these; more than this becomesunmanageable.2. Number the notes in any order.3. Draw a sequence of four or five paired boxes in a line. This will helpyou to map out the review of the literature together with the steps ofyour argument: one sequence of boxes to outline the steps in yourargument, the other paired boxes to list the supporting literaturenotes at each step (see Diagram 5.1).4. If you already have an idea about it, you might be able to insert theoutline of the argument in one line of boxes in the form of a shortstatement in each box in turn to trace the steps of the argument. Ifnot, leave these blank.5. If you have formulated the argument in this way, insert the numbersof the notes that underpin each step in the adjacent box.6. If you have not already inserted an argument, insert into the otherboxes the numbers of the notes. Group the ones that cover the sametopic into the same boxes and put the groups into a sequence thatleads from the general to the particular. For example, you may havenotes on general theoretical background to the subject, notes onviews about aspects of the subject that raise problems, notes on
solutions to those problems, evaluations of the already attemptedsolutions.7. Now work out the steps of an argument in the paired boxes.8. The last box should not have any literature notes numbers, as thelast step of the argument is to state the conclusion or, in effect, tostate the research problem.9. Add or subtract paired boxes as required: different arguments havedifferent numbers of steps.10. Review what you have constructed. Correct, strengthen or alter thesequence and content, or even explore new versions until you aresatisfied with the result.Diagram 5.1 Literature review boxes
Below, in Box 5.1, is an example of a literature review producedby a master’s student as an introduction to her proposal and todefine her research questions. Note how she uses the referencesto discuss the current theoretical standpoints on the subject andthen narrows down to the specifics of the situation she wishes toexamine, which then leads to her set of research questions.
Box 5.1 Example of a literature review in a research proposalForced migration: an understanding ofthe coping strategies of the Hazaracommunities in OxfordshireIt is necessary to understand Britain’s housing and employmentimmigration policy restrictions in order to comprehend the need forand type of coping mechanisms of the immigrant Hazara communitiesin Oxfordshire (Zetter and Pearl, 2000; UNHCR, 2013).Considering the UK immigration discourse, Article 1 (A) (2) of theUnited Nations Convention Relating to the Status of Refugees definesa refugee as, ‘someone with a well-founded fear of being persecutedfor reasons of race, religion, and nationality, membership of aparticular social group or political opinion’. In the light of this research,the terms ‘asylum seeker’ and ‘refugee’ will be aggregated together as‘forced migrants’ (FMO, 2015), to examine the coping strategies inboth phases of their lives.Looking at the Hazaras’ history, the largest Hazara migration anddisplacement (Ibrahimi, 2012) which included the author’s greatgrandfather, originated from the ‘Hazarajat’. It began between 1880and 1899, to Meshed-Iran, Russian Turkestan and Quetta-Pakistan(Gharjistani, 1986). It was as a result of persecution due to ethnicityand Shi’a sect (Bacon, 1958). The second wave of migration wasattributable to the resistance to the Soviet war (Ibrahimi, 2012). Andfinally, the Taliban/Lashkare- Jhangvi era led to another wave ofmigration from Quetta (HRW, 2014) and Afghanistan to Australia andEurope (Koser, 2013).Glancing at the main theme of ‘coping’, Baerenholdt and Aarsaether(1998:30) describe it as ‘how people engage in strategies which makesense to themselves’. Putman’s (1995; Zetter et al., 2006:9)pioneering typology of ‘social capital’, bonding (intra-community),bridging (inter-community) and linking (community-public agency) canbe used to understand the coping mechanisms of communities.However, ‘bonding capital’ may be necessary for refugees to ‘get by’(Begum, 2003), and can arguably cause social fragmentation(Haezewindt, 2003). A qualitative study by Begum (2003) and Korac’s(2005) ethnographic fieldwork found that, ‘bridging social capital’allowed successful integration with the host community. Mindful of thisresearch, it will be worth investigating this aspect amongst the Hazaracommunities in Oxfordshire.
Williams’ (2006) participatory observation study on the concept of‘coping strategy’ found that ‘transnationalism’ helped refugees stayconnected to their cultural identity. Likewise, Monsutti (2004) agreesthat the Hazaras in Afghanistan have developed strong translationalmigratory networks. This translational experience of refugees providesthe basis for the concept of ‘networks’ as suggested by Marx (1990) ina conceptual framework. More so, based on ethnographic evidencesMonsutti (2008) asserts that Hazara refugees use ‘tactics’ such astransnational networks to manage within limited means while stillmaking use of their social and cultural skills. It will be interesting tocompare how the diasporic discourse has helped the Hazaracommunities.Equally important, Ibrahimi (2012) emphasizes that, historically theinternational Shi’a religious institutions that are central to the Hazaraideology, laid the foundations of transnationalism. Such as,acknowledged by Creasy (2009:49) in her ethnographic study on theHazaras in Quetta, ‘physical space plays an integral role withinreligion’. Moreover, using RCOPE measure of religious coping,Pargament et al. (2000) found ‘religious capital’ (Baker and Miles-Watson, 2010) plays a multifunctional role by offering a variety ofways to manage in difficult situations. Therefore, it will be important tounderstand how religion and religious institutions in Oxfordshire havehelped the Hazara communities develop efficient coping mechanisms,in addition to how their ideology has evolved.Another important aspect is the work of Refugee CommunityOrganisations (RCOs) that offer invaluable cultural, emotional andeconomic support. In Britain, destitution amongst the refugees is theresult of the dispersal policy and inappropriate structure of socialsupport (Morrell and Wainwright, 2006). Remarkably, Crawley et al.(2011) in using the Participatory Ethnographic Evaluation andResearch (PEER) method discovered that, refugees rely heavily ontheir communities and their RCOs due to housing, employment andcultural barriers. However, this creates negative consequences ofdependency on informal contacts (Bloch, 2004) with low-paidexploitative work (Community Links and Refugee Council, 2011). It willbe of interest to understand in what capacity the informal networksand organisations have helped the Hazaras.Research questionsThe author’s interest in understanding the problems faced by theHazara community in Oxfordshire stems from her upbringing by herHazara grandparents in Quetta. This experience forms the foundationof the research question:
Forced migration: What are the coping strategies of the Hazaracommunities in Oxfordshire who have been subjected to forcedmigration?Moreover, several important questions were raised, such as:How has ‘transnationalism’ helped the Hazara community copeand has it played the main role in regaining the sense ofbelonging?What coping strategies have the Hazara adopted in the processof dealing with the UK’s asylum procedures?In the light of ‘bridging social capital’ how well is the communityintegrating within Oxfordshire?Fatima Hashmi, MA Development and Emergency Practice, OxfordBrookes University, 2015If you are doing a PhD direct, make sure that the review containssome of your own published (or even unpublished) work relevant tothe subject. You need to persuade the readers that you already haveexpertise in the chosen research topic, equivalent to the preparatoryMPhil stage of the work.The literature review within a dissertation or thesis will be muchmore extensive and detailed, covering all the aspects of theintended research. Here, in Box 5.2 is an example showing part ofa literature review for a PhD thesis. You will see that it is muchmore comprehensive than possible for a review in a researchproposal. It explores meanings of the concept of leadership.Notice how the terminology is carefully defined, the cited worksare used to provide evidence that there are many widely differinginterpretations of the concept and that there is no consensus onits main attributes. The review traces back to the first studies onthe subject and demonstrates that there is a conflict betweenthose who believe that leadership can be taught and those whobelieve it is an innate ability. It is clear that for the research tocontinue, a clear and detailed definition of the meaning of theconcept will have to be developed for the purposes of thisresearch.
Box 5.2 Example of a literature review in a thesisA part of the review that defines themeanings of the concept ‘leadership’Much research on leadership has been undertaken, with manyreferences taken from the 19th Century (Yukl, G. A., 2002). However,Alves et al. (2005, p. 9), posited that, ‘Initial leadership studiesfocused on major historical figures, such as Buddha, Mohammed,Gandhi, or Churchill, but the discussion of leadership as a processmay have been originated by Machiavelli in the sixteenth century’.Alves et al. (2009, p. 2) added that, ‘In the West, leadership has longbeen treated as a global field of study borrowing from Eastern andWestern classic teachings including Aristotelian, Confucian, andBuddhism, and more recent ones such as Machiavelli and Hobbes’.Yet Bennis and Nanus (1978) proposed it still remains one of the leastunderstood researched subjects to date. Leadership theories, modelsand definitions have emerged from numerous traditions and sectorsincluding those of industry, the military, the scientific world andacademia, and, more recently, philanthropy and informationtechnology.Due to these diverse sources of influence on the theory, there existsno single commonly-acknowledged or accepted definition forleadership (Cuban, L., 1988; Bass, B. M., 1990; Leithwood, K. et al.,1999; Yukl, G. A., 2002; Legace, M., 2010; Bush, T. and Glover, D.,2013). Jones (2005, p. 259) affirmed that, ‘The field of leadershipstudies has not succeeded in articulating a coherent, paradigm-shifting model or approach that both scholars and practitioners canaccept and work with’. Shebaya (2011, p. 65), in an astute explorationof leadership development, asked the following questions:If leadership is such a highly complex process, then where canwe find individuals that are able to master this complexity? Ifleadership is teachable and learnable, then how can we gainefficiency in developing its potential and the competencies itinvolves?However, according to Snook et al. (2012), the state of leadershipeducation lacked intellectual rigor and institutional structure. Just asthere is no consensus on whether there is one best leadership style(Legace, M., 2006), ‘There is clearly no consensus on the one bestway to teach leadership’ (Snook, S. et al., 2012, p. 3). Yet many
exponents have argued that leaders are born and not made (Carlyle,T., 1841; Galton, F., 1869; Bowden, A. O., 1926; Bingham, W. V.,1927; Vroom, V. H. and Yetton, P. W., 1973; Jago, A. G., 1982;Drucker, P. F., 1989; PIU, 2007). In contrast, many have stated thatleaders are made, not born (Fiedler, F. E., 1967; House, R. J. andDessler, G., 1974; Whitehead, J., 2002; Rooke, D. and Torbert, W. R.,2005; Bennis, W., 2009; Stacey, M. A., 2009). From bothperspectives, specific characteristics, competencies and attributesdeveloped, play a role in leadership. Discourse on the relationshipbetween the leader (as an individual in a position or role) andleadership (as a set of actions, processes or approaches) have onlyrecently come to the fore (Lyne de Ver, H., 2009; Mcllreavy, P. andNichols, C., 2013). Heifetz and Linsky (2002) differentiated ‘leader’from ‘leadership’, separating the act of leadership from one ‘greatman’ or person’s accomplishments. Further, leaders de-facto requirefollowers, be they groups, individuals or teams with and for whom theyfulfil numerous roles (Lyne de Ver, H., 2009). Earlier ideas, aspresented by Hofstede (1980, p. 57) proposed a more hierarchicalrelationship as, ‘The crucial fact about leadership in any culture is thatit is a complement to subordinateship’. Yet from an Asian SuccessFactors for the Leadership of IHDOs in South Asia perspective,leaders’ interactions with their teams are seen as more ‘interpersonal’than ‘hierarchical’ (Ling, W. and Fang, L., 2003).Source: Extract from PhD thesis by Adi Walker (2019).ConclusionsThe literature review forms the foundation of your research. It is inthis review that you place your research in the context of what hasalready been done and what is known, and argue for the need foryour proposed research in order to fill a gap in knowledge. It isthrough seeking this gap in knowledge that you form original ideasthat will inspire your own work and help to provide a focus for yourtopic.The review has several other purposes as well as that of definingyour project. It is a demonstration to the reader that you haveconsulted all the important literature on your chosen subject sothat you have a comprehensive understanding of the state of theart in the field and how it arrived at this stage. It is also used toexplore the methods that are commonly used to investigate the
particular issues, what data to collect and how, and how toanalyse them in order to come to conclusions. With this comes thedefinitions of the terminology used in the topic area in order toensure that correct meanings are attributed to the words used. So,in summary, remember the four directions of investigation for yourliterature review:1. Research theory and philosophy2. History of developments in your subject3. Latest research and developments in your subject4. Terminology and research methodsThe next steps: writing yourliterature review and defining yourresearch project, and exploringappropriate research methodsThe aims of this section are to get you to:use the notes you have made on your reading of the literatureto make a draft of your literature reviewdefine your research problem and derive your main researchquestiondetermine the appropriate type of research approach to usein answering the research questionexplore the types of appropriate research methods to use foryour data collection and date analysis.Checklist of activities that will progress yourresearchStep 1: Start your literature reviewUsing the framework in Diagram 5.1, construct the structure for yourliterature review. Use the notes you have collected so far to back up thesteps in your argument. You will probably not be able to complete it at this
stage, but you will be able at least to make a start – always the mostchallenging part of a task! As you read more and develop your ideas, youwill easily be able to adjust and add to the boxes until you (and yoursupervisors) are satisfied that the structure is sound. You can then write itout in full as part of your research proposal.Step 2: The first draftSubsequently, write and submit your completed first draft of your literaturereview to your supervisor in order to get feedback. Make sure that youhave covered the four directions of investigation and that your argumentfor doing your particular research project is clearly structured. Rely asmuch as possible on direct reports of research, i.e. those closest to theresearch efforts. Keep direct quotations to an absolute minimum, and onlyuse them if they brilliantly express a particular issue or controversy. Onlyuse overviews if these provide a useful wider perspective. Your reviewshould culminate in your main research question and better still, your sub-questions too.Step 3: Stop!Know when to stop! Only review literature that is relevant to your subjectand contains something valuable to inform your investigations and furtherthe argument of why you should select your topic of research.Step 4: Explore appropriate methodsThe two last aims of this section – determining the appropriate type ofresearch approach to use in answering the research question, andexploring the types of appropriate research methods to use for your datacollection and data analysis – should not be part of the literature review.You should make separate notes on these issues for use in your researchmethods chapter. List your main research question and the derived sub-questions and insert below each sub-question the kinds of data(quantitative, qualitative, etc.) that you will need to collect in order toanswer it, the methods you will utilize to collect that data, and those thatyou will use to analyse the mass of data collected in order to answer eachquestion. At this stage these descriptions will be quite tentative, based onwhat you have learned in the literature review. You will be able to firm upon them when you have learned more about the choice of researchmethods available from the various textbooks dedicated to types ofresearch methods.
Consolidation and assessmentWhen you are prepared, you should arrange a tutorial with yourtutor or supervisor. Referring to your first draft, explain how youhave structured your literature review with the aid of Diagram 5.1,and run through the argument as you have structured it. You mightalso be able to provide one or two examples of the notes that youwill use to back up the crucial steps of your argument. The mainoutcome of this discussion should be whether your argumentappears to be sound, and whether it reveals a research problemthat is researchable in the context of your studies. This is also agood time to discuss with your supervisor the types of datacollection and analysis methods you have read about in theliterature and if and how these could be utilized in your research.Further readingI have given a short description of what you may find useful in thefollowing books on doing a literature review. You will also findsections in all research methods books on this.Hart, C. (2001) Doing a Literature Review. London: Sage.Ridley, D. (2012) The Literature Review: A Step-by-StepGuide for Students, 2nd edn. London: Sage.Machi, L. (2016) The Literature Review: Six Steps toSuccess, 3rd edn. London: Corwin/Sage.Fink, A. (2013) Conducting Research Literature Reviews,4th edn. Thousand Oaks, CA: Sage.These books provide guides on how to search the literature in thesocial sciences. Useful strategies are described for efficientreading, conducting searches, organizing information, and writingthe review itself. Examples of best and worst practice drawn from
real literature reviews are included to demonstrate how theguidance can be put into practice.For books dedicated to doing literature reviews in morespecialized fields see:Aveyard, H. (2014) Doing A Literature Review in Health andSocial Care: A Practical Guide, 3rd edn. London: OpenUniversity Press.Bettanysaltikov, J. and Mcsherry, R. (2016) How to Doa Systematic Literature Review in Nursing, 2nd edn.London: Open University.
6 Honesty and Research Ethics:Establishing an Ethical CodeAimsTo establish honesty and responsibility in research workTo consider ethics in research projectsTo ensure the ethical treatment of research participantsTo check the ethics of all research methods usedTo understand the purposes of research organizations andethics committeesIntroductionAlthough there always was a keen awareness of ethical issues inprofessional research, increasing attention is being awarded toresearch carried on at student level. This is particularly the casefor students doing dissertation for research degrees, wherecarrying out the research goes beyond being an educationalexercise and is likely to be of wider significance and to bedisseminated to the public at a professional level.There are two perspectives from which you can view the ethicalissues in research. One is concerned with the values of honesty,frankness and personal integrity, the other with those of ethicalresponsibilities to the subjects of research, such as consent,confidentiality and courtesy.Whilst the principles underpinning ethical practice are fairlystraightforward and easy to understand, their application can bequite difficult in certain situations. Not all decisions can be clear-cut in the realm of human relations. It is therefore important thatyou discuss the ethical issues thrown up by your research withyour supervisors and others who are specialists in this field. If youwill be working with human participants, it is likely that you will
have to obtain some kind of ethical approval from your universityor organization. It is necessary for you to find out what conditionsapply in your situation.Honesty in your work and avoidingplagiarismFirst, consider those issues that are concerned with researchactivities generally, and the conduct of researchers in particular.Honesty is essential, not only to enable straightforward, above-board communication, but to engender a level of trust andcredibility to promote debate and the development of knowledge.This applies to all researchers, no matter what their subject.Although honesty must be maintained in all aspects of theresearch work, it is worth focusing here on several of the mostimportant issues.Intellectual ownership and plagiarismUnless otherwise stated, what you write will be regarded as yourown work; the ideas will be considered your own unless you sayto the contrary. The worst offence against honesty in this respectis called plagiarism: directly copying someone else’s work intoyour report, thesis, etc. and letting it be assumed that it is yourown. Using the thoughts, ideas and works of others withoutacknowledging the source, even if you have paraphrased intoyour own words, is unethical. Equally serious is claiming soleauthorship of work that is in fact the result of collaboration oramanuensis (‘ghosting’).PlagiarismIt is very easy to fall into the trap of copying and pasting text fromthe Internet or other resources and inserting it into your ownwriting as a quick way to get words on the page and make goodprogress. How will anyone know that they are not your ownwords? After all, the Internet is a huge resource and so difficult to
trace the origin of any texts! Don’t be fooled. Text analysisprograms such as Turnitin, used by almost all universities andother academic organizations will immediately reveal the amountof plagiarism and sources of material in any given writing. Thisincludes the work of other students submitted through Turnitin orother program in the past.Different countries have different attitudes towards the use of otherpeople’s writings and ideas. If you come from a country that has arelaxed policy towards citation and referencing, where copying andpasting from the literature and Internet without making reference tothe sources is tolerated, be careful not to follow this approach whenyou do work in a strict academic environment. Carefully cite andreference all material used from literature sources using thetechniques detailed in this chapter.Figure 6.1 The worst offence against honesty is called plagiarism
Even if you did not intend to use the downloaded material in yourwriting, it is often difficult to remember whether notes you madesome time in the past were your own or what you downloaded forfuture reference. How do you avoid falling into this trap? By clearlydistinguishing between your own observations and thoughts andthose that you have obtained from the literature.Acknowledgement of other people’s workthrough citation and referencingObviously, in no field of research can you rely entirely on your ownideas, concepts and theories; therefore standard practices havebeen developed to permit the originators of the work and ideas tobe acknowledged within your own text. These methods ofreference provide for direct quotations from the work of others andreferences from a wide variety of sources (such as books,journals, conferences, talks, interviews, TV programmes), andshould be meticulously used. You should also acknowledge theassistance of others and any collaboration with others.There is a level of judgement required about when you need tocite a source and when not. There are many things that you willwant to write about that are not your original thoughts but can beconsidered to be common knowledge, such as the issue of globalwarming, the existence of poverty, the high pay received byfamous footballers, scientific facts and suchlike. There is no needto find a reference for these – it would be a laborious andcumbersome obligation to back up every single statement youmake.Then there is the question of paraphrasing the work of others.Apart from inserting exact quotations from secondary sources,you will always be paraphrasing what was written, sometimesclosely aligned with individual statements, sometimes forming amuch broader overview of a collection of attitudes or beliefs. In thefirst case, who to cite is obvious, but in the second, it might entailciting the person or people who were the main proponents of anattitude or belief, or refering to a source that provided an overview.The safest approach to take is – when in doubt, cite!
There are many standard citation and referencing systems,each with their own style that indicates the exact form of citationand referencing to be used. The styles all rely on a shortindication of the fact that a particular section of your text, whetherin the form of a direct or indirect quotation, should be attributed tosomeone other than you. This part is called the citation. It can bein the form of a number or an author name and date sometimeswith page numbers too, or occasionally as a footnote to the page.The citation refers to a more detailed account of the source of theinformation, contained in a list of references, normally placed atthe end of the paper, chapter or report. This list provides enoughinformation for the reader to track down the actual paper, book,website, etc. from which the information was obtained.CitationYour university or other organization will have their preferred styleof referencing which is sometimes strictly enforced. Journalpapers and publishers will also have their chosen style. Checklist6.1 has a list of some of the more common styles in use.Checklist 6.1 Common styles of referencingHarvard, an example of author–date citation, is very commonand is used across most subjects.APA (American Psychological Association), also author–datecitation, is used by education, psychology, and sciences.MLA (Modern Language Association), author–page numbercitation, is a style used by the humanities.Chicago A (and B), footnote (author–date) citation, is generallyused in the humanities.Turabian style, either footnote or author–date citation, isgenerally used by business, history, the fine arts, languagestudies and social sciences.OSCOLA (developed by the Faculty of Law at the University ofOxford), footnote citation, is used to cite authorities, legislationand other legal materials. It is the predominant citation style forlegal academic writing in English.Vancouver (developed by the International Committee of MedicalJournal Editors (ICMJE)), numeric citation, is mostly used in theworld of medicine.
IEEE (Institute of Electrical and Electronics Engineers), numericcitation, is used in engineering disciplines.NLM (National Library of Medicine), numeric citation, is usedprimarily in the medical world.The Harvard system of referencing is explained in Chapters 4 and8. There are reference managing software programs (e.g.EndNote, Mendeley – seehttps://en.wikipedia.org/wiki/Comparison_of_reference_management_software for a comprehensive list) which are accurate andtime-saving utilities that enable you to develop your list ofreferences, and change them almost instantaneously from onestyle to another and act as a database for all your references andappended notes.Responsibility and accountability of theresearcherYou do have responsibilities to fellow researchers, respondents,the public and the academic community. Apart from correctattribution, honesty is essential in the substance of what you write.Accurate descriptions are required of what you have done, howyou have done it, the information you obtained, the techniquesyou used, the analysis you carried out, and the results ofexperiments – a myriad of details concerning every part of yourwork.Data and interpretationsThere is often a temptation to be too selective in the data usedand in presenting the results of the analysis carried out. Silentlyrejecting or ignoring evidence that happens to be contrary to one’sbeliefs constitutes a breach of integrity. What could be of vitalimportance in developing a theory could be lost. For example, thehypothetico-deductive method depends on finding fault intheoretical statements in order not only to reject them but to refinethem and bring them nearer to the truth. History is full of instancesof scientific fraud. Here are two examples. The first is a provencase of tampering with medical records in order to mislead others,
initially reported in The Times on 7 June 1995. This was aparticularly dangerous kind of fraud, which might have led tountested and potentially harmful treatments being carried out onpregnant women. The Times on 19 June 1995 reported thereaction of the academic establishment, and the estimates of theincidence of medical fraud. This is summarized in Box 6.1. It isevident from the article that vested interests and ambition canlead to strong temptations to ‘arrange’ evidence in order to reachthe ‘right’ results. This can occur even at a small scale at studentand practitioner level.Box 6.1 Fraud through fabrication and lyingGynaecologist’s research fraud promptsneed for ‘fraud squad’Calls were made by the Royal College of Physicians to set up a‘research fraud squad’ after an eminent gynaecologist was ordered tobe struck off the medical register because he fabricated research workand lied about performing a pioneering operation.The public should be assured that scientific research is free of bogusresearch claims. Science must be seen to be conducted with integrityand honesty, and any transgressions should not be covered up by theprofessions. Reliable estimates of fraud in medical science are difficultto come by, but enough concern is expressed by scientists to warrantimmediate action. A central body made up of experts and otherinfluential people will be established to coordinate investigations.The case which prompted this call was that of [consultant obstetricianMalcolm] Pearce, who invented research findings in ectopicpregnancies on the basis of non-existent trials. He obtainedsignatures on his fabricated reports from colleagues who were notable to verify his work. Later, he claimed to have successfullyconducted the first operation to successfully transfer an ectopicpregnancy. He was found guilty of creating a web of lies in an internalenquiry and was struck off the medical register to protect the public.The great danger of scientific fraud is that researchers follow in goodfaith the techniques and treatments described as being successful inpublished papers. The dangerous consequences of this to patientsand to the development of knowledge are obvious. The problem iscompounded by the reluctance of medical institutions to investigate
suspected instances of fraud because of a fear that their reputationsmay be damaged.(Based on an article in The Times, 19 June 1995)The second case is an attack on one of the most reveredintellectuals of our time for suspected falsification of evidence, asreported in The Times on 5 June 1995 and summarized below inBox 6.2. Apart from the accusation of deliberate falsification, thisreport demonstrates the hazards of basing the proof of a theoryon the results from just one case, which if discredited, willundermine the theory completely.Box 6.2 Fraud by falsifying evidenceJung was a fraud, claims harvard scholarRichard Noll, a prize winning academic at Harvard University, claimsthat he can prove that Carl Jung, a founder of psychoanalysis, was afraud. He believes that the proof that he falsified evidence to promotehis theories lies in archives of the Library of Congress, but that accessto them has been blocked by the Jung family. Jung developed atheory that humanity shares archetypal images in its collectiveunconscious, and asserted that this was supported by evidencecollected by one of his assistants, J.J. Honegger, in 1909. Accordingto Honegger, a male patient he interviewed, known now as SolarPhallus Man, reported that he saw a vision of the sun with a phallus.This image was similar to those found in the ancient Greek cult of thegod Mithras, thus proving, according to Jung, that this archetypalimage had persisted throughout the ages.Noll claims that Jung deliberately falsified the research notes in orderto substantiate his claims. Because the mythology of Mithras wasbeing popularized at that time in Germany, Jung changed the dates onthe notes to before the publication of popular books on this subject inorder to counter any possible assertions that the man had read aboutthe solar phallus and recalled it during the interview.This makes Jung, according to Noll, ‘the most influential liar of the20th century’.(Based on an article in The Times, 5 June 1995)
A more recent occurence is the case of Andrew Wakefield, who‘altered numerous facts about patients’ medical histories in orderto support his claim to have identified a new syndrome’ resultingin the MMR scare (https://www.bmj.com/content/342/bmj.c7452).It is difficult, and some maintain that it is impossible, to be freefrom bias. However, distorting your data or results knowingly is aserious lapse of honesty. Scientific objectivity should bemaintained (or attained as nearly as is practical). If you can seeany reason for a possibility of bias in any aspect of the research, itshould be acknowledged and explained. If the study involvespersonal judgements and assessments, the basis for these shouldbe given. The sources of financial support for the researchactivities should be mentioned, and pressure and sponsorshipfrom sources which might influence the impartiality of the researchoutcomes should be avoided.It is good practice to admit to limitations of competence andresources. Promising more than you can deliver can be seen as notonly foolhardy but also dishonest.Exercise 6.1Read the accounts in Boxes 6.1 and 6.2 about instances of proven andsuspected scientific fraud and answer the following questions:1. What type of fraud was Malcolm Pearce convicted of (Box 6.1)?2. What dangers might have resulted from the misinformation hedisseminated?3. What safeguards incorporated into scientific method are intended toovercome the possibility of this kind of deception?4. Is the call for a ‘fraud squad’ realistic? Give reasons for youropinions. What difficulties do you think it would face?5. Is the lie that Richard Noll accuses Jung of perpetrating in Box 6.2 ofthe same type as the medical case cited in Box 6.1? What are thedifferences, if any?6. What do you think are the reasons that make some researchers lie?Why do they not do it more often?
Where do you stand?There are often lively debates about how research should becarried out, and the value and validity of the results resulting fromdifferent approaches. The theoretical perspective orepistemology of the researcher should be made clear at theoutset of the research so that the ‘ground rules’ or assumptionsthat underpin the research can be understood by the readers and,in some instances, the subjects of the research.EpistemologyAlthough others might disagree with your epistemology, at least itwill be clear to all as to what it is. In many subjects it will initiallybe a challenging task to become aware of and to understand allthe current and past theoretical underpinnings to relevantresearch. One of the principal functions of doing backgroundresearch is to explore just this aspect, and to come to decisionson theory that will form the basis of your research approach. Youwill have the opportunity to make this clear in your researchproposal.Doucet and Mauthner stressed the need, while doing qualitativeresearch, to reflect on the intertwined ethical, methodological andepistemological processes in order to attain a high degree of‘epistemic responsibility’ through ‘knowing well’ and ‘knowingresponsibly’ (2002, p. 139). They maintained that the reflexivity ofa researcher should take into account those theoretical,epistemological and ontological models that inform and influenceour knowledge construction, and argued that it was important tomaintain relationships with participants who conflicted with thosemodels, particularly during the data analysis processes. Dataanalysis is an ethical issue and data analysis methods are notethically neutral. They are founded on both ontological andepistemological assumptions.Situations that raise ethical issues
Now let us consider ethics in terms of the personal relationshipsoften involved in research projects. Social research, and otherforms of research that study people and their relationships to eachother and to the world, need to be particularly sensitive aboutissues of ethical behaviour. As this kind of research oftenimpinges on the sensibilities and rights of other people,researchers must be aware of necessary ethical standards thatshould be observed to avoid any harm which might be caused bycarrying out or publishing the results of the research project.Checklist 6.2 provides a list of aspects of research that involveethical issues.EthicsChecklist 6.2 Aspects of research that involve ethical issuesResearch aims – what are the associated benefits or risks?Means and ends – what are the consequences of using theproposed methods?Relations to other people – what effect might the research haveon participants? This is a complex subject with many facets toconsider.In order to explore these issues further, here is a closerexamination of the characteristics of the ethical considerations ineach.Research aimsThe aims of the research can be analysed from an ethicalviewpoint. Is the research aimed merely at gaining greaterknowledge and understanding of a phenomenon? If so, this kindof quest, seen in isolation, has little or no ethical consequences:the expansion of scientific knowledge is generally regarded as agood thing. But what might be done with the knowledge gained?The developers of the atomic bomb relied on the research innuclear physics. Research in genetics resulted in the ability toclone a sheep, and in the future, perhaps humans.
Genetically modified crops are seen by some as a potentialsaviour for farmers in developing countries, and by others as agreat danger to our ecology. Is it the responsibility of theresearchers in exploratory research to predict what might bedeveloped as a result of their discoveries? The ethical debatearound genetic research, stem cell research and many otherinnovative advances is grappling with these issues. Is theargument by some scientists that if they do not do the work thenothers will good enough to overcome ethical scruples?The aims of applied research are more easily subjected to ethicalinvestigation. A series of questions can be posed to tease out theethical issues. First, are the aims clearly stated? Are the aimslikely to be achieved by the outcomes of the research? Will theresults of the research benefit society, or at least not harm it? Willthere be losers as well as gainers? You will have to argue that theaims of your research are in accordance with the ethicalstandards prescribed by your university or organization.It might not be easy to persuade everyone about the benefits thatmight accrue from your research because of the variety ofviewpoints and vested interests. For example, research into howsweatshop workers in developing countries could assert power togain better working conditions might be regarded by the factoryowners as subversive and a threat to the economy and thestability of society.Aims that are too ambitious, that cannot be achieved by theplanned research, can be seen as a form of deception, or at leastof self-delusion. It is necessary to be realistic. The intention toachieve the stated aims is a sort of promise that needs somecredibility of being fulfilled. There are always risks in doingresearch: the future cannot be predicted, but wild speculation andrash promises have no place in academic or professionalresearch. Not only are the subjects of your research relying on thefeasibility of the outcomes, but so too are the funders or othersupporters of the project.Means and ends
How the aims, however laudable, are achieved should also beexamined from an ethical viewpoint. ‘No gain without pain’ is apopular expression – but can this approach be justified in aresearch project? There are many famous controversies thatsurround this issue, for example the horrific medical experimentson detainees carried out by Mengele in Nazi Germany, the resultsof which are reputed to have been transported to the US after thewar to inform the development of defence methods; or, on a quitedifferent level, the experiments on animals for development andtesting of medicines; or the research at the British Common ColdResearch Centre, which used to induce a cold in volunteers in theattempt to find a cure; or the growing of test areas of GM crops onopen farmland.There might be several ways that the research aims can beachieved. You should look at the alternatives to see if there areany ethical implications in the choice. Also, examine if there mightbe unintended or unexpected side effects of your methods. If theonly means of achieving a beneficial end is at the expense ofdiscomfort or inconvenience, is it the sufferers that are likely to bethe beneficiaries? Whatever the situation, the nature and theextent of the ‘pain’ must be assessed, consent must be obtainedfrom the participants, and of course, clearance must be given bythe relevant ethical committee.Ethics in relation to other peopleQuite obviously, research ethics are principally concerned with theeffects of research on people, and importantly, on those peoplewho get involved in the research process in one way or another. Itis the researcher who plans the project who has the responsibilityto predict what the effect will be on those people that he/she willapproach and involve in the research, as subject, participant,respondent, interviewee, etc.TerminologyBefore going into details about the process of the research, it isworth discussing briefly the important influences of terminology
used during the research:1. in the way you use language in your written work2. in relation to people and their roles.Let us look at the use of language first. According to an OpenUniversity (2008) guide to language and image, there are fiveaspects to be aware of when writing, as given in Checklist 6.3.These are aimed at you being as neutral as possible in the use ofterminology involving people – who and what they are, and whatthey do.Checklist 6.3 What to be aware of when writingAccessible writing stye: use plain English to make the textaccessible to everyone.Age: avoid being patronizing or disparaging.Cultural diversity: avoid bias, stereotyping, omission,discrimination.Disability: avoid marginalizing, patronizing.Gender: avoid male-centricity, gender stereotyping.Sexual orientation: avoid prejudice, intolerance, discrimination.So many words and phrases are in common use that makeunwarranted assumptions and assertions about people, or are atleast imprecise and possibly insulting. Acceptable terminologychanges with time, so you should be aware that what is used insome older literature is not suitable for use now. You need to beconstantly aware of the real meaning of terms, and their usewithin the particular context. In order to test yourself on yourawareness, try the following exercise.Exercise 6.2Read the following extracts from a fictitious report about the results of asurvey and consider the suitability of the terminology used with regard topeople. You will notice that there are numerous words that arediscriminatory, derogatory or too imprecise. These would certainly beunacceptable today. Suggest better alternatives to these terms and correctanything that you find that makes assumptions that are inaccurate. For
official advice you could usefully refer to the UK National Statistics website‘Ethnic group, national identity and religion’ on:https://www.ons.gov.uk/methodology/classificationsandstandards/measuringequality/ethnicgroupnationalidentityandreligion#introduction-to-collecting-data-on-ethnic-group-religion-and-national-identityReportThe research project was centred on comparing the opinions on streetcrime of a large range of people, from old folk in care homes to youngsterson the street. The results of the research are summarized as follows:The old dears were, in general, most influenced by what they saw on thenews on television. Half of the normal ones feared going out after darkalone, and almost all of the disabled, such as the blind, the deaf, thewheelchair-bound and those with a mental handicap, would not go outunless accompanied.Middle-aged professional people such as businessmen, doctors,headmasters, etc., were not afraid to walk through the city centre at night,though they took simple precautions such as putting their wallets into theirtrouser pockets instead of in their coat pockets.Of the lower-class people, it was generally charwomen, deliverymen, 24-hour shop salesgirls and waiters who felt particularly vulnerable, as theytended to be out late at night.Policemen on the beat were aware of some problems in town centres atthe weekends, but most laymen thought that street crime was a problem atall times. British people in particular blamed the violence on other ethnicgroups, such as the blacks and other immigrants.When asked if street violence was a youth issue, youngsters generallysaid no, blaming it instead on druggies, homosexuals and tramps. Theycomplained about the intolerance of geriatrics who did not respect theirway of life. A typical young person was likely to think himself morestreetwise than any other age group, citing quick-wittedness and agility ashis main attributes.Roles of researcher and researchedHow will you present yourself in the role of the researcher? As astudent researcher, you can present yourself as just that, givingthe correct impression that you are doing the research as an
academic exercise that may reveal useful information orunderstanding, but do not have the institutional or political backingto cause immediate action. If you are a practitioner embarking onresearch, e.g. a teacher researcher, a nurse researcher or a socialworker researcher, then you have a professional status that lendsyou more authority and possibly power to instigate change. Thismight influence the attitude and expectations of the people youinvolve in your project.The words subject, participant, respondent, interviewee andother similar labels have shades of meaning that suggest ordenote relationships with the researcher and status in theresearch project.SubjectParticipantThe word ‘subject’ perhaps carries the implication thatsomething is being done to them, while the term ‘participant’implies that something is carried out in conjunction with them.During the research process, the distinction could be quiteimportant. (Oliver, 2010, p. 4)How one behaves with people during the research sends outstrong signals and might raise unforeseen expectations. Stoppingpeople in the street and asking them a set of standardizedquestions is unlikely to elicit much engagement by the subjects.However, if you spend a lot of time with a – perhaps lonely – olderperson, delving into his or her personal history, the more intimatesituation might give rise to a more personal relationship that couldgo beyond the simple research context. How ‘friendly’ should youbecome? Even more expectations can be raised if you areworking in a context of deprivation or inequality: will the subjectsbegin to expect you to do something to improve their situation?Thus the place and the degree of involvement are importantfactors in the characteristics of the research. Making observations
in a public place will not result in the subjects of the researchbeing affected by your demeanour, but the more private andindividual the context, the more attention will have to be paid toone’s personal attitude and behaviour.Your status and role might put you in a dominant position thataffects the free action of the participants in the research. If you arean employer researching your staff, a teacher with a project aboutyour students, or a doctor or nurse investigating your patients, itmight be difficult for the participants to refuse to cooperate orwithhold information or give true opinions. Perhaps, as a ‘wealthy’Westernized person engaged in research in a deprived rural areain a developing country, you might be accorded additional respectand authority. Care must be exercised not to take advantage ofthese situations.Thus practitioner researchers, both student and professional, faceparticular problems in avoiding any confusion between their rolesas practitioners and as researchers. Their concern must be how to‘manage’ their responsibilities towards clients, students, patients,etc., fellow practitioners, organizational bodies, other researchersand academic institutions, in ways that all parties should considerethical. The main issues of concern are those of confidentialityand negotiation within the research process, access to subjects,seeking informed voluntary consent, and whether or not toemphasize the role of practitioner when carrying out research(Bell and Nutt, 2002, pp. 70–5).ParticipantsParticipants, subjects, respondents or whatever term you wish touse for the people you will approach for information in yourresearch, need to be treated with due ethical consideration, bothon their own part and on the part of the information they provide.There is a series of issues that need to be considered when youuse human participants. Here in Checklist 6.4 is a summary ofthese issues for you to consider. This is followed by more detailedcomments of what to take into consideration in each issue.
Checklist 6.4 Ethical issues with regard to participantsChoosing participants – how you select participants in a fair wayFreedom from coercion: reward or not? – ensuring that no one isforced to take partInformed consent, written consent – being sure that participantsknow what they are letting themselves in forPotential harm or gain – avoiding detrimental effectsParticipant involvement – making sure that participants arecomfortable with what they are asked to doSensitive material – making decisions on how to handleimplications of intimate revelationsHonesty, deception and covert methods – attitudes towardsbeing upfront with your participantsChoosing participantsIn some cases, participants choose for themselves whether totake part in a survey. If you simply drop off a questionnaire at theirhouse, they are quite free to fill it in or not, assuming that there isnothing in the questionnaire that threatens or otherwise affects afree choice. There are situations, however, where pressure,inadvertent or not, might be exerted on participants. Enlistingfriends or relatives – people who feel they have an obligation tohelp you despite any reservations they may have – could result ina restriction of their freedom to refuse. Leaving too little time fordue consideration might also result in participants regretting takingpart.When you are making observations in a public place, the subjectsof your study are obviously not in a position to choose whetherthey are observed or not. Measures must be taken to protect theiranonymity (e.g. if photographed). If you cannot do this, you mustapproach the subjects to give them the choice of featuring in yourresearch or not.If the choice of participants is made by others, e.g. by a‘gatekeeper’ (someone with overall responsibility for a group ofpeople, usually within an organization), the basis on which thechoice is made should be agreed, and assurance gained that
participation is voluntary. Issues of bias should be considered, e.g.the head teacher might only want the best pupils to bequestioned, or the factory owner might only allow access tocompliant workers. Alternatively, the boss might bring pressure tobear on a participant to take part against his/her will. Considerwhether any restrictions imposed by the gatekeeper willcompromise the research.Figure 6.2 The subjects of your study are obviously not in aposition to choose whether they are observed or notFreedom from coercion: reward or not?Obviously dishonest means of persuasion, for example, posing asan official, making unrealistic and untrue promises, allowing thebelief that you have come to help, being unduly persistent, andtargeting people in vulnerable situations, must be avoided.Although it is easy to detect crass instances, you could employthem almost inadvertently if you are not alert to people’s situationsand reactions.The question of whether, what and how much to rewardparticipants is one that is not often posed in research studentprojects, as the financial means are rarely sufficient to cover suchincentives beyond perhaps the inclusion of reply-paid envelopes.However, in funded research this could be a real issue. Somecommensurate recompense for time and inconvenience canusually be justified. Rewards that might unduly influence an
individual’s decision to take part are not suitable, nor are thosethat might influence their responses to questions. You will have tojustify the need for and nature of any rewards in your proposal.Informed consent, written consentAn important aspect of the decision of individuals to take part ornot is the quality of the information they receive about theresearch, enabling them to make a fair assessment of the projectso that they can give informed consent. The form that thisinformation takes depends on the type of respondent, the natureof the research process and the context.Informed consentThere may be several layers of consent required. When workingwithin an organization, the managers or other people with overallresponsibilities may need to be consulted before the individualparticipants. There could be a conflict of interest between themanagement and the employees. It must be made clear andagreed at all levels how the investigation will be conducted, howconfidentiality will be maintained, and what issues are to bediscussed. This is a particularly sensitive matter in cases wherecriticism may be made of persons, organizations or systems ofwork or conditions. There must be some obvious form ofprotection for those making criticisms and those at the receivingend.Clarity, brevity and frankness are key attributes in providinginformation on which consent is based. Verbal explanations maysuffice in informal situations, though a written résumé on a flyercould be useful. Questionnaires should always provide thenecessary written information as an introduction. In many cases,the wording of any information may have to be simplified in orderthat the nontechnical person can understand; do not try to impresswith technical jargon. The participant needs to know: who is doingthe research, what are its aims and objectives, what theparticipant is being asked to do and why, how the informationgiven will be used, and what steps are being taken to ensure
confidentiality and anonymity. Time may be needed for theparticipant to consider the implications of taking part, and perhapsalso for consulting others.The form of consent must also be considered. Is verbal consentsufficient? Normally it is when doing informal interviews in publicplaces, and the questions are not too intimate or controversial. Inthe case of a simple questionnaire, the fact that the respondenthas completed it and returned it is usually sufficient.When dealing with organizations, written consent is alwaysrequired, not least by you, to cover you against any possiblefuture difficulties. In more complex forms of data-gathering, e.g.experiments, trials, extended interviews, it is good practice to getwritten confirmation from participants that they understand what isrequired and agree to take part.Getting consent from vulnerable people (this includes children,some older people, people who for a variety of reasons are unableto read or understand the language in which the research is beingcarried out, and those that are ill requires particular consideration,depending on the circumstances. The level of understanding mustbe assessed, and where applicable, those that are responsible forthe vulnerable people must be consulted. There are often difficultdecisions to be made. Special help to aid understanding might begiven, but not to the point of condescension or social labelling.The assistance of an intermediary might be appropriate, such as atranslator, relative or carer. In some cases, those responsible forthe care of a vulnerable person may be in a position to giveconsent on their behalf. It is a good idea to consult at an earlystage with people experienced in these issues, and with thoseresponsible for vetting the ethical standards of the proposedresearch.Notwithstanding any agreement to take part in a research project,participants must have the right to terminate their participation atany time. There could be many reasons why a participant maywant to stop taking part in the research: there may be aspects ofthe research that they did not fully understand, unwantedimplications that appeared, discomfort or embarrassment, or just
too much bother. No reason need be given for the termination,and it should be accepted immediately, however inconvenient tothe researcher.Carrying out the researchPotential harm and gainEthical research is aimed at causing no harm and, if possible,producing some gain, not only in the wider field, but for theparticipants in the project. A prediction must be made by theresearcher about the potential of the chosen research methodsand their outcomes for causing harm or gain. The implications ofinvolving people in your research are not always obvious, so ifthere are issues about which you are not certain, you shouldconsult with experts in the field who have had more experience.What sort of precautions should be taken? Find out how you canavoid risk to participants by recognizing what the risks might be,and choosing methods that minimize these risks. For example,psychological harm can be caused by investigations into tragic oruncomfortable experiences, particularly if the subjects are stillvery much emotionally involved with the event. Assuming thatthere is good reason for making these investigations, there maybe ways to minimize the harm by consulting people who are notso immediately affected by the event. Also, participants might bereassured about the benefits to others of the research; perhapssimilar events might thus be prevented. It may help to have arelative or friend present to support the participant.Other types of harm to avoid are those arising not so much fromthe encounter with participants, but from the results of theinvestigation. Could what is revealed in any way be harmful to thereputation, dignity or privacy of the subjects? Could it in any wayalter the status quo to the disadvantage of the participants, e.g. byunjustifiably raising their expectations or by souring theirrelationships with other people?
Particular care must be taken when the researcher is working inan unfamiliar social situation, e.g. in an institution or amongstpeople of a different cultural or ethnic background. It may be thatbecause the mores are not well understood, offence can becommitted inadvertently. Language barriers can also lead tomisunderstandings, as can religious differences. The best way toreduce the risk of harm in these circumstances is to carefullyweigh up the areas of doubt to get advice and guidance onmeasures to avoid potential pitfalls. Being aware of the problemsgoes half-way to solving them!Figure 6.3 Particular care must be taken when the researcher isworking in an unfamiliar social situationOne way to ensure that individuals or groups of participants arenot implicated in any way in the findings of the research is todevise systems to ensure anonymity and confidentiality in thematerial accessible by other people. This relates not only to thepublished work, but also to your collected data that might beaccessed by others. There is more about this subject in the latersection on dissemination.Interviews and questionnairesWhen recording data, particularly from interviews and openquestions, there is a danger of simplifying transcripts, and in theprocess losing some of the meaning. By cleaning up, organizing,
ignoring vocal inflections, repetitions and asides, etc., you start toimpose your own view or interpretation. This is difficult to avoid, asthe grammar and punctuation of written text impose their ownrules which are different from those of verbal forms. Losingsubtleties of humour can misrepresent emotional tone andmeaning. Alldred and Gillies pointed out that speech is a ‘messy’form of communication, and by writing it down we tend to make anaccount ‘readable’ and interpret ‘what was meant’ (2002, pp. 159–61).It is easy to impose one’s own particular assumptions, e.g. ininterviews, especially when questioning people of differentbackground, culture or social status. Is the content of yourinterview based perhaps on white, Western assumptions, or otherassumptions inherent in your own cultural milieu? As an example,Alldred and Gillies related how the notion of personal decision-making and individualism which underpinned interview questionsabout the nature of parenthood and agency made little sense to acouple from Bangladesh, who were being asked to narratethemselves as Westernized modernist subjects (2002, pp. 155–7).Participant involvement: experiments,observations, groupsIf your research entails close communication between you, theresearcher, and the participants, the issues of getting involved andthe question of rapport are raised. How will those involvedunderstand your actions, and are these in balance with yourjudgement about your own practice? Your intentions for yourresearch might be to gain as much revealing information aspossible, and by ‘doing rapport’ or faking friendship you mightencourage the interviewee to open up. The intimacy betweenresearcher and respondent can resemble friendship. This raisesthe question: is it taken so far as to deceive in order ‘to encourageor persuade interviewees to explore and disclose experiences andemotions which – on reflection – they may have preferred to keepto themselves or even “not-to-know”’ (Duncombe and Jessop,2002, p. 120)?
Sensitive materialResearch into human situations, whether in the workplace, insocial settings, in care institutions or in education, can throw upinformation that is of a sensitive nature. This means that theinformation is such that, if revealed, it could do damage to theparticipants or to other people. Revelations about the treatment ofindividuals due to the actions of others or due to the workings ofan organization may call for action on the part of the researcherthat is outside the remit of the project. If, for example, a pupilcomplains to you about being bullied or an employee claims to bethe victim of discrimination, what attitude should you adopt abouttaking action to remedy the unfair situation? What should you do ifsomeone admits to cheating or an even more seriousmisdemeanour? Should you report the information to the relevantmanagement or authority, or should you keep it confidential?Sometimes, the researcher is seen by participants as someonewho could help or who has special knowledge or power toinfluence the situation.Every case must be judged individually, and careful thought mustbe given to the implications of divulging the information to anythird party. It may be possible to give advice to the participantabout whom to contact for help, such as a school tutor, tradeunion representative or ombudsman. It is not advisable to getpersonally involved as this could lead to unforeseen andunfortunate consequences that could not only cause harm to theparticipant and other people, but endanger the integrity of yourselfand the research project. Take advice from your supervisor orethics officer if the decisions are difficult.Honesty, deception and covert methodsAn ethically sound approach to research is based on the principleof honesty. This precludes any type of deception and the use ofcovert methods. However, it may be argued that some kind ofinformation that could be of benefit to society can only be gainedby these methods, because of obstruction by people ororganizations that are not willing to risk being subjected toscrutiny. Injustices might be brought to light that are otherwise
obscured by lack of information, such as discrimination, unfairworking practices or the neglect of duties. If the argument isbased on the principle of doing good without doing harm, it mustbe recognized that prediction of the outcomes of the research arespeculative. How can one be sure of the benign consequences ofthe actions? The risks involved are such as to make the use ofdeception and covert methods extremely questionable, and evenin some cases dangerous (one need only think of the risks run incases where investigative journalists have infiltrated a gang offootball hooligans and other criminal organizations in order touncover their covert operations).Using assistants or delegating tasksYou may be in the lucky situation where you can call onassistance to carry out some of the fieldwork, particularly if youare doing funded research. Your ethical responsibilities will be, ofcourse, delegated to your assistants, so you must be sure thatthey are aware of these. A thorough training in what they shoulddo (and not do) is essential so as to avoid problems arising fromignorance and confusion. Regular feedback from the assistants isessential to ensure standards are maintained.Social media researchThe popularity of social media makes it a rich source of researchopportunities, but these are faced with particular problems whenconsidering ethical standards. According to Moreno et al., in adetailed review of the current situation, the most commonconcerns can be explored by considering the use of observational,interactive, and survey/interview methods (Moreno et. al., 2013).The first consideration in all of these methods is whether the datacollected and analysed is in the public domain – and therefore isnot subject to privacy and anonymity restrictions, or whether it isprivate – and therefore subject to these. There are many formatsof social media: Twitter, YouTube, Facebook, LinkedIn, etc. eachwith their own methods of login in and sharing data, whether in theform of words, sounds or videos. How can you distinguish
between what is in the public domain, and what must beconsidered to be private?For observational research, if the material, such as a YouTubevideo, is publicly posted and available to any Internet user withoutany interaction with the person who posted it, the material is notprivate and not subject to ethical considerations. This is the casefor other public information concerning individuals on social mediawebsites such as Facebook, Twitter and LinkedIn, as long youensure that these individuals cannot be identified in any way, arenot minors, or that the information gained cannot place thesubjects at any kind of risk. It is relevant that the users of thesewebsites can choose to keep their information private if theychoose, so the logic follows that if they do not choose to do so,they are willing to share their information publicly (Moreno et al.,2013).Interactive research requires you to have contact with theparticipants providing the data and their consent for you to accessit. You can achieve this by requesting to be a ‘friend’ or ‘follower’of the participant, who will then be aware of making the materialthey uploaded available to you. The interaction in these cases isminimal. But, again, you need to be aware of the importance ofthe issues of anonymity and avoidance of doing harm.When doing survey or interview research, through which youintend to get answers to specific questions, it is necessary to getinformed consent from your participants, and parental consent ifthey are minors. There is a concern about the lack of face-to-facecontact, particularly with regard to the consent process andquestioning on sensitive matters, in that the reactions of thesubjects cannot be observed. However, this is also the case withmailed questionnaires. What is important is that participants caneasily contact you to raise queries or concerns online. Again,maintaining confidentiality is of the utmost importance. This is notalways simple. For example, direct quotations can often be tracedto their writer by online searches, and if researching a particularorganisation, personal information might lead to identificationwithin that organization (Moreno et al., 2013).
The Association of Internet Researchers produced detailedguidelines on ‘Ethical Decision-Making and Internet Research’ in2012, available at http://aoir.org/reports/ethics2.pdfIn Box 6.3 are two examples of how the ethical issues were dealtwith in two Internet research projects. Notice that both areconcerned with maintaining the anonymity of participants and howto deal with sensitive information.Box 6.3 ExamplesEthics when doing internet researchprojectsExploring anti-depressant medicationuse: using postings to an Internet medicaldiscussion boardJennifer Davis-Berman and Frances G. Pestello, 2018The key ethical issues that one should consider with this kind ofresearch, according to Annette Markham and colleagues (2012), areprivacy, the nature of human subjects, and text versus persons. Weconsidered these in thinking about the message board postings asdata. These issues are not novel to Internet data, but the nature of theInternet does require some additional thinking about how to addressthem in each research project. The message boards we used wereclearly labeled as public Internet spaces open to all and available forvarious uses, even though some posters may not think about thepublic nature of the forum when they are posting. To some extent, thisis true of all public spaces. Implicitly there is informed consentbecause one agrees to this as one logs on to the message boardwebsite to post. Although individuals post their thoughts andexperiences, we did not sample people, but rather conducted asystematic sampling of the posts on the forum. We did use somequotes from the data to explicate the themes that our analysisidentified, which might increase ethical concerns, but the posts had noidentifying information and they were not retained electronically by us(see also Sixsmith & Murray, 2001). This would make it difficult, if notimpossible, to connect any quote to a person.
ReferencesMarkham, A., Buchanan, E., & AoIR Ethics Working Committee.(2012). Ethical decision-making and internet research: 2012.Recommendations from the AoIR Ethics Working Committee.Retrieved from http://aoir.org/reports/ethics2.pdfSixsmith, J., & Murray, C. D. (2001). Ethical issues in the documentarydata analysis of Internet posts and archives. Qualitative HealthResearch, 11, 423–432. doi:http://dx.doi.org/10.1177/104973201129119109Researching gun control attitudes withweb surveysLacey N. Wallace, 2018As with any project using human participants, my project needed to beapproved by my university’s Institutional Review Board. To maintainappropriate ethical standards related to informed consent, I restrictedmy study to English-speaking adults able to give consent. Everyrespondent was asked to read an informed consent document beforecompleting the survey. An informed consent document explains thepurpose of the study, any potential risks or benefits, the respondent’srights as a participant, and provides contact information for theresearchers. Informed consent is necessary to ensure that aparticipant understands that participation is voluntary. For in-personstudies, respondents are often asked to sign an informed consentdocument to indicate that they understand the procedures, risks, andso on. As this survey was completed online, it was not feasible to askrespondents to physically sign the informed consent page. Instead,respondents were asked whether they understood the informedconsent document and agreed to participate in the study. If therespondent selected yes, the respondent was able to continue withthe web survey. If the respondent selected no, the survey ended.Exercise 6.3This exercise involves you looking up a Web page and considering theethical issues involved in a particularly controversial psychologicalexperiment carried out by psychologist Stanley Milgram at Yale Universityin the 1950s and 1960s. He wanted to see how far people would be willingto turn up the dial, if ordered to do so, on a machine that pretended to give
electrical shocks to people in the next room. He was surprised how manypeople were willing to go all the way, even though some broke down intears after hearing fake screams coming from the other room.The Web address iswww.wadsworth.com/psychology_d/templates/student_resources/0155060678_rathus/ps/ps01.html (sorry about the length!) based on Milgram(1963), with the permission of Alexandra Milgram. This is a detailedaccount of why and how the experiment was carried out, and there is a setof questions at the end on the ethical issues raised. I reproduce those thatare directly related to ethics.1. Milgram’s methods made the subjects extremely uncomfortable. Doyou believe that his research was ethical? Why or why not?2. Milgram undertook his research in the late 1950s and early 1960s,before ethics review committees were established. Thesecommittees weigh the value of proposed research against itspotential harm. If you were proposing the research to such acommittee, what arguments could you make in its favour? If youwere a member of such a committee, what arguments might youmake to block the research?3. Can you think of a way in which Milgram’s research might have beenconducted without deceiving the subjects?4. Milgram ‘debriefed’ subjects after they participated in the study. Thatis, he explained that they did not really shock anyone. Debriefing isintended to protect subjects from harm. If you had participated in thestudy, which would you have found more stressful: belief that youhad or had not shocked anyone? Why?Dealing with dataStoring and transmitting dataThe data that you have collected will often be sensitive, i.e. willcontain confidential details about people and/or organizations. It istherefore important to devise a storage system that is safe andonly accessible to you. Paper-based and audio data should belocked away, and computer databases should be protected by apassword. If it is necessary to transmit data, take measures toensure that the method of transmission is secure. Emails and filetransfers can be open to unauthorized access, so precautions
should be taken to use the securest transmission methodavailable.The Data Protection Act 2018 covers virtually all collections ofpersonal data in whatever form and at whatever scale in the UK. Itspells out the rights of the subjects and the responsibilities of thecompilers and holders of the data. You can search for a copy ofthis on the UK government website www.gov.uk/data-protectionand equivalent regulations on sites in other countries (seehttps://i-sight.com/resources/a-practical-guide-to-data-privacy-laws-by-country for a comprehensive overview with useful links).Analysing the data and getting resultsOne might assume that, once the necessary data have beenacquired in an ethical fashion, the analysis of those data is anethics-free activity. After all, you are only trying to make sense ofthe data and to reveal the hidden ‘truths’ by applying standardanalytical techniques that are known for their reliability. However,Doucet and Mauthner conclude that:Data analysis is an ethical issue because it exposes powerand privilege in relationships, decision-making aroundmaintaining or curbing relationships with research subjects,and the potential for profound relational violations. In arguingthat data analysis processes are theoretically infused we alsosuggest that data analysis methods are not neutraltechniques. Rather, they are methods that embrace bothmethodological and epistemological assumptions. (2002, p.139)They stress the importance of maintaining relationships withresearch subjects during the analysis stage, particularly those thatdo not ‘fit’ into your theoretical, epistemological and politicalframeworks. Specifically in qualitative research, the combinationof influences from your own personal life, your choice of theacademic texts that frame your research and analysis, and theinstitutional context will guide you towards a particular way of
‘seeing’ and ‘hearing’ respondents’ accounts (Doucet andMauthner, 2002).The way to gain clarity and accountability in the analysis stages is tobe as transparent as possible about your personal and theoreticalattitudes, which have a bearing on the analytical and interpretive partsof the research. This requires a high degree of reflexivity about one’sown cultural assumptions, theoretical stance and personal attitudes,and knowledge of other ways that the analysis could be approached.Checking data and draftsWhat are your responsibilities in relation to the data that you havecollected from respondents? To what extent should you allowparticipants to see your observation notes, interview transcripts,etc.? Oliver (2010, p. 63) draws the distinction between raw dataand data that have been subjected to some interpretation oranalysis. He suggests that it would be reasonable to allow arespondent to have a copy of the tape of their recorded interviewor their completed questionnaire (i.e. raw data), but not transcriptsor observational data, as these have been subjected to a level ofcoding, selection or analysis that renders them the intellectualproperty of the researcher. It is not practical to let respondentsread and edit large amounts of primary data owing to the delaysthis would cause, and as they are unlikely to have the necessaryskills to judge their validity and accuracy.It is normal practice to produce drafts of your work in order for youand others to check it for spelling and grammatical errors and forstructure and content. It is appropriate to pass the drafts on tocolleagues or supervisors for comment, with the proviso that thecontent is kept confidential, as at this stage it is not ready forpublication and dissemination. It is generally not appropriate,however, to allow sponsors to make comments on a draft becauseof the danger that they may demand changes to be made toconclusions that are contrary to their interests. This couldundermine the intellectual independence of the findings of thereport.
DisseminationIf you are doing a research degree, the first stage of disseminationis the submission of the thesis for examination. In some countries,the examination at PhD level is a very public affair. The candidatemust present the thesis to a panel of examiners, who question theexaminee in front of a large invited audience, members of whichcan also pose questions. Mercifully, this is not generally the casein the UK, where the viva is much more intimate. After yoursuccessful completion of a thesis or dissertation, there will bestandard procedures, depending on your university or college, formaking your work available to the public.You may wish to disseminate your work further by publishing theresults in the form of conference or journal papers, a website orother types of publication. As this process inevitably involvesreducing the length of the material, and perhaps changing thestyle of the writing for inclusion in professional journals ornewspapers, you must be careful that the publication remains trueto the original. Oversimplification, bias towards particular resultsor even sensationalization may result from targeting a particularreadership.In most cases, the intellectual ownership of sponsored researchremains with the researchers. Apart from any specific agreementsbetween the researchers and sponsors, dissemination is theconcern of the authors of the research report. Oliver (2010, p.121) suggests one situation where the sponsor may wish to seethe final report before dissemination. If the findings of the reportare contrary to the sponsor’s interests, or the sponsor disagreeswith the conclusions, seeing the report will enable them toformulate a statement in reply that can be disseminated at thesame time as the report. This will leave readers to make their ownjudgements.The question of authorship in team efforts needs to be addressed.Agreement should be reached between those that contributed tothe research and production of the report as to who should belisted as authors and in which order the names should appear.Acknowledgement should be made to others who contributed
substantially to the work, and to the sponsors. This may not be aneasy process in larger projects where many people have beeninvolved at various stages of the work and in various capacities(see Townend, 2000, pp. 92–6).Participants might request to be informed about the outcomes ofthe research. This raises certain questions about ownership of thefinal report. If the report has been commissioned by a sponsor itcould, depending on the contract, be their property and you mighthave no right to disseminate it to others without permission. If theresults of the research are published in the public domain,however, there is no reason why participants should not be sent acopy of the article or paper. In some cases, a short summary ofthe findings might suffice.Disposing of recordsWhen the data have been analysed and are no longer needed, asuitable time and method for disposal should be decided. Ideally,the matter will have been agreed with participants as a part oftheir informed consent, so the decision will have been made muchearlier. However, this issue might not be as easy as it appears.There are several reasons why you might want to keepconfidential data for longer than needed for your immediateproject. You may want to keep the data in case there are anyqueries about them at a later date, or if you expect to use them asthe basis of further publications. Perhaps you only used a part ofthe collected data, and wish to use some more of them for anotherproject. You might even know another researcher who could makegood use of the data you have collected.There are several precautions that can be taken when faced withthese issues. If the conditions about disposal of data were agreedwith participants, and you wish to change the conditions, youshould contact them again to ask if they object to your newproposals. This might cause practical problems of finding theparticipants again, in which case a pragmatic solution should besought that protects their interests. One basic policy is to ensurethat all the data are anonymous and non-attributable. This can be
done by removing all labels and titles that could lead toidentification.When destroying data, make sure that they are disposed of insuch a way as to be completely indecipherable. This might entailshredding documents and discs, wiping drives, erasing audio orvideo files.Ethics policies, permissions andcommittees in organizationsAll organizations that are involved in research with humanparticipants will have set up a code of practice for theirresearchers. To see typical examples of these types of guidelines,you can refer to the Web page produced by the BritishEducational Research Association(www.bera.ac.uk/publication/ethical-guidelines-for-educational-research-2018), and the British Sociological Associationstatement of ethical practice (https://www.britsoc.co.uk/ethics),and the British Psychological Society (www.bps.org.uk/news-and-policy/bps-code-ethics-and-conduct). For more guidance anduseful links go to www.brookes.ac.uk/research/research-ethics/policiesYour university will certainly have set up its own code of practice.As an indication of what you might expect, Box 6.5 gives anabbreviated version of the Oxford Brookes University code. Do,though, refer to your organization’s code when preparing yourown ethical policy.Box 6.4 ExampleEthical standards for research involvinghuman participants (Oxford BrookesUniversity, 2016)
Code of practice1. IntroductionThe University adheres to the principles of research ethics as laid outby the ESRC (2015) guidelines and which comprise the following:Research participants should take part voluntarily, free from anycoercion or undue influence, and their rights, dignity and (whenpossible) autonomy should be respected and appropriatelyprotected.Research should be worthwhile and provide value thatoutweighs any risk or harm.Researchers should aim to maximise the benefit of the researchand minimise potential risk of harm to participants andresearchers. All potential risk and harm should be mitigated byrobust precautions.Research staff and participants should be given appropriateinformation about the purpose, methods and intended uses ofthe research, what their participation in the research entails andwhat risks and benefits, if any, are involved.Individual research participant and group preferences regardinganonymity should be respected and participant requirementsconcerning the confidential nature of information and personaldata should be respected.Research should be designed, reviewed and undertaken toensure recognised standards of integrity are met, and qualityand transparency are assured.The independence of research should be clear, and any conflictsof interest or partiality should be explicit.The University requires that the integrity of any research depends notonly on its rigour but also on its ethical adequacy. The Universityexpects that research should contribute to knowledge development.Research involving human participants is undertaken by manydifferent disciplines and conducted in a broad range of settings andinstitutions. While some issues are specific to professional groups, allresearch should be guided by a set of fundamental ethical principlesto ensure the protection of human participants and their personaldata. Good ethical research practice is the responsibility of theresearcher and or the research team and or the supervisory team.The underpinning principle of research conducted in or through OxfordBrookes University is to do no harm to research participants. TheUniversity supports a breadth of research and approaches andacknowledges that this range of research may result in an array ofethical issues.
The following standards have been developed to guide staff andstudents undertaking research involving human participants. They areintended to cover general principles, but they may not address allsituations and the researcher should seek further advice from theirFaculty’s Research Ethics Officer, the University Research EthicsCommittee and the relevant professional code of practice for researchethics as appropriate.This code of practice acknowledges key regulatory frameworks suchas the latest versions of:Data Protection ActEquality ActFreedom of Information ActHuman Rights ActHuman Tissue ActThe Mental Capacity ActResearch Governance Framework for Health and Social CareSafeguarding/DBS requirementsSafeguarding Vulnerable Groups ActLegislation relating to Children including The United NationsConvention on the Rights of the Child2. No research should cause harmNo research undertaken should cause harm to participants,researchers or other persons directly or indirectly involved in theresearch1. A judgement needs to be made as to whether a particularresearch practice is likely to affect the well-being of participantsand any potential risks to participants which might arise in thecourse of the research should be identified.2. Research approaches must be justified, explaining whyalternative approaches cannot be used, for example in cases ofpotentially intrusive research.3. The potential benefits of the research to participants, the widercommunity and/or society must be clearly stated.4. Any cultural, religious, gender or other differences in a researchpopulation should be sensitively and appropriately handled byresearchers at all stages. Relevant subject specific requirementsand norms of research practice should be sought out andadhered to.3. Honesty
Honesty in the planning, conducting, analysing and reporting ofresearch is required. Honesty should be central to the relationshipbetween researcher, participant and institutional representatives1. The deception of participants should be avoided. If necessarydeception must be justified by the researchers and the reasonsshould be explained to participants after the study.2. Covert research may be appropriate within certain contexts andwithin certain subject areas but must be justified. However the‘do no harm’ principle must be adhered to.3. Any conflicts of interest must be declared to the University,research participants and in any dissemination of findings.4. CoercionParticipants should be free from coercion and should not be pressuredto participate in a study1. Incentives for participation should be commensurate with thetasks involved.2. Compensation for loss of income should not be consideredinducements. Reimbursement of participants’ expenses, forexample travel, can be provided.3. Risks involved in participation should be acceptable toparticipants, even in the absence of inducement.4. Participants must be free to withdraw from the study at any time,without any repercussions to the participant.5. Confidentiality and anonymityParticipants’ confidentiality and anonymity should be maintained as acore principle1. Researchers should adhere to the University’s Code of Practicefor Academic Integrity, including procedures for investigatingallegations of misconduct in research.2. Researchers and other collaborators should deal with all dataobtained through their project in such a manner so as not tocompromise the personal dignity of the participant orcompromise the participant’s right to privacy, through all thecollection, storage, analysis and disposal stages of the research.3. All information obtained in the course of a research projectshould be considered privileged information and should under nocircumstances be publicly disclosed in a fashion that wouldidentify any individual or organisation except when required bylaw, or with the express consent of the participant.
4. When personal identifiers are used in a study, researchersshould explain why this is necessary and how confidentialitywould be protected. In cases where participants’ anonymity maybe at risk, the participants will be informed at the outset of theresearch.6. ConsentParticipants in a research study are usually required to give theirinformed consent before participating6.1 Informed consentParticipants should understand the purpose and nature of the study,what participation in the study requires, and what benefits areintended to result from the study1. Informed consent, should usually be recorded from anyparticipant who is able to give such consent, either by impliedconsent (see point c below) or explicitly.2. It is the researcher’s responsibility to seek ongoing consentduring the course of a study, as appropriate.3. Consent may be implied by the completion and return of surveyquestionnaires, removing the need for written consent.4. Individual consent is not always required for some types ofresearch activities, for example studies involving observation ofpublic behaviour in public spaces.6.2 Third party consentWhere third parties are affected by the research, consent should beobtained1. When third parties, for example spouses, teachers or healthcareprofessionals, are directly affected by the research and areinvolved in the care, education or treatment of the potentialparticipants, consent should also be obtained from them.6.3 Gatekeeper consent1. Certain situations may necessitate the granting of permission byeducational institutions, business organisations, or hospitals, toconduct the research. Where this is required permission shouldbe sought prior to the start of the research and any relevantpolicies or guidelines should be followed.
6.4 Vulnerable participant consentThe consent of vulnerable participants or their representatives’consent should be sought by researchers1. In cases where people are unable to comprehend theimplications of research, assent to participate may come from arepresentative, such as a legal guardian, immediate relative orcarer with the appropriate authority to do so.2. Consent witnessed by an appropriate person is required forvulnerable participants who have intellectual difficulties orlinguistic differences limiting speech or understanding, but whoare deemed capable of giving consent.6.5 Research involving children and their consentTo the extent that it is feasible, which will vary with age, the willingconsent of participants who are children should also be sought.Generally, children over age 16 may be assumed to be capable ofgiving informed consent, but this will vary depending on the nature ofresearch. Adherence to the UNICEF guidelines for Ethical ResearchInvolving Children should be followed (http://childethics.com/)1. If children are involved in a research study, then parents or otherlegal guardians must be informed and may be required to givetheir consent for inclusion of the child in the study.2. In the case of research within an educational setting such as aschool, specific school policies or procedures should befollowed.7. Dependent relationshipsWhere relationships may be characterised by inequalities of power orstatus, the impact of this relationship on all parties needs to beconsidered1. Research involving participants who are in a potentiallydependent relationship with the researcher (e.g. students,employees and patients) requires careful consideration,justification and mitigation where possible.8. Data protectionThe processing of research information containing personal data mustcomply with the Data Protection Act 1998 principles listed below
1. The University is registered as a data controller with theInformation Commissioners Office (ICO).There are eight good practice guidelines which must be adheredto when processing personal data:1. Personal data shall be processed fairly and lawfully.2. Personal data shall be obtained only for one or morespecified and lawful purposes.3. Personal data shall be adequate, relevant and notexcessive.4. Personal data shall be accurate and, where necessary,kept up to date.5. Personal data processed for any purpose or purposes shallnot be kept for longer than is necessary.6. Personal data shall be processed in accordance with therights of data subjects under this Act.7. Appropriate technical and organisational measures shall betaken to protect personal data.8. Personal data shall not be transferred to a country orterritory outside the European Economic Area withoutadequate protections.2. Additional care must be taken when processing sensitivepersonal data which may contain the following information1. racial or ethnic origin2. political opinions3. religious beliefs or other beliefs of a similar nature4. membership of a trade union5. physical or mental health or condition6. sexual life7. commission or alleged commission of any offence8. any proceedings for any offence committed or alleged tohave been committed.3. In the event that personal data have been lost, stolen orcompromised in any way, the incident must be reported at thefirst opportunity to the information compliance officer.Further information on data protection is available:www.legislation.gov.uk/ukpga/1998/29/contents9. Digital, Electronic and Social Media DataThis refers to research involving the use of online surveys,questionnaires, use of social media (e.g. Facebook, on-line groups,chat rooms, Skype etc.) for recruitment, data collection or researchwithin this area. Continuing technological advances can introduce
additional and nonobvious complexities in adhering to ethicalprinciples1. Members of the University who carry out research using thesemethods should ensure that they are familiar with currentdebates on the ethics of Internet-mediated research.2. Gaining informed consent, negotiating access, assessing theboundaries between the private and the public and ensuring thesecurity of data transmissions may be problematic, however theethical principles referred to in this document apply as much forthese forms of research as they do for all other forms ofresearch.3. Adequate controls must be put in place to preserve theanonymity and/or protect the privacy of participants, appropriateto the technology used and the type of data which is to becollected.4. Researchers are reminded to read the terms and conditions ofany such service that they use to ensure that the service is fit forpurpose. UK Legislation such as The Regulation of InvestigatoryPowers Act (RIPA) may require lawful disclosure ofcommunication data. In such circumstances this must beescalated to the Information Compliance Officer.5. The type of consent obtained (e.g. documented using a separateform, in hard or soft copy, evidenced via other ‘implied’ means)should be proportional to the risk of the research to participants.6. The distinction between public and private domains should beconsidered from the point of view of the participant and notdefined simply in terms of ability of the researcher to access thesite, material or participant. Care must be taken in both consentand data management processes to respect individuals and theirprivacy.7. Further useful guidance can be obtained from: The BritishPsychological Society www.bps.org.uk and the Association ofInternet Research (AoIR) http://aoir.org/ethics.10. Research involving sensitive and or illegalsubjectsThe University is committed to providing an environment in whichstudents and staff are able to engage in debate and research acrossall subject areas within legal boundaries1. Oxford Brookes University acknowledges the need to be vigilantin identifying and managing, where possible, research involvingsensitive subjects such as but not limited to terrorism, abuse andillegal substances. In relation to terrorism the University aims to
prevent individuals from being drawn into terrorism in so far as ispossible.2. The researcher should be aware that researching certain subjectareas carries with it the potential for an individual being flaggedby the security or other public protection services and thisflagging can be irreversible.3. For information on security sensitive material and activity pleaserefer to the University’s guidelines.11. Research in public contexts and with groupsAwareness should be demonstrated when research is conducted inpublic spaces and or with groups1. In certain research contexts obtaining consent from eachindividual may be impractical or unfeasible or indeed detrimentalto the research being undertaken. In such instances the groupsshould, where possible, be informed that the research is takingplace and no individual identifying detail should be included inany subsequent research report. Sensitivity should be shown tolocal group cultural norms to avoid the perception of the invasionof privacy despite being situated in a public space.2. In participatory research care should be taken to ensure thatparticipants within a group are made aware that they are beingobserved for research purposes.12. Personal safety and responsibilities of theresearcherThe researcher and or research team and or supervisory team shallensure that the research to be undertaken creates no more risk to theresearcher than that of normal every day life1. Where it is considered that the risks are more than those inevery day life a justification for those risks and suitable mitigationof the risks should be provided.2. Where the research involves the use of specialised equipmentthe researcher, research team and/or supervisory team willensure that there is adequate knowledge and training in the safeoperation of that equipment.3. The researcher and/or research team and/or supervisory teamshall be responsible for ensuring that appropriate insurances arein place through communicating to the University any overseastravel plans for research purposes in addition to the nature oftheir research.
13. Communication with participantsPotential participants should receive clearly communicatedinformation from the researcher in advance1. Most research should be explained on an information sheetwritten in clear language that is easily understood by thepotential research participant.2. The information sheet should set out: the purpose of theinvestigation; the procedures; the risks (including psychologicaldistress); the benefits, or absence of them, to the individual or toothers in the future or to society; a statement that individualsmay decline to participate and also will be free to withdrawwithout giving a reason; the level of feedback to be offered; thetime required and an invitation to ask questions.3. The information sheet should also provide contact details of theFaculty’s Research Ethics Officer or Chair of UREC so thatparticipants may report any concerns about the conduct of thestudy.4. Participants should be given, under normal circumstances, aminimum of 48 hours to study the information sheet, and consultrelevant parties where necessary.5. The information sheet and the consent materials (seeAppendices for templates) should form part of the application forethics approval.14. Dissemination of research findingsResearchers have a duty to disseminate their research findings to allappropriate parties1. Participants and relevant stakeholders will usually be offeredaccess to a summary of the research findings whereappropriate.2. Reports to the public should be clear and understandable andaccurately reflect the outcome of the study.3. Research outputs should wherever possible be disseminatedwidely and openly to maximise their value.Approved by the University Research Ethics Committee on: 9 March2016Approved by the University Research and Knowledge ExchangeCommittee on: 24 May 2016Approved by Academic Board on: 13 July 2016
Source: Oxford Brookes University, Ethical Standards for ResearchInvolving Human Participants: Code of Practice, 2016Ethics committeesThe role of ethics committees is to oversee the research carriedout in their organizations in relation to ethical issues. It is they whoformulate the research ethics code of conduct and monitor itsapplication in the research carried out by members of theirorganizations. Your university or other institution will probablyhave a system that makes it possible for its research committee todo its job. This, inevitably, involves filling in forms.In order to help researchers decide whether their researchproposals should be submitted for approval by the schoolresearch ethics officer or committee, your university will alsoprovide an ethics review checklist that can be adapted for theparticular circumstances of the research in each subject area.Obtain one of these to check whether your activities willnecessitate ethics approval. Examples of questions you mightexpect in the checklist are in the ‘next steps’ section later in thischapter.If you answer ‘yes’ to any of the questions, you will have to submityour proposal using the appropriate ethics approval applicationform to your school research ethics officer. It is advisable, whenany doubt arises in relation to the above, to always forward yourproposal to the school research ethics officer for him/her to review.The forms will ask a series of questions on a range of ethicalissues and will require detailed information on your researchmethods. You will have to justify why you have selected theparticular methods proposed and to explain how you will deal withall the attendant ethical issues. Checklist 6.5 gives a guide, takenfrom the Oxford Brookes University consent form guidelines, towhat you will probably need to include.Checklist 6.5 Issues for the ethics consent formTitle of the study
Purpose of the studyWhy participant was selectedDescription of procedures, purpose, length of time required andhow participant will be involvedDiscomforts, inconveniences expectedRisks, if anyBenefits, if anyWithholding standard care/treatment or an alternative, if anyCompensation to be expected, if anyHow confidentiality, anonymity and privacy will be maintainedRight of participant to refuse to participate or withdraw at anytime for any reasonSources for information and assurances that researcher willprovide further and ongoing information (e.g. name and contactphone number of the researcher)Signature of the researcher and the participant or theparticipant’s representativeSignature of the witnesses where appropriateSource: Oxford Brookes UniversityThe research committee will then review your forms, and all beingwell, give you clearance to continue with the research. If there areissues that are not clear or not fully dealt with, the committee willwant to have clarification before giving consent.If you are embarking on a research degree, this processcommonly occurs immediately after you have submitted yourresearch proposal and are registered to do the degree. Similarly, ifyou are doing a master’s degree dissertation, seek ethicsapproval if necessary after your research proposal has beenapproved. For contract research, answering the ethical issues willnormally be part of preparing the initial research proposal. In anycase, the ethics consent should be obtained before any fieldworkor contact with participants has begun. It is up to you to ensurethat you have clearance from the relevant bodies, including otherorganizations if they are involved (e.g. National Health Service,school, company) before you begin the research work.
When applying principles to any practical activity, there are oftenconflicts and dilemmas; the application of ethical standards inresearch activities is no exception. You will have to be able to arguethat you have taken reasonable, best possible measures to conform tohigh standards of research ethics.Figure 6.4 Forms of behaviour and etiquette desirable in thecivilized pursuit of knowledgeBeyond the moral obligations of research, there are forms ofbehaviour and etiquette desirable in the civilized pursuit ofknowledge which should be observed when communicating withpeople. A considerate and courteous attitude to people will alsohelp to improve their readiness to assist you and provide you withthe information you require. Remember that you are relying ontheir cooperation and generosity to make your research possible,and this should be acknowledged in your attitude and behaviour.You should devise a systematic method of making requests forinformation, interviews, visits, etc., together with one for
confirmation of appointments, letters of thanks, and some follow-up and feedback where appropriate. More detailed considerationhas been given in Chapter 5, where research techniques andmethods were discussed.ConclusionsThe value of research depends as much on its ethical veracity ason the novelty of its discoveries. How can we believe in the resultsof a research project if we doubt the honesty of the researchersand the integrity of the research methods used? It is easy to cheatand take shortcuts, but is it worth it? The penalties resulting fromdiscovery are stiff and humiliating. It is also easy to follow thesimple guidelines on citation which avoid violations of intellectualproperty laws, and which also enhance your status as being wellread and well informed about the most important thinkers in yoursubject.To treat participants in your research with respect and dueconsideration is a basic tenet of civilized behaviour. Officialconcern about the ethical issues in research at any level thatinvolves human subjects is growing. This means that there is agreater need to analyse the methods used in research in detailand to account for the decisions made when seeking officialapproval. Admittedly, the issues can become quite complicated,with no clear-cut solutions. It is therefore important that youconsult with others, especially advisers appointed for thatpurpose.Miller and Bell suggest that keeping a constant record of decisionsmade is a good safeguard against sloppy thinking and inadvertentoverlooking ethical issues:Using a research diary to document access routes anddecisions made throughout the research process is onepractical way of developing an ethics checklist. This practiceof regular reflection helps ensure that ethical and
methodological considerations are continually reassessed.(2002, p. 67)This chapter has presented a brief overview of the principal ethicalissues in research, and what actions you need to take to ensurethat your research will be ethically sound and that you haveobtained the necessary clearances from relevant authorities.The next steps: plan your code ofethicsSound ethical procedures are the basis of good research. Theextent to which ethical issues impinge on your work depends onhow and how much you will be working with people (and perhapsanimals?). Beyond the basic ethical matters of honesty etc. youwill have to make an assessment of how you will tackle theproblems of fair dealing, gaining consents, providing information,privacy and anonymity, etc. Obtain the guidelines and an ethicsapplication form from your ethics department to see whatquestions you will have to answer. You may be surprised at theextent of information you will need to give!This is bound to raise a lot of issues and questions, some of whichyou will not be able to resolve yet. You will need to record whatethical issues are presented by your research, to be able todiscuss how you have come to your conclusions about ways ofovercoming problems, and to list issues that you think importantbut have not been resolved. This will provide a good basis fordiscussion with your supervisors and ethics advisers to help youcompile a set of priorities for further investigation and clarification.The aims of this section are for you:to consider the ethical aspects of the projectto find solutions to ethical problemsto consult with others on unresolved issuesto be in a position to apply for approval (if necessary) fromthe ethics committee or other responsible body.
Checklist of activities that ensure ethical standardsBy taking simple precautions you can ensure that your researchcomplies with every aspect of the expected ethical standards. Byfollowing the steps below, you will check that you have considered themost important factors. There are a lot of things to consider if yourresearch is going to involve human participants and you have tosubmit your planned methods and explain your decisions to the ethicscommittee for their approval.Step 1: Check your methods of citationIn order to make sure that you cannot be accused of plagiarism, checkthat you have not used other people’s ideas or writing in such a wayas to deceive the reader about the real author of the work. If you haveused the ideas and words of others, which is in no way reprehensible,make sure that you have rigorously followed a system of citation andreference. Check that your citations make it easy for the reader toidentify the relevant section of text, and that the reference list isconsistent in its order of information, punctuation, use of italics,inverted commas, etc. Remember that even if you do not know theauthor of the work (which is often the case on websites and innewspaper articles) you should still acknowledge the source so as notto pass it off as your own work.Step 2: Clarity of data presentation,analysis, outcomes and conclusionsThis is about being open and honest about what you do, what you findout and how you interpret it. This is all rather in the future, as yourwork is still at the beginning. But it is obviously a big issue and willpermeate the whole of your work. It requires you to cultivate an openand honest attitude, especially when things do not turn out asexpected. There is no shame in finding the unexpected; in fact this iswhat research is about. Even if your theories are proved wrong, yourassumptions are without ground and your findings are inconsistent,there is always something to be learned for you and others. At leastthey will not have to make the same mistakes as you to learn thelesson. Check that you have used non-sexist, non-ageist, non-cultural,etc. terminology throughout your writing.Step 3: What ethical issues are involved?
Will your research involve human participants? If not, then you canread through the following with interest rather than to find out what todo. If it does involve human participants, first ensure that yourintended actions comply generally with all the standard ethical issuesas discussed above. Here is a short checklist:Your relationship to participantsHow you choose participantsHow you inform participants and what the information consists ofAssurance of the right to terminateGuarantee of confidentiality and anonymityGaining permissions from organizationsImpartiality in data collectionImpartiality in data analysisManagement of sensitive informationFreedom from deception and cover methodsAdequate training of assistantsCareful management of data and eventual disposalFeedback to participantsForms of dissemination that ensure no harm is doneResponsibilities to fundersThen go through the following steps to make sure that you pick up allaspects that need special attention.Step 4: Vulnerable participantsWill the study involve participants who are particularly vulnerable orunable to give informed consent (e.g. children, unconscious patients,people under 18 or with learning difficulties, difficulties inunderstanding or communicating)? You will have to take specialmeasures to ensure that those responsible for them are fully informedand give permission where appropriate on their behalf, and that theparticipants do not get confused or harmed in any way.Step 5: ChildrenIf the participants are school children, will you be able to obtainpermission to involve children under 16 from the school or parent? Ifnot, what are the reasons? You will need to justify your approach. Inthe UK, any person employed to work with children needs to get anofficial character clearance. Find out what the situation will be withregard to your intended work.
Step 6: Medical proceduresWill the research involve medical procedures? You will need to bequalified to carry these out, and they should be approved byknowledgeable experts. A rigorous risk assessment will have to becarried out. Similar precautions are required if drugs, placebos orother substances (e.g. food substances, vitamins) are to beadministered to the study participants, or blood or tissue samples areto be obtained. In all cases, the proposed procedures will have to bepainstakingly explained and evaluated.Step 7: Pain or discomfort or potentialharmIs pain or more than mild discomfort likely to result from the study?There will have to be good reasons why this should be so, and howthe level of pain or discomfort can be predicted and monitored. What ifthe study involves invasive, intrusive or potentially harmful proceduresof any kind (e.g. manipulation, aromatherapy oils, exercise routines,splints, specialized equipment)? Again, you will have to argue for thebenefits of this type of research outweighing the possible harm, andthe lack of alternative, more benign, risk-free methods. This is equallynecessary if it is predicted that the research could inducepsychological stress or anxiety, cause harm or have negativeconsequences for the participants (beyond the risks encountered intheir normal lifestyles). Prolonged or repetitive testing could also leadto fatigue and psychological or even physical harm.Step 8: Financial inducementsWill financial inducements (other than reasonable expenses andcompensation for time) be offered to participants? The worry is thatthe results could be affected by a form of bribe that puts theparticipant under pressure to provide the desired answers. For mostpoor postgraduate students, this issue is unlikely to be relevant!Step 9: DeceptionWill the study involve deception of participants, their carers,dependants, support networks or others, in any way (e.g. omitting orgiving false information about the researcher’s background, reason forstudy or anticipated outcomes)? This is usually not permissible, butperhaps, in certain circumstances, covert methods are the only way to
get unsullied data. This is a really problematic issue, as one of themost important principles of ethical research is that participants arefully informed about the research as far as it affects them. You willhave to have very good reasons why deception is necessary, andexamine what possible effects may result when the truth is revealed.Step 10: Right to remain anonymousAre there problems with the participants’ right to remain anonymous,or to have the information they give not be identifiable as theirs? Thiscould possibly lead to embarrassment and much worse (in somecases even threat to life), so this right should not be abused. Ifanonymity cannot be assured, then the participant must be madeaware of this before agreeing to take part in the research.Step 11: Right to withdrawThe right to withdraw from the study at any time is a basic condition ofethical research. Will it be withheld, or not made explicit? It may bevery inconvenient for the researcher if participants withdraw withoutwarning, but this should be the prerogative of voluntary participation.Why should you not allow this?Step 12: Access to groupsWill the study require access to groups or individuals because of theirmembership of a group, organization, place of study or dwelling place(e.g. university students, self-help group, nursing home, police force,National Health Service)? If so, several issues will need to beaddressed, such as gaining permission from the organizations andcomplying with their regulations. You will even perhaps need to getethics clearance from them.Step 13: Confidential informationDo you require access to confidential information? This could beperhaps health records, prison or conviction records, even school oruniversity grades. Obviously, because such records are confidential,they are only allowed to be given to certain authorized people ororganizations. How will you get access to them, and how will youensure that they remain confidential? These are questions you willcertainly have to answer.
Step 14: Special circumstancesAre there any other special circumstances? This is a ‘catch-all’question. Consider if there is any situation in which the ethical rulesmight be transgressed. Research is a very complex business, withdecisions to be made that cannot be clear-cut. Become aware of theshades of interpretation that can be made in human situations, andthe network of responsibilities that are incurred when doing researchthat impinges in some way or other on fellow human beings. Constantalertness and sensitivity are required to detect any possible problems;it helps to always consult with others, as they might see matters quitedifferently from you.The questions in the above, from step 2 onwards, are compiled fromlists provided by the ethics officers in a range of schools in OxfordBrookes University. The students are instructed that if they answer‘yes’ to any of them, they will have to fill in an ethics application formfor approval by the University Ethics Committee. It is likely that youwill have a similar system at your university or college. Check it out!Consolidation and assessmentArrange a tutorial with your tutor or supervisor. Consider theissues of ethics and privacy, and describe any aspects of themwhich you think will have a particular bearing on the type ofresearch you are proposing. You may find this checklist useful(Burnett, Sheehy and Digby, 1994, p. 15):Whose interests are involved?Is there any risk of damaging or embarrassing participants?What issues of confidentiality are involved?What issues of law and professional conduct need to bechecked out?What are the effects of this research and/or publication on theparticipants?What are the effects of this research and/or publication onother interested parties?Further reading
Although ethical behaviour should underlie all academic work, it isin the social sciences (as well as medicine etc.) that the reallydifficult issues arise. Researching people and society raises manyethical questions that are discussed in the books below. The firstset of books is aimed generally at student and professionalresearchers, and the second set contains examples of morespecialized books – though the issues remain much the same forwhoever is doing research involving human participants.Oliver, P. (2010) The Student’s Guide to Research Ethics,2nd edn. Maidenhead: Open University Press.This is an excellent review of the subject, going into detail on allaspects of ethics in research, and providing useful examples ofsituations where ethical questions are raised. It demonstrates thatthere are not always simple answers to these questions, butsuggests precautions that can be taken to avoid transgressions.Laine, M. de (2000) Fieldwork, Participation and Practice:Ethics and Dilemmas in Qualitative Research. London:Sage.The main purposes of this book are to promote an understandingof the harmful possibilities of fieldwork; and to provide ways ofdealing with ethical problems and dilemmas. Examples of actualfieldwork are provided that address ethical problems anddilemmas, and show ways of dealing with them.Mauthner, M., Birch, M., Jessop, J. and Miller, T. (eds)(2012) Ethics in Qualitative Research, 2nd edn. London:Sage.This book explores ethical issues in research from a range ofangles, including access and informed consent, negotiatingparticipation, rapport, the intentions of feminist research,epistemology and data analysis, tensions between being aprofessional researcher and a ‘caring’ professional. The bookincludes practical guidelines to aid ethical decision making rootedin feminist ethics of care.
Mertens, D. and Ginsberg, P. (2009) The Handbook ofSocial Research Ethics, 5th edn. Thousand Oaks, CA:Sage.With almost 700 pages, this is a comprehensive volume that offersa deeper understanding of the history, theory, philosophy andimplementation of applied social research ethics. It addresses theethical issues that arise in the theory and practice of researchwithin the technologically advancing and culturally complex worldin which we live.Geraldi, O. (ed.) (2000) Danger in the Field: Ethics andRisk in Social Research. London: Routledge.Read this if you are going into situations that might be ethicallyhazardous.Burton, D. (ed.) (2001) Research Training for SocialScientists. London: Sage.See Chapters 7–9 for good advice on the ethics when doingresearch.Woodfield K. (ed.) (2018) The Ethics of Online Research(Advances in Research Ethics and Integrity), Volume 2.Bingley: Emerald Publishing.A wide-ranging discussion on many aspects of ethics when doingonline research by a number of experts in the field.Internet Research Ethics (2016) Stanford Encyclopediaof Philosophy.https://plato.stanford.edu/entries/ethics-internet-researchA really detailed examination with lots of examples and a hugefurther reading list.
There are also books about ethics that specialize in certain fields.Here are some examples. You could search out some in yoursubject perhaps.Bryson, B. (2002) The Penguin Dictionary of TroublesomeWords, 3rd edn. Harmondsworth: Penguin.Whitbeck, C. (2011) Ethics in Engineering Practice andResearch, 2nd edn. Cambridge: Cambridge UniversityPress.Graue, M.E. (1998) Studying Children in Context: Theories,Methods, and Ethics. London: Sage.Burgess, R.G. (ed.) (1989) The Ethics of EducationalResearch. London: Falmer.Rosnow, R. L. (1997) People Studying People: Artefactsand Ethics in Behavioural Research. New York: Freeman.Alderson, P. (2011) The Ethics of Research with Childrenand Young People. A Practical Handbook. London: Sage.
7 Research Methods: Choosing theBest Methods for Your ProjectAimsTo show the range of research methods available to theresearcher in order that a choice can be made about suitablemethods for the research project for inclusion in the researchproposalTo describe the nature of different research methods used forcollecting quantitative and qualitative dataTo indicate appropriate methods of analysing dataIntroductionAs this book is focused on helping you to formulate your researchproject in the form of a research proposal, but not on actuallygoing into the detail of carrying the project out, there is not spacehere to provide exhaustive instructions on how to implement eachresearch method. So this chapter provides you with the essentialinformation on the applicability of a wide range of data collectionand analytical methods, depending on your subject of study.As a result, this chapter is rather long, but you really do not haveto read it all from beginning to end. The intention of this chapter isto provide you with a very brief description of commonly usedresearch methods so that you have a basic idea of the optionsopen to you when you set about planning your research. Anessential ingredient of your research proposal will be to suggestappropriate methods by which you will carry out your research, soa prior understanding of the choice of methods open to you willhelp you to decide which might be pertinent to your form of study.Although the number of methods available for research into thenatural sciences, the social sciences and the humanities isenormous, those that could reasonably be explained in this
chapter is limited, so only a general indication can be given ofsome of the principal methods. Definitely read the first section ondata collection and analysis, and then browse through the rest toselect those methods that are relevant to your project. You canuse the information in your proposal. Later, when you come to dothe research, you will have to delve deeper into the selectedmethods for data collection and analysis by referring to morespecialist books on research methods, of which there are amultitude!It would, at first glance, be easy to cover the subject by neatlydividing it under two headings: methods of data collection, andmethods of data analysis. The first part would be a straightforwardsummary of how to collect data from different sources usingdifferent methods, and the second would be a description oftechniques of analysis, both quantitative and qualitative. However,it soon becomes evident that both of the activities of datacollection and data analysis tend to be inextricably bound up withthe research strategies – historical, survey, case study, etc. – andcannot easily be discussed without some reference to these. Italso becomes obvious that there are often strong links betweenthe type of data collected and the type of analysis appropriate forthem. Even the clear distinction between data collection and dataanalysis can become problematic, as in some cases datacollection, collation and analysis go hand in hand, and in othersfurther valuable data are produced by the process of analysis.So, as an introduction, we will first look briefly at the threeperspectives from which one can review research methods,examine the characteristics that make them distinctive, anddescribe how they interact. The three perspectives are: datacollection and analysis, quantitative, qualitative and mixedmethods research, and research strategies.Data collection and analysisOnce the research problem has been formulated, it shouldbecome evident what kind of data will be required to study theproblem, and also what kind of analysis will be appropriate to
analyse the data. The reasons for choosing particular datacollection and analysis methods are always determined by thenature of what you want to find out, the particular characteristicsof your research problem and the specific sources of information.In fact, it will often be appropriate to decide first on the type ofanalysis – quantitative or qualitative – that will be required toinvestigate your research problem, and then on the type of datathat needs to be collected in order to make that analysis.Note, however, that the categories of data collection and analysisare not always as distinct as they might at first appear. They can,depending on the particular research aims, be closely interlinked.On the one hand, for example in exploratory research, the datamay be continuously analysed as they are collected, the analysisgiving clues as to the most fruitful area of further data collectionand subsequent analysis. On the other hand, when a particularphenomenon is investigated according to a specific predeterminedmethodology, it might not even be possible to begin the analysisuntil all the relevant data have been collected.When considering what data you might require, consider carefullythe sources, the availability and the possible methods of collectingthe data. When considering analysis, think about the tools,techniques and resources required. The different researchstrategies have often distinctly different methods for datacollection and analysis.Quantitative, qualitative and mixed methodsresearchQuite a strong distinction is generally made between quantitativeand qualitative research. Not only do the appropriate data havedifferent characteristics, but they also require different techniquesfor their analysis. Natural science has traditionally concentratedon ‘hard’ quantitative (positivist) analysis, and this was adopted bythe human sciences until its shortcomings became evident. As itbecame increasingly obvious to some researchers that subjectivehuman feelings and emotions were difficult (or impossible) toquantify, qualitative (anti-positivist) analytical methods wereevolved, which took more account of the ‘soft’, personal data. You
are not forced, however, to make a choice between the twoapproaches in your research project. When appropriate, a mixtureof quantitative and qualitative research is possible, a mixedmethods approach as discussed later in this chapter (see alsoCohen et al., 2017, Chapter 2 for a discussion of linking qualitativeand quantitative data analysis). Within certain limits, all types ofresearch strategy are suitable for quantitative and qualitativeresearch.Figure 7.1 It became increasingly obvious to some researchersthat subjective human feelings and emotions were difficult (orimpossible) to quantifyResearch strategiesThe different types of research (or research strategies ormethodologies, as they are often termed) are commonly put intofive major categories: experimental, survey, archival analysis,
historical and case study (I have added several more in Chapter1, which expands on this categorization). In some projects, morethan one may be used, e.g. historical and case study. Eachprovides an alternative way, with its own logic, of collecting andanalysing empirical evidence. Although each has its advantagesand disadvantages, they can all be used for three customarypurposes of research: exploration, description and explanation.The different nature of the various strategies is illustrated in Box7.1, which indicates the forms of research question that they canappropriately answer, whether they require control over the eventsstudied, and whether they focus on contemporary events.Box 7.1 Research strategiesSource: Based on Yin, 2018, pp. 10–13The links between perspectivesIt should be evident by now that from whichever perspective onelooks at research methods, the other two perspectives must betaken into account. The interlinkages are simply portrayed in athree-dimensional matrix in Diagram 7.1.The problem remains of how to provide a clear summary ofmethods within this matrix, which will enable you to understandthe basis of each, and how to help you to decide which methodsare likely to be appropriate for your own research project.
We will look first at the pure data collection methods used inliterary searches (secondary data) and in archival and historicalstudies (secondary and primary data) and then at the quest forprimary sources normally associated with the survey researchstrategy and case study research. In all cases, quantitative andqualitative options are discussed, and the important issue ofsampling is treated at some length.Primary sourcesExperimental research and associated models are generallyinteractive in their use and production of data through analysis, sothese are considered next. The attributes of a range of pure andquasi-experimental methods are described, as are a diverserange of models suitable for quantitative and qualitativeanalysis.Quantitative analysisQualitative analysisThe analysis of quantitative data is then considered, particularlystatistical techniques, used widely to analyse quantitative dataand readily discussed as a distinct analytical method. Statisticsare used in the appropriate situations in most of the researchstrategies, particularly where a large amount of quantitative datais produced and there is a requirement to identify relationshipsbetween variables.Diagram 7.1 Links between perspectives: Matrix diagram
The major activities involved in qualitative analytical methods aresubsequently described, explaining the essential interplay of datamanagement and theory production, which is a feature of thesemethods.Collecting secondary dataAll research studies require secondary sources for thebackground to the study. You will inevitably need to ascertain whatthe context of your research question/problem is, and also get anidea of the current theories and ideas. No type of project is donein a vacuum, not even a pure work of art. However, it is quitecommon in student-level research to rely on secondary data forthe actual research investigations rather than generating newprimary data from the field. Wherever there exists a body ofrecorded information, there are subjects for study. You canimagine using existing resources when doing a historical study
(i.e. of any past events, ideas or objects, even the very recentpast) or a nationwide or even a local study that uses officialstatistics as the principal data.Secondary sourcesThe advantages of using sets of secondary data is that they havebeen produced by teams of expert researchers, often with largebudgets and extensive resources way beyond the means of asingle student, and so they cut out the need for time-consumingfieldwork. Data that have been collected over a long period of timewill provide the opportunity to do a longitudinal study (tracing thedevelopments over time), impossible to do with data collected inshort projects. Secondary data can also be used to compare withprimary data you may have collected, in order to triangulate thefindings and put your data into a larger context. Data in the publicrealm are also open to scrutiny by others and are a permanentresource on which to base your research.The disadvantage is that you miss out on the experiences andskills gained by having to generate your own primary data fromreal-life situations. The data will also have been collected with apurpose that might not match easily with what you need for yourresearch focus. Though many sources of data are free, others arecostly to acquire and may be difficult to access. When using datafrom a variety of sources, there are likely to be mismatches interminology and the way the data has been collected andanalysed, making aggregation of the data difficult or open tochallenge.As a researcher, you will face several problems when seeking andvalidating previously recorded historical data. Chief among theseare listed in Checklist 7.1.Checklist 7.1 Problems with seeking and validating historical dataLocating and accessing themAuthenticating the sourcesAssessing credibilityGauging how representative they are
Selecting methods to interpret themLocating secondary data can be an enormous topic. Activities caninvolve anything from rummaging through dusty archives in anobscure library to downloading the latest government statisticaldata from the Internet or even unearthing city ruins in the desert.Even current data might be difficult to get hold of. For instance, alot of current economic data are restricted and expensive to buy.To be really useful, data sets should have good documentationthat includes descriptions of all the variables and codes, as wellas the recording methods that were used.Types and sources of secondary dataThere are numerous types of secondary data, the main beingdocumentary sources in the form of written and non-writtenmaterials, and survey data in the form of statistical information.Written materials – organizational records such as internalreports, annual reports, production records, personnel data,committee reports and minutes of meetings; businessaccounts, communications such as emails, letters, notes;publications, such as books, journals, newspapers, advertisingcopy, government publications of all kinds.Non-written materials – television programmes, radioprogrammes, tape recordings, video tapes, films of all types,including documentary, live reporting, interviews, works of art,historical artefacts, etc.Survey data – government census of population, employment,household surveys, economic data, organizational surveys ofmarkets, sales, economic forecasts, employee attitudes. Thesemay be carried out on a periodic basis, with frequent regularityor continuously, or on one-off occasions. They may also belimited to sector, time and/or area.It sometimes depends on how you want to use the data whetherthey should be regarded as primary or secondary data. Forexample, if you are analysing a work of art in the form of apainting, you could use it as primary data by looking at the
subject, materials and techniques used in the painting, theproportions, etc. particular to that individual painting or artist.Alternatively you could use it as secondary data when examiningit for aesthetic trends, as evidence for developments in art history,or as a commentary on the society of the time. The same could besaid for pieces of music, films or television programmes.It is impossible to give a full description of all sources ofsecondary data, as the detailed nature of the subject of researchdetermines the appropriate source, and, of course, the possiblerange of subjects is enormous. However, here, in Box 7.2 aresome of the principal types of data and the sources where theycan be found.Box 7.2 Principal sources of secondary informationLibraries and archives: These are generally equipped withsophisticated catalogue systems which facilitate the trackingdown of particular pieces of published data (and evenunpublished data such as diaries and letters) or enable a trawl tobe made to identify anything which may be relevant. The Internetmakes remote searching possible – see your own libraryspecialists for the latest techniques. Apart from thesemodernized libraries and archives, much valuable historicalmaterial is contained in more obscure and less organizedcollections, in remote areas and old houses and institutions. Theattributes of a detective are often required to track down relevantmaterial, and those of a diplomat to gain access to private orrestricted collections.Museums and collections: These often have efficient cataloguingsystems which will help your search. However, problems may beencountered with searching and access in less organized andrestricted and private collections. Larger museums often havetheir own research departments which can be of help.Government departments and commercial/professional bodies:These often hold much statistical information, both current andhistoric.Public sector: Non-government organizations (NGOs) and otherpublic bodies keep records of their activities and statisticalinformation from within their field of action.The Internet: Not only the most convenient portal to many of theabove sources of information, but also a rich mine of data of allkinds not found elsewhere. This is particularly useful for getting
access to films, television programmes, recorded music, etc.which are difficult or expensive to access.The field: Not all historical and modern artefacts are contained inmuseums. Ancient cities, buildings and urban developments,archaeological digs, etc. are all available for study in situ. Here,various types of observation will be required to record therequired data.Suitability of data for your projectIt is worth making several checks before you commit yourself tousing secondary data to make sure that the characteristics of thedata can fulfil your research objectives and answer your questions(see Checklist 7.2).Checklist 7.2 Are your data suitable for your project?Do measures match those you need, e.g. economic,demographic, social statistics?Coverage – is there sufficient data of the required type, and canunwanted data be excluded?Population – is it the same as required for your chosenresearch?What variables are covered – the precise nature of these mightnot be so important for descriptive work but could be essentialfor statistical tests or explanatory research?Will the benefits be greater than your costs?Will you be allowed access?Authentication and credibilityWhen you use data from a source where you have had no controlin its collection, you will have to assure yourself that the data arereliable and sufficiently comprehensive and suitable for yourneeds. This is the process of authentication. With regard toreliability, a quick assessment can be made by examining thesource of the data – what is the reputation of the organizationsupplying the data? Government statistics and data provided bylarge, well-known organizations are likely to be authoritative, as
their continued existence relies on maintaining credibility. Recordsheld by smaller organizations or commercial companies will bemore difficult to check for reliability. In these cases, it is importantto make a check on the person or institution responsible for thedata, and to explore whether there are any printed publicationsrelating to the research which might give them more credibility.AuthenticationAlthough it may be impossible in the case of commercial providersof statistical data, in other cases you should try to make anassessment of the methods of data collection and analysis usedto produce the data. Internet-based data sets may provide thisinformation through hyperlinks, and reports will normally have asection devoted to the research methods used. Issues to beassessed are the sampling method used, the response rate ofsurveys, the context in which the data were collected andrecorded and the analytical methods used to process the data.Analysing secondary dataAnalysis of secondary data can aim at looking for patterns ortrends across the results, to track progressions through time, or toseek out repetition of certain results to build up a strong case. Ofthe many different ways to analyse secondary data, many are nodifferent to those used for primary data, which are described in thelater chapters. However, it is worth mentioning three methods thatare particularly suitable for secondary sources: content analysis,data mining and meta-analysis.Content analysisContent analysis is a quantitative form of analysis that consists ofan examination of what can be counted in text of any form(articles, advertisements, news items, etc.) or other media such aspictures, television or radio programmes or films, and livesituations such as interviews, plays and concerts. The analysis isdone very often, but not necessarily, from secondary sources. The
method was developed from the mid-1900s chiefly in America,and is a rather positivistic attempt to apply order to the subjectivedomain of cultural meaning. It is done by counting the frequencyof phenomena within a case in order to gauge its importance incomparison with other cases. As a simple example, in a study ofracial equality one could compare the frequency of different racesin the illustrations in fashion magazines in various Europeancountries. Much importance is given to careful sampling andrigorous categorization and coding in order to achieve a level ofobjectivity, reliability and generalizability and the development oftheories.Checklist 7.3 lists five basic stages to this method.Checklist 7.3 Five stages in content analysisState the research problem, i.e. what is to be counted and why.This will relate to the subject of the study and the relevantcontents of the documentary source.Employ sampling methods in order to produce representativefindings. This will relate to the choice of publications or othermedia, the examples selected and the sections within theexamples that are investigated.Retrieve the coded fragments. This can be done manually, butcomputer-based search systems are more commonly used whenthe text can be digitalized.Do quality checks on interpretation. This covers issues of:the units of analysis (can the selected stories or themesreally be divided from the rest of the text?)classification (are the units counted really all similarenough to be counted together?)combination of data and formation of ‘100 per cents’ (howcan the units counted be weighted bylength/detail/authoritativeness and how is the totality of theelements to be calculated?)Analyse the data (what methods of interpretation will you use?)Content frames and codingThis is a preliminary analytical method that tabulates the initialresults of content analysis in a content frame. A single publication,
programme or film is analysed in order to establish codes that canbe used as the basis for the units of analysis to be counted. Aseparate content frame is devised to investigate each generalquestion, and each column in the frame is headed by a sub-question that is a component of the general one. The answers tothese sub-questions provide the codes that suggest appropriateunits of measurement. A coding manual can then be producedthat lists descriptions and code numbers of all the codes that willbe searched for and recorded. The codes can consist of a word ora phrase. An example of this is shown in Diagram 7.2 whichrelates to what stories appeared in particular news programmeson television.Diagram 7.2 Content frameTabulation of resultsThe numerical data that form the results of a content analysis aremost conveniently presented in tabular form. The units ofmeasurement are listed and number of appearances noted,together with the percentage this is of the total. Diagram 7.3
shows the results of the number of appearances in one newsbroadcast of the units of analysis listed in Diagram 7.2.What content analysis on its own cannot do is to discover theeffects that the publications, programmes, films, etc. have on theiraudience. Other research methods (e.g. questionnaires,interviews) must be used to gain this type of information. What itcan uncover, however, is how the communications areconstructed and which styles and conventions of communicationare used by authors to produce particular effects. This form ofanalysis allows large quantities of data to be analysed in order tomake generalizations.Diagram 7.3 Tabulation of results of a content analysis showingnumber of appearances of units of analysis in content frameData miningData mining is a technique used extensively by businessmanagements to extract meaningful information from the hugedatabases that are generated by electronic and other methods inmodern businesses. It is often the starting point in decision-basedresearch. Company data are stored in databases called datawarehouses or data marts, containing data from all over the worldin international companies.Data mining uses statistical tools to explore the data forinteresting relationships that can be exploited by the business,such as finding a gap in the market or pinpointing areas ofincreasing demand. This involves pattern discovery and theprediction of trends and behaviours.Data visualization techniques help the analyst to gain a clearunderstanding of the data in pictorial form. These may involve therepresentation of the data in clusters, networks or tree models, or
the arrangement of a set of classifications (e.g. of thecharacteristics of creditworthy customers). Underlying patterns inthe data may be displayed in the form of associations, often basedon market analysis. The patterns might also be considered bytaking in the time element and presenting a sequence-basedanalysis. More complex mining operations employ fuzzy logic,genetic algorithms and fractal-based transformations.Meta-analysisMeta-analysis consists of making an analysis of the results of anumber of results of previous research – an analysis of acollection of analyses, hence ‘meta’-analysis. This is not the sameas a literature review, as it is a statistical analysis of theaccumulated data from the results of previous studies rather thana commentary and critical appraisal of the research projects andtheir outcomes.Checklist 7.4 describes the actions in the stages in this type ofanalysis.Checklist 7.4 Stages in doing meta-analysis1. Define the issue to be investigated – e.g. the effects of class sizeon student learning.2. Collect the studies according to the issue defined at the outset.They may be published or unpublished research results. Caremust be taken to select similar studies of good quality, to avoidvery different types and qualities of data.3. Find common methods of measurement of variables used todetect significant relationships.4. Select the purpose of analysis of results data, either acomparison to detect how much the results of the studies varied,or to track a particular variable across all the studies andaccumulate the results to indicate its importance.5. Carry out the statistical analysis to compare or computesignificance levels. An estimation of the size of the effect of onevariable on another is another aspect to be explored. Sometimesit may be useful to divide the studies into sub-groups to clarifythe outcomes.6. Report the results and discuss the limitations of the researchand recommend further research in the subject.
There are plenty of problems associated with meta-analysis. Themain one is that the probably wide range of methods andstatistical techniques used in the various studies makecomparison and combination difficult to justify. Another is that thepublished works only record successful outcomes wherestatistically significant results are achieved, leaving all the othertest results unrecorded. This can lead to an overoptimistic result inthe meta-analysis. Despite these problems, it is a useful way toassimilate the results of numerous studies dedicated to one issue.Collecting primary dataCollecting primary data involves collecting information directlyfrom the subject of study. A wide range of data collection methodscan be used, such as asking questions (when the subject involvespeople), making observations and measurements, conductingexperiments and tests, and carrying out evaluations. All of thesewill be briefly considered in the following sections.Survey researchSurvey research depends heavily on the process of sampling andon asking questions, through questionnaires, interviews, accountsand tests. You do need to acquire skills to do this properly.‘Anyone’ can do a bad survey … To do a good surveyrequires expertise and professionalism at every stage: thedesign, sampling, questionnaire development, interviewing,analysis and reporting, based on an extensive theoreticalframework well grounded in practice and methodologicalresearch. (Morton-Williams, 1993, p. 2)The world is full of potential data. You will, however, only beinterested in collecting data that are relevant to your study andspecifically required in order to investigate your research problem.Even so, the amount of information you could collect on your
specific subject is likely to be enormous, so a method must beused to limit the amount of data you must collect to achieve youraims. The main technique for reducing the scope of your datacollection is to select and study a sample, i.e. a small section ofthe subject of your study. There are several things one mustconsider in selecting a sample, so before discussing the differentmethods of data collection, let us first deal with the issue ofsampling.SamplingWhen conducting any kind of survey to collect information, orwhen choosing some particular cases to study in detail, thequestion inevitably arises: how representative is the informationcollected of the whole population?When we talk about population in research, it does notnecessarily mean a number of people. ‘Population’ is a collectiveterm used to describe the total quantity of cases of the type whichare the subject of your study. So a population can consist ofobjects, people or even events, e.g. schools, miners, revolutions.A sample is a selected number of cases in a population.PopulationSampleIf you wish to survey the opinions of the members of a small club,there might be no difficulty in getting information from eachmember, so the results of the survey will represent those of thewhole club membership. However, if you wish to assess theopinions of the members of a large trade union, apart fromorganizing a national ballot, you will have to devise some way ofselecting a sample of the members who you are able to question,and who are a fair representation of all the members of the union.Sampling must be done whenever you can gather informationfrom only a fraction of the population of a group or a phenomenonwhich you want to study. Ideally, you should try to select a samplethat is free from bias. You will see that the type of sample you
select will greatly affect the reliability of your subsequentgeneralizations.There are basically two types of sampling procedure – randomand non-random. Random sampling techniques give the mostreliable representation of the whole population, while non-randomtechniques, relying on the judgement of the researcher or onaccident, cannot generally be used to make generalizations aboutthe whole population.Random samplingThe simplest form of random sampling is to represent all the units(sometimes called elements or cases) in a population on slips ofpaper, put them into a hat, and draw out the slips in a randomfashion. As with all samples, the larger the sample the better.However, for a researcher facing the practicalities of a specificresearch situation, this example is not very useful, so a set ofguidelines is called for.First, a question should be asked about the nature of thepopulation: is it homogeneous or are there distinctly differentclasses of cases within it? Different sampling techniques areappropriate for each. The next question to ask is: which processof randomization will be used? The following gives a guide towhich technique is suited to the different populationcharacteristics.Simple random sampling is used when the population isuniform or has similar characteristics in all cases, e.g. aproduction batch of cars of a particular model from which randomsamples are selected for testing as to their quality.Simple random samplingWhen the population is not quite as uniform or one-dimensional asa particular model of a car, simple random sampling is not assimple as it sounds. The procedure should aim to guarantee thateach element (person, group, class, type, etc.) has an equal
chance of being selected and that every possible combination ofthe elements also has an equal chance of being selected. While itis virtually impossible to achieve this in practice, several methods,some using randomly generated numbers, have been devised toproduce some form of a fair lottery in which each combination ofelements has an equal chance in coming up.Simple stratified sampling should be used when cases inthe population fall into distinctly different categories or strata) (e.g.a business whose workforce is divided into the three categories ofproduction, research and management).Simple stratified samplingWith the presence of distinctly different strata in a population, inorder to achieve simple randomized sampling, an equally sizedrandomized sample is obtained from each stratum separately toensure that each is equally represented. The samples are thencombined to form the complete sample from the whole population.Proportional stratified sampling is used when the cases ina population fall into distinctly different categories (strata) of aknown proportion of that population (e.g. a university in which theproportions of the students studying arts and sciences is 61% and39%). When the proportions of the different strata in a populationare known, then each stratum must be represented in the sameproportions within the overall sample. In order to achieveproportional randomized sampling, a randomized sample isobtained from each stratum separately, sized according to theknown proportion of each stratum in the whole population, andthen combined as previously to form the complete sample fromthe population.Proportional stratified samplingIn cluster sampling, cases in the population form clusters bysharing one or some characteristics but are otherwise asheterogeneous as possible (e.g. travellers using main railway
stations). They are all train travellers, with each clusterexperiencing a distinct station, but individuals vary as to age, sex,nationality, wealth, social status, etc. Also known as areasampling, cluster sampling is used when the population is largeand spread over a large area. Rather than enumerating the wholepopulation, it is divided into segments, and then several segmentsare chosen at random. Samples are subsequently obtained fromeach of these segments using one of the above samplingmethods.Cluster samplingSystematic sampling is used when the population is verylarge and of no known characteristics (e.g. the population of atown). Systematic sampling procedures involve the selection ofunits in a series (e.g., on a list) according to a predeterminedsystem. There are many possible systems. Perhaps the simplestis to choose every nth case on a list (e.g. every tenth person in atelephone directory or electoral register). In using this system, it isimportant to pick the first case randomly, i.e. the first case on thelist is not necessarily chosen. The type of list is also significant:not everyone in the town owns a telephone or is on the electoralregister. Diagram 7.4 illustrates these sampling techniques.Systematic samplingNon-random samplingAlthough non-random sampling can be useful for certain studies, itprovides only a weak basis for generalization.Accidental sampling (or convenience sampling) involvesusing what is immediately available (e.g. studying the buildingyou happen to be in, examining the work practices in yourfirm). There are no ways of checking to see if this kind ofsample is in any way representative of others of its kind, so theresults of the study can be applied only to that sample.
Quota sampling is regularly used by reporters interviewingon the streets in an attempt to balance the sample interviewedby selecting responses from equal numbers of differentrespondents (e.g. equal numbers from different politicalparties). This is an unregulated form of sampling, as there is noknowledge of whether the respondents are typical of theirparties (e.g. Labour respondents might just have come from anextreme left-wing rally).Theoretical sampling is a useful method of gettinginformation from a sample of the population that you thinkknows most about a subject. A study on homelessness couldconcentrate on questioning people living in the street. Thisapproach is common in qualitative research where statisticalinference is not required.Accidental samplingQuota samplingTheoretical samplingThree other methods can be briefly mentioned: purposivesampling, where the researcher selects what he/she thinks is a‘typical’ sample; systematic matching sampling, when two groupsof very different size are compared by selecting a number from thelarger group to match the number and characteristics of thesmaller one; and snowball techniques, where you contact a smallnumber of members of the target population and get them tointroduce you to others, e.g. members of a secret society.Diagram 7.4 Examples of random sampling methods
Figure 7.2 Purposive sampling, where the researcher selectswhat he/she thinks is a ‘typical’ sampleHaving selected a suitable sampling method, the remainingproblem is to determine the sample size. There is no easy answerto this problem. If the population is very homogeneous, and thestudy is not very detailed, then a small sample will give a fairlyrepresentative view of the whole. In other cases, you shouldconsider the following.The greater the accuracy required in the true representation of thepopulation, then the larger the sample must be. The size of thesample also should be in direct relationship to the number ofquestions asked, the amount of detail required in the analysis ofthe data, and the number of controls introduced. It is generallyaccepted that conclusions reached from the study of a largesample are more convincing than those from a small one.However, the preference for a large sample must be balancedagainst the practicalities of the research resources, i.e. cost, timeand effort.
The amount of variability within the population (technically knownas the standard deviation) is another important factor indetermining a suitable sample size. Obviously, in order that everysector of a diverse population is adequately represented, a largersample will be required than if the population were morehomogeneous. If statistical tests are to be used to analyse thedata, there are usually minimum sample sizes specified fromwhich any significant results can be obtained. A later part of thischapter deals briefly with statistical methods.No sample will be exactly representative of a population. Ifdifferent samples, using identical methods, are taken from thesame population, there are bound to be differences in the mean(average) values of each sample owing to the chance selection ofdifferent individuals. The measured difference between the meanvalue of a sample and that of the population is called thesampling error.Sampling errorExercise 7.1Answer the following questions:1. Why do researchers use sampling procedures?2. Why should you not make generalizations about a group by theobservation of a single case? Are there instances when this is done?Give one example.3. a What are the two basic types of sampling procedure, and what isthe difference between them?b When is it appropriate to use them?4. What factors must you examine when deciding on an appropriaterandom sampling method?5. What are the critical issues which determine the appropriate samplesize?
Perhaps in your own research project you will not need to do anysampling. However, because the need to select samples is socommon it is essential that you understand the principles involved,even if only to argue why you did not find it necessary to use asampling method. Remember that any form of generalization fromparticular cases implies that those cases are somehow representativeof the wider whole.QuestionnairesAsking questions is an obvious method of collecting bothquantitative and qualitative information from people. Using aquestionnaire enables you to organize the questions and receivereplies without actually having to talk to every respondent. As amethod of data collection, the questionnaire is a very flexible tool,but it must be used carefully in order to fulfil the requirements of aparticular piece of research. While there are whole books on theart of questioning and questionnaires, it is possible to isolate anumber of important factors to consider before deciding to use aquestionnaire.Before examining its form and content, we will briefly considerwhy one might choose this form of data collection, and ways inwhich the questionnaire can be delivered.One of the main features of a questionnaire is its impersonality.The questions are fixed, i.e. they do not change according to howthe replies develop, and they are the same for each respondent,and the person posing the questions is remote. The responsescan be completely anonymous, allowing potentially embarrassingquestions to be asked with a fair chance of getting a true reply.Another feature is that there is no geographical limitation withregard to the location of the respondents: they can be anywhere inthe world as long as they can be reached by post or online.Questionnaires can be a relatively economic method, in cost andtime, of soliciting data from a large number of people. Time forchecking facts and pondering on the questions can also be takenby the respondents, which tends to lead to more accurateinformation.
There are two basic methods of delivering questionnaires:personally and remotely. The advantages of personal delivery arethat respondents can be helped to overcome difficulties with thequestions, and that personal persuasion and reminders by theresearcher can ensure a high response rate. The reasons whysome people refuse to answer the questionnaire can also beestablished, and there is a possibility of checking on responses ifthey seem odd or incomplete. This personal involvement of theresearcher enables more complicated questionnaires to bedevised. Obviously, there are problems in both time andgeographical location which limit the scope and extent to whichthis method of delivery can be used.Postal and online questionnaires do not suffer from these twolimitations. However, the most serious problem is that the rate ofresponse is difficult to predict or control, particularly if there is nosystem of follow-up. The pattern of non-response can have aserious effect on the validity of the sample by introducing bias intothe data collected. Cost is often a determining factor in choosingpostal distribution: it is cheap compared with interviewing. It mightalso be the only method of questioning people spread over a largearea or situated in relatively inaccessible regions if you do nothave access to their email addresses or they are not online.There are simple rules to devising a questionnaire, though it is notalways easy to carry them out perfectly. The rules are shown inChecklist 7.5. A questionnaire should be pre-tested on a smallnumber of people in what is called a pilot study. It is best to testit on people of a type similar to that of the intended sample, so asto anticipate any problems of comprehension or other sources ofconfusion. When sending out the questionnaire, you shouldcourteously invite recipients to complete it, and encourage themby explaining the purpose of the survey, how the results could beof benefit to them, and how little time it will take to complete.Simple instructions on how to complete the responses are alsorequired. Some form of thanks and appreciation of their effortsshould be included at the end. If you needed to be sure of aresponse from particular persons, you should send a preliminaryletter, with a reply-paid response card, to enquire whether theperson is willing to complete the questionnaire before you send it
(Leedy and Ormrod, 2015; see also Moser and Kalton, 2016,Chapter 13; Hague, 1993; Oppenheim, 1998).Pilot studyChecklist 7.5 Rules for devising a questionnaire1. You must establish exactly which variables you wish to gatherdata about, and how these variables can be assessed. This willenable you to list the questions you need to ask (and those thatyou do not!) and to formulate the questions precisely in order toelicit the responses that are required.2. The language must be unmistakably clear and unambiguous andmake no inappropriate assumptions. Leedy and Ormrod give anilluminating example of a seemingly simple question that is quiteambiguous: ‘How many cigarettes do you smoke each day: morethan 25, 25–16, 15–11, 10–6, 5–1, none?’ (2015, p. 167). Thereis no problem answering this if you are a regular smoker, butwhat if you smoke only at weekends, at parties or in theholidays?3. In order to enhance the response rate, questions generallyshould be kept simple, and the questionnaires kept as short aspossible. This minimizes the effort required by the respondent.4. Clear and professional presentation is another essential factor inencouraging a good response.5. Consider how you will process the information from thecompleted forms. This may influence the layout of thequestionnaire, e.g. by including spaces for codes and scoring.DiariesYou can think of the use of a diary in research as a kind of self-administered questionnaire. The form of text can be totallyunstructured or can be a response to a series of questions.Although diaries can provide a huge amount of data for very littleeffort from the researcher, the content can be quite haphazarddepending on the interpretation of the task by the respondent.Bias is especially prone to creep in if respondents believe thatsome actions will be taken as a result of the contents, andtherefore record events in such a way that might be beneficial to
them. It is therefore important to ensure that respondents areclear as to what they have to do, why and when, and that theyagree to full cooperation. Confidentiality is obviously a majorissue.A diary method can be used as a substitute for observation wheredirect observation is difficult or impossible, e.g. in intimate orprivate situations. It can also be usefully used as a precursor tointerviews: the information gained will provide pointers to anappropriate list of questions to be asked in the interview (Robsonand McCartan, 2016, pp. 273–75).Interviews: structured, semi-structured andopenWhile questionnaire surveys are relatively cheap and are effectivein preventing the personality of the interviewer having effects onthe results, they do have certain limitations. They are not suitablefor questions that require probing to obtain adequate information,as they should only contain simple, one-stage questions. Thereare also problems in gaining the required response from thecomplete sample, especially as the questionnaires tend to bereturned by the more literate sections of the population. Whileinterviewing is suitable for quantitative data collection, it isparticularly useful when qualitative data are required.When you wish to question samples of people, the interview is avery flexible tool with a wide range of applications. There are twomain methods of conducting interviews: face-to-face andtelephone.Figure 7.3 Face-to-face interviews can be carried out in a varietyof situations
Face-to-face interviews can be carried out in a variety of situations– in the home, at work, outdoors, on the move (e.g. whiletravelling) – and can be used to question members of the generalpublic, experts or leaders, specific segments of society (e.g.elderly or disabled people, ethnic minorities) both singly and ingroups. Interviews can be used for subjects both general orspecific in nature and even, with the correct preparation, for verysensitive topics. They can be one-off or, for longitudinal studies,repeated several times over a period to track developments. Theinterviewer is in a good position to be able to judge the quality ofthe responses of the subjects, to notice if a question has not beenproperly understood, and to reassure and encourage therespondent to be full in his/her answers. Visual signs, such asnods, smiles, etc., are valuable tools in promoting completeresponses.
Telephone interviews avoid the necessity of travelling to therespondents, and all the time and problems associated withcontacting people personally. With wider telephone ownership,particularly in developed countries, it is often possible to contact asuitable sample of the target population by this method. Surveyscan be carried out more quickly than face-to-face, especially if thequestionnaire is short (20–30 minutes is the accepted maximum).However, visual aids cannot be used to explain questions, andimportant visual clues between interviewer and interviewee (e.g.eye contact, smiling, puzzled looks) are absent. Using video callscan help to allay these issues. Non-response tends to be 5–10%higher on telephone surveys than with equivalent face-to-facesurveys (Morton-Williams, 1993, p. 156). It is interesting to notethat voice quality is an important factor in successful phoneinterviews. Interviewers with the highest success rates spokerapidly and loudly, used standard pronunciation and soundedcompetent and confident (1993, p. 169).The structuring of the interview depends on the type of informationyou wish to elicit. For very precise answers to very precisequestions, used for quantitative and statistical analysis, a tightlystructured interview is required with closed questions formulatedin a method similar to a questionnaire. At the other extreme, if youneed to explore a situation and wish to get information that youcannot predict, a very open and unstructured form of interview isappropriate. A semi-structured interview falls between the two,achieving defined answers to defined questions, while leavingtime for further development of those answers, and including moreopen-ended questions. The most important issue when you set upan interview is to know exactly what you want to achieve by it, andwhat you intend to do with the information gained, and to choosethe appropriate structure of interview to achieve this.Interviews, because of their flexibility, are a useful method ofobtaining information and opinions from experts during the earlystages of your research project. It is important, however initialyour research enquiry, to devise an efficient method of recordingwhat is discussed in the interview and to form a simple structurefor organizing the data collected. These early interviews oftenprovide very important information which helps not only to indicate
your future research direction, but also to identify and prioritizeissues. You should aim to achieve a balance between openquestioning to explore issues, and obtaining responses that cansubsequently be easily examined and compared.There is a great difference in technique for conducting interviews‘cold’ with the general public and interviewing officials or expertsby appointment. In both cases, however, the personality andbearing of the interviewer are of great importance. There is notspace here to go into detail about the different techniques thatshould be acquired for the various interview situations. These areexamined in detail in Morton-Williams (1993). What must bestressed, however, is the need for preparation on the part of theinterviewer, both in the groundwork (i.e. writing letters forappointments, explaining the purpose of the interview) and inpresenting the interview, cultivating confidence, friendliness, goodappearance, etc. and selecting an appropriate method ofrecording the responses (tape recording, writing notes, completingforms, etc.).Standardized scales and testsThese can perhaps be seen as versions of structured interviewsor of self-completion questionnaires. There is a wide range ofstandardized scales and tests that have been devised by socialscientists and psychologists to establish people’s abilities,attitudes, aptitudes, opinions, etc. A well-known example of one ofthese is the IQ or intelligence test. The objective of the tests isusually to measure in some way the abilities etc. of the subjectsaccording to a standardized scale, so that easy comparisons canbe made. One of the main problems facing the researcher is toselect or devise a suitable scale for measuring the often ratherabstract concepts under scrutiny, such as attitude to school meals,military service, capital punishment, etc.A common precaution to prevent oversimplification of responsesis to ask many questions about the same topic, from a variety ofdifferent angles. This form of triangulation helps to build up a morecomplete picture of complex issues. The use of arbitrary scalesshould be avoided. It is safer to use well tried and tested standard
scales, of which there are several, each taking differentapproaches according to the results aimed at. To mention a few:the Likert scale, one of the most common, which uses asummated rating approach; the Thurlstone scale, which aims toproduce an equal-appearing interval scale; the Guttman scale, aunidimensional scale where items have a cumulative property;and the semantic differential scale, where the subjective meaningof a concept to a respondent is gauged, rather than his/herattitude towards it. At this stage you are unlikely to need to knowthe details of these testing methods, unless they form a centralpart of your research. If you want to know more, Robson andMcCartan (2016, pp. 307–15) give a useful, succinct introductionand further references.AccountsAccounts is a method of qualitative data collection, used mainly insociological research; it seeks to find information on people’sactions and feelings by asking them to give their owninterpretation, or account, of what they experience. This form ofstudy is called ‘ethnogenic’, and is founded on the belief thathuman beings are plan-making, self-monitoring agents, who areaware of goals and deliberately consider the best ways ofachieving them (Cohen et al., 2017, p. 292; and if you want amore comprehensive account of the ethnogenic approach seeHarré, 1977). Accounts can consist of a variety of data sources: aperson’s spoken explanations, behaviour (such as gestures),personal records of experiences and conversations, letters anddiaries. As long as the accounts are authentic, there should be noreason why they cannot be used as an argued explanation ofpeople’s actions.Since the information must come directly from the respondents,care must be taken to avoid leading questions, excessiveguidance and other factors that may cause distortion. Checkingthe authenticity of the accounts is an important element in thisform of data collection. This can be done by cross-checking withother people involved in the events, examining the physicalrecords of the events (e.g. papers, documents, etc.) and checkingwith respondents during the account-gathering process.
The transformation of the collected accounts into workingdocuments that can be coded and analysed is the next procedureto be considered. At this stage, additional checks on theauthenticity of the accounts can be made as well as on thereliability of the transformation. The documents can then beanalysed qualitatively or quantitatively, depending on the type ofresearch problem and the nature of the enquiry. More details ofthis process can be seen in Brown and Sime (1981).Observations and physical surveysObservation is a method of recording conditions, events andactivities through the non-inquisitorial involvement of theresearcher. The non-participant researcher takes a detachedstance to the phenomena, and aims to be ‘invisible’, either in factor in effect (i.e. by being ignored). The participant observer isseen as involved in the process or activity. Observation can alsobe used for recording the nature or conditions of objects (e.g.buildings). This type of observation is often referred to as asurvey, and can range from a preliminary visual survey to adetailed survey using a range of instruments for measurement. Asan activity, as opposed to a method, observation is of courserequired in many research situations, for example, observing theresults of experiments, the behaviour of models and even thereactions of people to questions in an interview. Observation canbe used to record both quantitative and qualitative data.Observation can record whether people act differently to whatthey say or intend. They can sometimes demonstrate theirunderstanding of a process better by their actions than by verballyexplaining their knowledge. For example, a machine operator willprobably demonstrate more clearly his/her understanding of thetechniques of operating the machine by working with it than byverbal explanation.Observation is not limited to the visual sense. Any sense (e.g.smell, touch, hearing) can be involved, and these need not berestricted to the range perceptible by the human senses. Amicroscope or telescope can be used to extend the capacity of theeye, just as a moisture meter can increase sensitivity to the
feeling of dampness. You can probably think of instruments thathave been developed in every discipline to extend theobservational limits of the human senses.On the one hand, observations of objects can be a quick andefficient method of gaining preliminary knowledge or making apreliminary assessment of its state or condition. For example,after an earthquake, a quick visual assessment of the amount andtype of damage to buildings can be made before a detailed surveyis undertaken.On the other hand, observation can be very time-consuming anddifficult when the activity observed is not constant (i.e. much timecan be wasted waiting for things to happen, or so much happensat once that it is impossible to observe it all and record it).Instrumentation can sometimes be devised to overcome theproblem of infrequent or spasmodic activity (e.g. automaticcameras and other sensors).Certain basic hints on how to carry out observations are given inChecklist 7.6.Checklist 7.6 Hints on how to do observations1. Make sure you know what you are looking for. Events andobjects are usually complicated and much might seem to berelevant to your study. Identify the variables that you need tostudy, and concentrate on these.2. Devise a simple and efficient method of recording theinformation accurately. Rely as much as possible on tickingboxes or circling numbers, particularly if you need to record fast-moving events. Obviously, you can leave yourself more timewhen observing static objects to notate or draw the datarequired. Record the observations as they happen. Memories ofdetailed observations fade quickly.3. Use instrumentation when appropriate or necessary. Instrumentswhich make an automatic record of their measurements are tobe preferred in many situations.4. If possible, process the information as the observationsprogress. This can help to identify critical matters which needstudy in greater detail, and others which prove to beunnecessary.
5. If you are doing covert observations, plan in advance what to doif your presence is discovered, to avoid potentially embarrassingor even dangerous situations.Using the Internet for primaryresearchThe use of the Internet to find secondary data has beenmentioned earlier in this chapter. But can it be used to gainprimary data? Yes: according to Hewson et al. (2016), surveysand questionnaires are the most widely implemented Web-basedmethodology. Although the typical Web-based survey involves astructured questionnaire, they maintain that the Internet can beused for the whole range of surveys, from structuredquestionnaires to unstructured interviews, and even observationalstudies and experimental designs. However, there is, as yet, littlepublished research on the important factors, such as responserate, bias, non-response, sincerity and completion rates. In thefield of marketing, however, there has been plenty of discussionon the subject of maximizing responses.The attractions of using a Web-based survey are that it canpotentially reach an enormous number of respondents anywherein the world, the costs and time involved in distribution andcollection of questionnaires are minimized, and analysis of datathat are already in electronic format is made easy. Anonymity ofthe researcher and respondent are also enhanced, which can helpto overcome bias engendered by nationality, sex, age, etc. and toencourage frankness and higher response rates. There are cheapand easy-to-use survey packages such as Survey Monkey andQualtrics that provide ready-made questionnaire formats and evenprovide basic statistical analyses of the collected data.Additional features which Web-based methods can readilysupport, include skip logic (moving to particular subsequentquestions depending on participants’ answers to previousquestions), question piping (customising later questions using
text entered by participants in earlier responses), andchecking for response completeness and correct format. Thisgives Web-based approaches a big advantage when it comesto considerations of flexibility, consistency, reliability andvalidity. (Hewson et al., 2016)But there are particular problems that must be faced, such as lackof control by the researcher over the quality of the responses thatcan lead to questions about the reliability and validity of the data.There is also the issue of sampling. The most common approachuses a non-random sampling technique, making announcementson websites or newsgroups to recruit volunteer participants. Thisraises concerns about the generalizability of the results, as thereis little knowledge either of the population or of the samplereached (Benfield and Szlemko, 2006). Better control and morereliable sampling techniques can be applied by using email as themedium of the survey. The problem then posed is how to compilea suitable list of addresses.Email and, even better, ‘chatrooms’ lend themselves to conductinginterviews over the Web. Although the extra-linguistic cues aremissing in this form of communication, there might be advantagesdue to the anonymity of the persons taking part in the discussion.As for observation studies, the written medium of the Web limitsthese to mainly linguistic projects focusing on Internetcommunication itself. Compared with more conventionalobservation studies, the observer presence bias is removed andthe researcher is able to access a wide range of material to locatesomething suitable for the project. Non-language, behaviouralobservation research can be devised using simulation, gaming orrole-play techniques. The use of webcams for direct observationscan provide automatic recordings of events. Web-basedexperiments using printed material, video, sound or interactivetasks or games are possible if the design allows the researcher tomanipulate the independent variables in order to measure theeffect on the dependent variables.There is not enough space here to consider this medium forresearch in detail. A host of technical issues must be faced as well
as those of challenges to the quality of the research itself. Theseare all comprehensively explained by Hewson et al. (2016), whoselist of ten general principles for use in developing an Internetsurvey is reproduced in Checklist 7.7.Checklist 7.7 General principles for developing an Internet survey1. Collect information about participant demographics (gender,education, income, nationality, occupation, and frequency ofInternet use).2. Aim for sampling procedures that allow measurement ofsampling frame and response rates.3. Control access to the survey (for example, by using passwords).4. Send a preliminary request for participation prior to sending thesurvey itself.5. Include an introduction to the survey that gives affiliation details,and provides a professional appearance.6. Aim for simplicity and clarity in layout and presentation.7. Keep procedures and software requirements as low-tech aspossible, as appropriate for the nature of the study.8. Undertake extensive piloting across different platforms.9. Maintain participant anonymity unless it is essential to the studyto obtain participant identity.10. Use server-side scripting (as opposed to the ‘Mailto:’ command)in order to collect data if resources allow. (Hewson et al., 2016,pp. 37–9)With the integration of Internet technologies into every aspect ofour lives, and the increase in familiarity with smartphones, tabletsand laptops, there is an enormously growing potential for usingInternet-mediated research in more imaginative ways. Thecapability of smartphones and tablets to display, make andtransmit photos and films, provide location data, download apps,and record, play and transmit music increases the range of datathat can be collected by the researcher, which not only transformsthe ways that questionnaires can be designed, but also allowsreal-time transfer of multimedia data. This is aided by thespeeding up of Internet communications through more powerfulmodems, fibre cables and 4G and now the beginner of 5G mobilewireless networks. Virtual reality methods are now being rapidlydeveloped and the required equipment simplified and less costly,
opening up further possibilities for research into people’sperceptions and reactions.Exercise 7.21. What is wrong with these questions, designed to be part of aquestionnaire?1. Where do you come from?2. How many children do you have? Circle one number: 0, 1, 2, 3,4.3. Do you like modern architecture? Score your liking by giving anumber between 0 and 100: 0 = dislike very much, 100 = likevery much.4. When did you last drink a bottle of champagne: yesterday, lastweek, last month, last year?5. Which one of the following concepts – war, society, publicrelations, revolution, education – is described in this text: ‘thetemporary outcome of the attempt to limit the infinite play ofdifferences in the site of the social, to domesticate the potentialinfinitude of symbiosis corroborated by the principle ofindeterminacy of meaning, and to embrace it within the finitudeof an order’?2. What are the main differences between structured interviews, semi-structured interviews, open interviews and accounts?3. Are sampling techniques used only when aiming to getrepresentative information from a population consisting of people, forexample, when making a survey of user attitudes?When making generalizations from the conclusions of your research, itis most likely that you have come to your conclusions on the basis ofwhat you have learned from only a few cases amongst all possiblecases. It is virtually impossible to examine every case of anyphenomenon, so sampling, in one form or other, is one of the mostprevalent features of research.Combined data generation andanalysis
As a method of analysing phenomena, and gaining moreinformation about them, it is sometimes possible artificially tocreate a microcosm of the real world, where only the elementswhich are regarded as important are included, so that these canbe studied in detail by directly manipulating the variables. Thelaboratory experiment is such a method. There are many otherlocations where experiments can be carried out, but the laboratorysituation is the one which provides the greatest possibilities forcontrol. Use of models is another method of obtaining informationabout the real world in a controlled situation. In these methods,the collection and analysis of data are inextricably linked. Thepreliminary data on which the experiments and models are basedare used to create new data which, in their turn, can be used forfurther analysis. Checklist 7.8 summarizes the difference betweenexperiments and models.Checklist 7.8 The difference between an experiment and a modelAn experiment is used to examine actual phenomena, which arecontrolled in scope and size.A model provides an artificial version of the phenomenon forstudy, either by mimicking it at a reduced scale or by abstractingit diagrammatically or mathematically (often using computertechniques).Experiments are powerful and versatile techniques that can bedesigned to fulfil virtually any of the analytical functionsmentioned in the next main section.Generally, experiments are designed and carried out in order toexamine causes and effects (studying dependent andindependent variables), and are used to find explanations forthem, e.g. What happens if…? And why? The design of theexperiments depends on the type of data required, the level ofreliability of the data required, and practical matters associatedwith the problem under investigation.Checks should be carried out on both experiments and models totest whether the assumptions made are valid. In experiments, acontrol group is used to provide a ‘baseline’ against which theeffects of the experimental treatment may be evaluated. Thecontrol group is one that is identical (as near as possible) to theexperimental group, but does not receive experimental treatment
(e.g. in a medical experiment, the control group will be givenplacebo pills instead of medicated pills).Figure 7.4 Experiments are powerful and versatile techniquesLaboratory and field experimentsWhat are the significant differences between doing experiments inthe contrived setting of a laboratory and those done in a real-lifesetting? Social science is concerned with what is happening in thereal world, so isn’t the laboratory the wrong place to do socialexperiments?Laboratory experiments have the advantage of providing a gooddegree of control over the environment, and of studying theeffects on the subjects involved. With the aid of some deception,the subjects might not even be aware of what effects they arebeing tested for. Despite the artificiality of the setting, this canprovide reliable data that can be generalized in the real world.
However, according to arguments by leading academics referredto in Robson and McCartan (2016, p. 112) the disadvantages oflaboratory experiments are that they may:lack experimental realism – the conditions may appear to beartificial and not involve the subjects in the same way as in arealistic settinglack mundane realism – real-life settings are always muchmore complicated and ambiguous than those created in alaboratorylead to bias through demand characteristics – the expectationof the subjects that certain things are expected or demandedof them, and reaction to the knowledge that they are beingobservedlead to bias through experimenter expectancy – the oftenunwitting reactive effects of the experimenters that lead to abiased view of the findings to support the tested hypothesis.It is not always easy to distinguish between laboratory and fieldexperiments. Realistic simulations of rooms in a laboratory, or the useof normal settings as laboratories for the purposes of experimentsmake it difficult to know where to draw the line. There may also be asense of artificiality in a natural setting when people are organized forthe purposes of the experiment, or people are just aware that they aresubjects of investigation.In field experiments, planned interventions and innovations arethe most useful strategies for natural experiments as they providepossibilities to apply relatively reliable experimental designs,involving control groups and getting information prior to theinterventions. External validity (generalizability to the real world) isobviously more easily achieved when the experiments are carriedout in normal life settings. Subjects are also more likely to reactand behave normally rather than being affected by artificialconditions. In most cases, it is also easy to obtain subjects to takepart in the research as they need not make any special effort toattend at a particular time and place. However, the move out of
the confined and controllable setting of the laboratory raises someproblems:Faulty randomizationLack of validityEthical issuesLack of controlTypes of experimentShadish, Cook and Campbell (2001) divided experiments into fivegeneral types, as shown in Checklist 7.9, which can be used tocompare the characteristics of different types of research and thelevel of their rigour and thus the reliability of their outcomes.Checklist 7.9 Types of experimentPre-experimental – with only partial or no control of the variablesand with no control groupTrue experimental – with total control of the selected variablesand with a control group, using a random sampleQuasi-experimental – various techniques with less rigour thanwith true experimentsCorrelational – looking for cause and effect situationsEx post facto – systematic investigations after an event hasoccurredTo explain how they work, here is a brief summary of the differentdesigns under these headings, each illustrated with an example ofstudying the same simple phenomenon (the effect of a revisioncourse on a group’s exam results).Pre-experimental designsOne-shot case study (after only). This is the most primitive type ofdesign where observations are carried out only after theexperiment, lacking any control or check.Example: A group does the revision course and the exam resultsare reviewed. Do good results mean that the course was
effective?One group, pre-test–post-test (before–after) design. Here thesubject (group) is examined before the experiment takes place.Example: The group does the exam before taking the course andthe results are reviewed. The group does the course and a furtherexam is taken. These exam results are compared with theprevious ones. Better results in the second exam may lead to theconclusion that the course was effective.Static group comparison (before–after) design. Similar to theprevious design except that a control group is introduced.Example: Two groups are selected at random. The experimentalgroup does the course, the control group does not; both groupstake the exam and the results are compared. If the control groupdoes less well, we might conclude that the course enhances examsuccess.The trouble with these designs is the lack of control of thevariables, which can seriously affect the outcomes, e.g. whathappens if some of the groups are already well prepared for theexam or are not interested in learning?True experimental designsPre-test–post-test control group (before–after) design. This is thecommonest true experimental design.Example: Two groups are selected in the same randomprocedure. Both do the exam (pre-test); one group does thecourse, the other not. The groups are examined again; the examresults are compared. Best results are gained if both samplesachieve identical results in the pre-test exam.Solomon four-group (before–after) design. This is a refinement ofthe previous design, using four samples, which additionally teststhe effects of the pre-test.
Example: Four groups are selected in the same randomprocedure. Two do a pre-test exam; one of these then does thecourse. Of the other two, one does the course. All four then do theexam; the results are compared. It will be detectable if the pre-testexam affected subsequent performance by comparing those whodid the pre-test exam with those who did not.Post-test only – control group design (after only) design. This isused when a pre-test is not possible, e.g. in a one-off situation likean earthquake or during a continuous development, or if the pre-test would destroy the material. In this case, let us assume thatonly one set of exam questions is allowed.Example: Two groups selected in the same random procedure.One does the course, the other not. Both do the exam; the resultsare compared. The validity of this test critically depends on therandomness of the sample.Quasi-experimental designsNon-randomized control group, pre-test–post-test design. Whenrandom selection cannot be achieved, the control group and theexperimental group should be matched as nearly as possible.Example: Two sets of adjacent unpainted wood panels on abuilding façade are tested for water absorption. One set ispainted, the other left unpainted; both are retested for waterabsorption and the results compared.Time-series experiment. Repeated identical experiments aremade, then one variable is changed to produce a new outcome,and the new experiment is repeated to check if the variableconsistently creates the changed outcome.Example: The water absorption of an unpainted piece of wood isrepeatedly tested. The same sample is painted; the waterabsorption of the painted sample is repeatedly tested. The dangerwith this design is that, over time, other unknown factors mightaffect the results (e.g. the water used in these tests may breakdown the wood and make it less absorbent).
Control group, time-series. The same process as above, but witha parallel control group, which does not undergo the variablechange.Example: As above but with a parallel sample that remainsunpainted and is used to compare outcomes.Correlational and ex post factoCorrelational is a design that is prone to misuse. After acorrelation between two factors is statistically proved, a claim ismade that one factor has caused the other. Life is rarely sosimple! There may be many other factors that have not beenrecognized in the research, one or some of which could be thecause or could have contributed to the cause.Ex post facto is not really an experimental approach in that theinvestigation begins after the event has occurred so no controlover the event is possible. The search for the cause of the event(e.g. a plane crash or the outbreak of an unknown disease) relieson the search for, and analysis of, relevant data. The most likelycause has to be discovered from amongst all possible causes, sothere are many opportunities to search in the wrong area! This isa common form of scientific investigation, and needs the skills of adetective in addition to those of a scientist.ValidityThe quality of data gained from true experimental designs shouldgenuinely reflect the influence of the controlled variables(internal validity) and should enable generalizations to bemade beyond the immediate experimental situation (externalvalidity). Cohen et al. (2017, p. 155) list the factors that cause athreat to validity, and these are worth summarizing briefly here.For example, unnoticed interfering events between pre-test andpost-test observations that affect results, or the subjects of theexperiment may change (e.g. samples deteriorate with age).Faulty or inappropriate measuring instruments and shortcomingsin the performance of human observers lead to inaccurate data,
as can bias in the samples due to faulty or inadequate samplingmethods. Dropout of experimental subjects (not necessarilythrough death!) during the course of a long-running experimenttends to result in bias in what remains of the sample. Also, peopletend to react differently if they know that they are the subject of anexperiment (known as the Hawthorne effect) or can learn ways ofmanipulating the results during an experiment.Internal validityExternal validityVague identification of independent variables will make it impossiblefor subsequent researchers to repeat the experiments.ModelsModels are a method of selectively mimicking reality in a form thatcan be manipulated, in order to obtain data about the effects ofthe manipulations. Broadbent (1988, p. 91) described the threebasic functions of models to be: descriptive, concept structuringand exploratory. He split these basic functions into more detailedones, which include almost the entire list of reasons for doinganalysis given in the next section (see Checklist 7.10).Checklist 7.10 Functions of modelsDescriptive. Isomorphic, i.e. a true representation in all details ofthe object (e.g. a wiring diagram); homomorphic, i.e. a simplifiedrepresentation showing a selection of features (e.g. a touristroad map)Concept structuring. Data defining; data collecting; dataordering; data predictingExploratory. Hypothesis testing; hypothesis generating
In order to make a model, it is necessary to understand thesystem that lies behind the phenomena in reality, to understandwhich are the important variables and how they interact. Theactual form of the model can be diagrammatic, physical ormathematical. Qualitative models emphasize the relationshipsbetween entities without trying to quantify them, while quantitativemodels not only describe the relationships but also accuratelymeasure their magnitude.Checks should be carried out on models to test whether theassumptions made in order to set up the model are valid. Theresults obtained by using a model should also be checked againstdata collected from the actual case that is being modelled.Diagrammatic modelsA simple method of representing systems is the creation ofdiagrams. A wide range of techniques can be used to portrayvarious aspects of a system; most of them are qualitative. Here, inChecklist 7.11 is quite a comprehensive list, with shortexplanations of their nature.Checklist 7.11 Different types of diagrammatic modelsMultiple cause diagram. Shows the causal links between variablesin a system. The links can be sequential or contain loops.Systems map. A ‘snapshot’ showing the components of a systemand its environment, revealing any subsystems and significantoverlaps.Influence diagram. Represents the main structural features of asituation and the important relationships (not necessarily causal)that exist among them.Sign-graph diagram. A development of the multiple causediagram, it records positive or negative causal links, and situationsof positive and negative feedback.Cognitive mapping. A diagrammatic record of a person’s thoughtsabout a particular issue or situation, showing their structure andthe connections between ideas and attitudes.Organization chart. Shows the paths of command andauthorization in an organization.Critical path diagram. Displays the sequence of events andactivities in a process, and reveals the critical elements. It is useful
in predicting, planning and controlling.Linear responsibility diagram. A development of the critical pathdiagram to show responsibilities of, and relationships between, theparticipants.Technical drawing or map. Reduces to two dimensions andsimplifies actual objects or systems (e.g. London Undergroundmap).Flow chart. Not so much the portrayal of a system as a step-by-step set of rules to calculate or instruct.Physical modelsThese are three-dimensional representations of an object orbuilding at a reduced scale. The lavish architectural display modelis a familiar spectacle in the developer’s boardroom, but is notusually the type of model which is useful for research purposes.Models devised for the purpose of research must be specificallydesigned to test the variables which are central to the problembeing investigated. Thus, a model made to test the acoustics of aconcert hall is very different to one made to test mechanicalmethods of raising and lowering the stage floor elements. Theycan be qualitative or quantitative in nature.One of the main problems when using spatial physical models forproducing quantitative data is the problem of scaling. It raises thequestion: do materials and forms of energy behave identically atany scale? Usually, the answer is no. In a model made to test theacoustics of a concert hall, the sound frequency and amplitude ofthe sound source may have to be scaled down, as may thetexture and density of the materials used to line the hall. Inmechanical models, the different behaviour of materials at a smallscale must be compensated for. To overcome the scaling problem,full-scale prototypes are used where possible.Mathematical modelsThese are invariably quantitative models and are divided into twomajor categories, deterministic and stochastic models. Thesecategories relate to the predictability of the input: deterministicmodels deal only with predetermined inputs, whereas stochastic
models require unpredictable inputs to produce satisfactoryresults (often produced by using a random number generator). Tomake this clearer, here are two examples. A mathematical modelthat calculates the heat loss through walls of different materialsand constructions is likely to be a deterministic model: the inputsfor the insulation qualities of the materials will be known, as willthe temperature difference between inside and outside etc.Another term for this type of model is an algorithm. A modelthat investigates the effects of a traffic-light control system in atown centre might be designed on a stochastic model, determiningthe traffic flows from each direction randomly, in order to test thesystem under widely different and unpredictable conditions.AlgorithmThe computer is an invaluable tool in the construction ofmathematical models. Spreadsheet programs provide asystematic two-dimensional framework for devising models, andfurnish facilities for cross-calculations, random numbergeneration, setting of variable values, and the build-up of series offormulae.Scope and limitations of modelsThe essential qualities of a model are that it should be constructedfor a particular purpose and that it should, in some way, reducethe complexity of the real situation. It is important that thepurposes of the model and the assumptions on which it is built areclearly enunciated. The scope of applicability should also bedescribed. You should note that the same object or event can bemodelled in a number of different ways. However, because eachmodel is devised for a particular purpose, it is potentiallydangerous to use it for some other, unrelated, purpose. It couldalso produce misleading information if it is used beyond the rangeor area of applicability for which it was designed. Models arenever perfect because of the many difficulties faced by theresearcher.
The main factors that limit the capability of models to accuratelymimic reality are listed in Checklist 7.12.Checklist 7.12 Model limitations for mimicking realityData limitations. It is obviously impossible to incorporate arelevant variable into a quantitative model if you have nomeasurement of its values. If you have incomplete or onlyapproximate values, then you will have to rely on guesswork tocomplete the information. Additionally, if you do not know inwhich ways the variables interact with each other, then yourmodel will require much conjecture to complete it.Structural limits. If incorrect assumptions are made about therelevance of the variables and the manner in which they interact,then the model will fail, sooner or later, to accurately reflectreality. For example, the early models of the solar system,devised on the basis of the earth at the centre, failed torepresent the future motions of the planets accurately. Thus, it iswrong to assume that however closely the model reflects reality,it is the ‘right one’.Chaos. It has long been assumed that if a system wasaccurately modelled, the behaviour of the model would closelyreflect reality. Any small errors in the initial settings of the model(say of 1%) would produce only correspondingly small deviationsin the results compared with reality (say 1–2%). It is now knownthat some sorts of systems do not behave in this way, and thattiny changes in the variables’ values can result in dramaticsubsequent changes. Chaotic systems are virtually impossible tomodel in order to make long-range predictions (as weatherforecasters and investors in the stock market will be quick to tellyou).Exercise 7.31. What types of experiment do you think it would be possible to carryout in each of the following situations? Describe the different stepsthat you would take in carrying out the experiment.1. You are a lecturer of art, and you wish to test during one of yourlectures how many names of artists the students can rememberafter having seen named pictures by them during the lecture.2. You have been given a collection of different types of clay brickfrom around the world, and you want to establish which type of
brick is affected by frost after it has been soaked in water. Youhave a tank of water and a large deep-freeze cabinet at yourdisposal. Some bricks are baked and others are not.3. You have designed a new type of bed, which you maintainmakes people sleep longer than other beds. You wish todemonstrate, by a scientific experiment, that this is the case.2. What kind of models could you use to demonstrate the operationswithin:1. the solar system2. central government3. a building4. the personal relationships within a family5. a business contract?3. In the following experiments and models, what sort of data will beneeded and what sort of data will be produced? Will it be possible touse the data produced to feed back into the experiment or model inorder to increase its precision?1. An experiment to test how long it takes to hard-boil differentsizes of freshly laid eggs.2. A physical model of a bed-sitting room with a model set ofstandard furniture, to investigate how many furniturearrangements are possible in the given space.3. A computer model that mimics the journeys of buses on a busroute in order to investigate when and why they tend to arrive atthe bus stop in threes.4. An experiment to test how the framing of glass affects thestrength of windows during impact.Some of the above examples of models and experiments are rathersimplistic in order to keep the answers manageable. However, even inactual research situations, one of the major problems is to try toreduce complex situations or phenomena to manageable formats inorder to model them or experiment with them. You must make well-argued and often difficult decisions about the choice of important orrelevant variables. Another aspect that you will always need toinvestigate is the availability of resources, both technical and financial.You might find it necessary to beg or borrow the necessary softwareor equipment!Figure 7.5 The essential qualities of a model are that it should beconstructed for a particular purpose
Analysing primary dataLittle sense can be made of a huge collection of data; therefore anessential part of research is the analysis of the data. This analysismust be carried out in relation to the research problem. You arewasting your time, and that of the eventual readers of your work, ifyou carry out analysis irrelevant to the aims of your study. You areprobably also wasting your time if you amass data that you areunable to analyse, either because you have too much, or becauseyou have insufficient or inappropriate analytical skills or methodsto make the analysis. We say ‘probably’, because research is nota linear process, so it is not easy to predict exactly how much datawill be ‘enough’. What will help you to judge the type and amountof data required is to decide on the methods you will use toanalyse them. In turn, the decision on the appropriateness of
analytical methods must be made in relation to the nature of theresearch problem and the specific aims of the research project.There are several reasons why we analyse data, some of whichare the same as the reasons for doing research. These are shownin Checklist 7.13.Checklist 7.13 Main reasons for analysing dataTo measureTo make comparisonsTo examine relationshipsTo forecastTo test hypothesesTo construct concepts and theoriesTo exploreTo controlTo explainA large variety of methods is employed to carry out theseprocesses. The more common are summarized below, togetherwith a mention of the above processes for which they may besuitable. You will notice that some methods can be used for morethan one process.A strong distinction is usually made between quantitative andqualitative analysis: many textbooks on research devote differentchapters to each, and whole books are written on the techniquesinvolved in each approach. The following descriptions aretherefore ordered in this way, although some of the analyticalmethods can be used both quantitatively and qualitatively. Theseare mentioned where appropriate.Quantitative analysisQuantitative analysis uses the syntax of mathematical operationsto investigate the properties of data.
We can express with numbers what is impossible to state inwords. You cannot pile up words and deduce an average fromthem. You cannot take the square root of a sentence. It isimpossible to square a word, a phrase or a paragraph. (Leedyand Ormrod, 2015, p. 99)Obviously, data that can be analysed in this fashion must bequantitative in nature. The levels of measurement used in thecollection of the data, i.e. nominal, ordinal, interval and ratio, arean important factor in choosing the type of analysis that isapplicable, as is the number of cases involved.StatisticsThere are two meanings to the word ‘statistics’. The first is:The science of collecting and analysing numerical data,especially in, or for, large quantities, and usually inferringproportions in a whole from proportions in a representativesample. (OEED, 1991)The second refers to any systematic collection or presentation ofsuch facts: for example, population statistics are records ofpopulation numbers and make-up. Statistical methods deal purelywith quantitative data, or with qualitative data that are expressedin numerical terms.As you well know by now, one of the primary purposes of scientificinvestigation is to discover relationships among phenomena inorder to explain, predict and possibly control their occurrence. It isin the discovery and quantification of these relationships thatstatistical methods are a valuable tool. We are talking here aboutcorrelational, rather than causal, relationships.Correlational techniques generally aim to answer three questionsabout two variables or two sets of data. Does a relationship existbetween the two variables or sets of data? If so, what is thedirection of the relationship? And what is its magnitude? A wide
range of techniques can be used, depending on the nature of thevariables being analysed, and they bear exotic names likeKruskal’s gamma, Kendall’s coefficient of concordance, Guttman’slambda, and the chi-square and Kolmogorov–Smirnov tests.However, there is no reason to be nervous about this for, asLeedy and Ormrod suggested, a simple definition of statisticsmight be:A language that, through its own special symbols andgrammar, takes the intangible facts of life and translates theminto comprehensible meaning. (2015, p. 156)You do not even have to be a mathematician to use this speciallanguage, as user-friendly computer packages (such as SPSS)will do all the calculations for you. However, you must be able tounderstand the function and applicability of the various tests toyour own sets of data. In this respect, Preece advised:For any researcher, the first rule of statistics is ‘Always consulta qualified statistician.’ The second rule is, ‘Know enoughabout statistics to be able to view the advice critically.’ (1994,p. 159)There is not space (or need) in this book to explain in detail therange of tests and their uses. It will help your understanding,though, if a description of the realm of statistics is provided, andthe various branches of the discipline are outlined. Diagram 7.5portrays the main classes of statistics and their characteristics,and will serve as a guide to the explanations that follow.Diagram 7.5 Types of statistics
Parametric statisticsThe two major classes of statistics are parametric and non-parametric statistics. An understanding of the meaning of aparameter, which in this context refers to a function of thepopulation, is essential in order to appreciate the differencebetween these two words. A parameter of a population is aconstant feature that it shares with other populations: a commonone is the ‘bell’ curve of the normal frequency distribution. Mostpopulations display a large number of more or less ‘average’cases with extreme cases tailing off at each end. For example,most people are of about average height, with those who areextremely tall or small being in a distinct minority. The distributionof people’s heights shown on a graph would take the form of thenormal or ‘Gaussian’ curve (Diagram 7.6). Although values varyfrom case to case, the generality of this type of curve amongstpopulations is so strong that statisticians take it as a constant – abasic parameter – on which the calculations of parametricstatistics are based. For those cases where this parameter isabsent, non-parametric statistics may be applicable.
GeneralityChecklist 7.14 Classes of parametric statisticsDescriptiveInferentialDiagram 7.6 Gaussian curveDescriptive statistics provide a method of quantifying thecharacteristics of the data, where their centre is, how broadly theyspread and how one aspect of the data relates to another aspectof the same data. The ‘centre of gravity’ of the data, their point ofcentral tendency, can be determined by finding the ‘mode’ or the‘median’ and any one of several ‘means’. These measures havetheir own characteristics and applications and should be chosenwith regard to the data being analysed.Descriptive statistics
The measure of the dispersion (or spread) of the data, how flat orsteep the Gaussian curve appears, is an indication of how muchof the data closely resembles the mean. The flatter the curve, thegreater is the amount of data that deviates from the mean, i.e. thefewer that are close to the average. The horizontal length of thecurve also gives an indication of the spread of values and theextent of the extremes represented in the data, while theoccurrence of a non-symmetrical curve indicates skewness in thedata values.Apart from examining the qualities of a single set of data, the mainpurpose of statistical analysis is to identify and quantifyrelationships between variables. This is the type of researchcalled correlation research. But remember, the mere discoveryand measurement of correlations are not sufficient on their own toprovide research answers. It is the interpretation of thesediscoveries which provides the valuable knowledge that will giveanswers to your research question.The technical term for the measure of correlation is thecoefficient of correlation. There are many types, thePearson’s r being the most common. It is possible to measure thecorrelation between more than two variables if you use theappropriate tests. However, one must be wary about assumingthat, because a strong statistical correlation between variablescan be demonstrated, there is necessarily a causal bond betweenthe variables. It may be purely chance or the influence of otherfactors apart from design that, say, leads to areas of high-densitydevelopment in cities having high crime rates. The researchermust carefully question the premises on which such a causalassertion is made, and review the facts to examine if suchcausality is verifiable in other ways.Coefficient of correlationInferential statistics goes beyond describing thecharacteristics of data and the examination of correlationsbetween variables. As the name implies, it is used to producepredictions through inference, based on the data analysed. What
is not so clear from its name is that inferential statistics is alsoused to test statistically based hypotheses.Inferential statisticsThe predictive role is limited to estimating population parametersfrom sample statistics. This is not as abstruse as it might at firstseem. Simply stated, this entails making predictions about thequalities of a total population on the basis of the qualities of asample. This is an exercise that is commonly carried out in qualitycontrol in production processes, where a sample of the productionis tested in order to estimate the qualities of the total production.Three parameters (qualities) are commonly estimated: centraltendency (proportion of products which are close to the norm (e.g.within permitted size tolerance); variability (e.g. range of sizesoccurring); and probability (e.g. the proportion of acceptableproducts produced). As with all predictions made from samples,the representative quality of the sample is crucial to accuracy.Figure 7.6 A sample of the production is tested in order toestimate the qualities of the total production
Two types of estimate can be made of population parameters fromsample statistics: point estimates and interval estimates. Pointestimates attempt to pinpoint the population parameter throughthe sample statistic value (e.g. the standard deviation andaverage size of a sample of manufactured components, giving anestimate of the standard deviation and average size of the wholeproduction run of the same component). While this produces aprecise estimate of the parameters of the population, the valuesare crucially dependent on the truly representative quality of thesample.Interval estimates of parameters use the sample statistics topredict the band within which almost all (typically 95% or 99%) ofthe values will lie. The expected range of the statistical values ofthe population is established, and if any values fall with asignificant frequency (i.e. more than 5% or 1%) outside this range,then this variability will be considered to be caused by somethingmore than mere chance. For example, the range of sizes of a
manufactured component is estimated from a sample to benormally within a certain tolerance, e.g. ± 1 mm, when theproduction machines are running correctly. If more than say 5% ofthe components are found to be beyond these sizes, then theproduction is being affected by some unusual influence other thanchance (e.g. a fault in the production machine). Obviously, thelarger and more representative the sample from which the samplestatistics were taken, the more likelihood that the prediction iscorrect.The statistically based hypothesis is commonly referred to as thenull hypothesis. Inferential statistics are used to test this typeof hypothesis. As a very simple example of the principles involvedusing the above case of the manufactured component, theengineer who designed the production machines couldhypothesize that, according to his/her design, the components willbe manufactured within the size tolerance ±1 mm. Assuming thatthe machines are properly built and working correctly, if samplestaken of the components produced show that 95% of thecomponents fall within the size tolerances (i.e. no significantdifference is observed between the predicted and the observedparameters: this is where the null comes from), then his/herhypothesis can be seen to be supported. If more samples aretaken with the same result, then more support is given. If,however, the sample shows that more than 5% of the componentsexceed the size tolerance, then there is some cause for thisbeyond mere chance. The engineer’s hypothesis will be rejected.For what reason his/her hypothesis was proved to be wrong isanother matter; perhaps his/her design was incorrect, orcalculations faulty.Null hypothesisNon-parametric statisticsNot all data are parametric, i.e. samples and populationssometimes do not behave in the form of a Gaussian curve. Datameasured by nominal and ordinal methods will not be organized ina curve form. Nominal data tend to be in the dichotomous form of
either/or (e.g. this is a cow or a sheep or neither), while ordinaldata can be displayed in the form of a set of steps (e.g. the first,second and third positions on a winners’ podium). Statistical testsbuilt around discovering the means, standard deviations, etc. ofthe typical characteristics of a Gaussian curve are clearlyinappropriate for analysing this type of data. Non-parametricstatistical tests have been devised to recognize the particularcharacteristics of non-curve data and to take into account thesesingular characteristics by specialized methods. In general, thesetypes of test are less sensitive and powerful than parametric tests;they need larger samples in order to generate the same level ofsignificance.Tests can be used to compare the qualities of two or more groupsor samples (e.g. Wilcoxon matched pairs signed rank test, signtest, Kruskal–Wallis test) or to analyse the rankings made bydifferent judges (Kendall coefficient of concordance, Spearmanrank order correlation), or to compare the data from observed andtheoretical sources (chi-square test). Detailed information aboutwhich tests to use for particular data sets can be obtained fromspecialized texts on statistics, and of course, your own expertstatistical adviser. This is perhaps a good place to warn you thatcomputer statistical packages (e.g. SPSS) will not distinguishbetween different types of parametric and non-parametric data.In order to avoid producing reams of impressive looking, thoughmeaningless, analytical output, it is up to you to ensure that the testsare appropriate for the type of data you have.Qualitative analysisUnlike the well-established statistical methods of analysingquantitative data, qualitative data analysis is still in its fairly earlystages. The certainties of mathematical formulae anddeterminable levels of probability are not applicable to the ‘soft’nature of qualitative data, which are inextricably bound up with
human feelings, attitudes and judgements. Unlike the largeamounts of data that are often collected for quantitative analysis,which can readily be managed with the available standardstatistical procedures conveniently incorporated in computerpackages, there are no such standard procedures for codifyingand analysing qualitative data. According to Turner (1994, pp.208–13), it is perhaps not even desirable that there should behomogeneous, classifiable types of qualitative analysis;researchers must structure their analytical approaches to fit thenature of the data with which they are faced.However, there are common steps that have to be made in allqualitative data analysis. Bryman and Burgess (1994, pp. 6–8)identified two general processes: the building of typologies andtaxonomies, and the generation of concepts and theory. Robsonand McCartan (2016, p. 469) went further to provide a list oftactics for drawing conclusions from qualitative data. These alsofall into the two processes above: counting, patterning, clusteringand factoring into the first; and relating variables, building causalnetworks and relating findings to general theoretical frameworksinto the second. Miles and Huberman (2013, p. 12) added a thirdcomponent in their definition of the components of analysis, thethree concurrent flows of activity being data reduction, datadisplay, and conclusion drawing and verification.The inclusion of data display is important. The unwieldy mass ofinformation that normally provides the basis for analysis, evenwhen coded, clustered, summarized, etc., cannot be easilyunderstood when presented as extended text. Information in textis dispersed, is sequential rather than concurrent, and is bulky anddifficult to structure. The mind is not adept at processing largeamounts of information, but prefers to simplify complexinformation into patterns and easily understood configurations.Using suitable methods to display the data (in the form ofmatrices, graphs, charts and networks) aids not only in thereduction and ordering of the data, but also in its analysis. Milesand Huberman advocate ‘more systematic, powerful displays andurge a more inventive, self-conscious, iterative stance toward theirgeneration and use’ (2013, p. 13).
Preliminary analysis during data collectionWhen conducting extended field research it is important tomaintain an analytical attitude to the type and amount of databeing collected, and to acknowledge the assumptions andhypotheses that brought you to this stage. It will be easier tostructure the information whilst the details are fresh in the mind, toidentify gaps and to allow new ideas and hypotheses to challengeyour assumptions and biases. Raw field notes, often scribbled andfull of abbreviations, and tapes of interviews or events, need to beprocessed in order to make them useful. Much information will belost if this task is left for weeks or months.The process of data reduction and analysis should be a sequentialand continuous procedure, simple in the beginning stages of thedata collection, and becoming more complex as the projectprogresses.To begin with, one-page summaries can be made of the results ofcontacts, e.g. phone conversations, visits. A standardized set ofheadings will prompt the ordering of the information: contactdetails, main issues, summary of information acquired, interestingissues raised, and new questions resulting from these. Similarone-page forms can be used to summarize the contents ofdocuments.Typologies, taxonomies and codingAs the data accumulate, a valuable step is to organize theshapeless mass of data by building typologies and taxonomies,that is, by identifying differences in the data and thereby formingsubgroups within the general category. Using these newtypologies helps to clarify the relationships among concepts.The development of a coding system is an important aspect offorming typologies, as it facilitates the organization of copiousdata (in the form of notes, observations, transcripts, documents,etc.) and provides a first step in conceptualization. Codes arelabels or tags used to allocate units of meaning to the collected
data. This helps to prevent ‘data overload’ resulting frommountains of unprocessed data in the form of ambiguous words.CodingMiles and Huberman (2013, p. 81) formulated four generalcategories of code: descriptive, interpretive, explanatory andastringent. More specifically, Lofland et al. (2005, pp. 14–15)devised a classification of ‘social phenomena’ on which it wouldbe possible to build a coding scheme. The six classes are shownin Checklist 7.15.Checklist 7.15 Classes of social phenomenaActsActivitiesMeaningsParticipationRelationshipsSettingsThe process is analytical, requiring you to review, select, interpretand summarize the information without distorting it. It is normal tocompile a set of codes before doing the fieldwork, based on thebackground study, and then to refine it during the data collection.Coding is rarely regarded by different researchers as exactly thesame kind of activity. However, two essentially different types ofcoding can be distinguished, one used for the retrieval of textsequences, the other devised for theory generation. The formerrefers to the process of cutting out and pasting sections of textfrom transcripts or notes under various headings. The latter is amore open coding system used as an index for ideas: reflectivenotes or memos, rather than merely bits of text.While the process of identifying segments of records for orderlycopying and filing under topics might be seen as an onerous,boring and essentially clerical activity, Richards and Richards(1994) maintain that actually this judgement obscures the fact that
much of the activity involved in coding for retrieval is a theorizingprocess. The analytical decision to create a particular code has adefining effect on what you will find in your collated data: ‘datacontrol methods must be processes of analysis, not merely of datadisposal’ (1994, p. 149).Coding is an important part of much qualitative research, and youwill have to refer to more specialist texts for details of itsdevelopment and application (see Miles and Huberman, 2013, pp.71–94; Strauss and Corbin, 2015; Strauss, 1987). Severalcomputer programs used for analysing qualitative data (such asEthnograph and NUD*IST) also have facilities for filing andretrieving coded information. They allow codes to be attached tothe numbered lines of notes or transcripts of interviews, and forthe source of the information/opinion to be noted. This enables arapid retrieval of selected information from the mass of materialcollected.Pattern coding, memoing and interim summaryThe next stage of analysis requires us to begin to look for patternsand themes, and explanations of why and how these occur. Thisrequires a method of pulling together the coded information intomore compact and meaningful groupings. Pattern coding can dothis by reducing the data into smaller analytical units such asthemes, causes/explanations, relationships among people andemerging concepts, to allow the researcher to develop a moreintegrated understanding of the situation studied. This willgenerally help to focus later fieldwork and lay the foundations forcross-case analysis in multi-case studies by identifying commonthemes and processes.Miles and Huberman (2013, pp. 70–1) describe three successiveways that pattern codes may be used, as shown in Checklist 7.16.Checklist 7.16 Finding patterns using codesThe newly developed codes are provisionally added to theexisting list of codes and checked out in the next set of fieldnotes to see whether they fit.
The most promising codes are written up in a memo (describedbelow) to clarify and explain the concept so that it can be relatedto other data and cases.The new pattern codes are tested out in the next round of datacollection.Generating pattern codes is surprisingly easy as it is the way by whichwe habitually process information. However, it is important not to clinguncritically onto initially developed patterns, but to test and developand, if necessary, reject them as your understanding of the datadevelops, and as new waves of data are produced.To develop a greater understanding of events and to make moresense of the relationships between codes, short analyticaldescriptions can be compiled based on the evolving ideas of theresearcher reacting to the data and the development of codes andpattern codes. Compiling memos is a good way to explore linksbetween data and to record and develop intuitions and ideas. Thiscan be done at any time – but is best done when the idea is fresh!Remember that memos are written for yourself, and so the lengthand style are not important, but it is necessary to label them sothat they can be easily sorted and retrieved. The activity ofmemoing should continue throughout the research project; youwill find that the ideas become more stable with time until a‘saturation’ point is achieved, i.e. the point where you are satisfiedwith your understanding and explanation of the data.MemosData collection in qualitative research is inherently more flexibleand less predictable than with quantitative research. Therefore itis essential, at probably about one-third of your way through thedata collection, to seek to reassure yourself and yoursupervisors/sponsors by checking the quantity and quality of whatyou have found out so far, your confidence in the reliability of thedata, and the presence and nature of any gaps or puzzles thathave been revealed, and to review what still needs to be collected
in relation to the time available. This exercise will result in theproduction of an interim summary, that is, a provisional report(10–25 pages are usually sufficient).Interim summaryThis report will be the first time that everything you know about acase will be summarized, and presents the first opportunity tomake cross-case analyses in multi-case studies and to reviewemergent explanatory variables. Remember, however, that thesummary is provisional, and perhaps will be sketchy andincomplete. It should be seen as a useful tool for reflecting on thework done, for discussion with your colleagues and supervisors,and for indicating any changes that might be needed in the codingand in the subsequent data collection work. In order to check onthe amount of data collected about each research question, a dataaccounting sheet can usefully be compiled. This is a table thatsets out the research questions and the amount of data collectedfrom the different informants, settings, situations, etc.Shortcomings can thereby easily be identified.Further techniques during data collectionThree other useful techniques for analysis during the datacollection period are described by Miles and Huberman (2013, pp.182–5). These are vignettes, pre-structured case analysis andsequential analysis. Some very brief remarks about these will beuseful.A vignette is a short, contextually rich narrative story that is usedto encapsulate a typical event or attitude in a straightforward,direct way. It results in a vivid, compelling and persuasiveinterpretation of an issue, an abstraction rather than arepresentation of an original event. It can be used in interimreports to formulate core issues and also in longer case reports inorder to highlight important subjects.Pre-structured case analysis is used to streamline data collectionand to produce good comparability between data sets from
different cases by collapsing the processes of data collection,analysis and report writing into one evolving procedure. Itassumes that the research questions are closely specified andthat the conceptual framework is well developed. The methodconsists of compiling a standard case report outline and datadisplay formats before data collection begins. The collectedinformation from each case can then be ‘slotted into’ the preparedframework. The case reports are filled in, edited and refined asthe information is collected, enabling the reports to be finalizedalmost as soon as the data collection is complete. In this way,coding can be applied to raw field notes and entered straight intothe selected displays, obviating the need to compile coded write-ups. Any shortage of information is easily detected, and can benoted for future collection. Because the process of analysis directfrom the field is difficult to illustrate, various forms of triangulationshould be used to verify the results obtained.Sequential analysis allows the researcher to review and deepenunderstanding of the findings at particular points during theresearch project, and to inform the data-gathering process in thenext stage. It helps to develop a progressively profound focus onevolving important issues as the data-gathering can become moretargeted on vital questions. Each session of sequential analysisshould involve discussion with your supervisor and criticalcolleagues, to help you to review and, if necessary, reorient youranalysis of the facts and theory development.Main analysis during and after data collectionThe traditional text-based report tends to be a lengthy andcumbersome method of presenting, analysing, interpreting andcommunicating the findings of a research project. Not only is itforced by its nature to present the evidence and argumentssequentially, but it also tends to be bulky and difficult to graspquickly because information is dispersed over many pages. Thispresents a problem for the writer, as well as for the final reader,who rarely has time to browse backwards and forwards throughmasses of text to gain full information, let alone compare a seriesof such reports. Graphical methods of data display and analysiscan largely overcome these problems and are useful for exploring
and describing as well as explaining and predicting phenomena.They can be used equally effectively for one-case and cross-caseanalysis (see Checklist 7.17; see Miles and Huberman, 2013, fora detailed guide to a wide range of displays).Checklist 7.17 Graphical displays fall into two categoriesMatricesNetworksMatricesThe two-dimensional arrangement of rows and columns inmatrices can summarize a substantial amount of information.They can easily be produced informally in a freehand fashion toexplore aspects of the data, to any size, and computer programsin the form of databases and spreadsheets can aid in theirproduction. Matrices can be used to record variables such astime, levels of measurement, roles, clusters, outcomes and effects(Diagram 7.7). Latest developments allow the formulation of three-dimensional matrices.MatricesNetworksA network is made up of blocks (nodes) connected by links.Maps and charts featuring networks can be produced in a widevariety of formats, each with the capability of displaying differenttypes of data. Flow charts are useful for studying processes orprocedures. Not only are they helpful for explanation, but theirdevelopment is a good device for creating understanding.Organization charts display relationships between variables andtheir nature, e.g. formal and informal hierarchies. Causal networksare used to examine and display the causal relationships betweenimportant independent and dependent variables, causes andeffects.
NetworkDiagram 7.7 Matrix example: measures taken to reduce the initialskill requirements of the self-builders, initiation and design phaseThese methods of displaying and analysing qualitative data areparticularly useful when you compare the results of several casestudies, as they permit a certain standardization of presentation,allowing comparisons to be made more easily across the cases.The information can be displayed in the form of text, codes,abbreviated notes, symbols, quotations or any other form thathelps to communicate compactly. The detail and sophistication ofthe display can vary depending on its function and on the amount
of information available; displays are useful at any stage in theresearch process. The different types of display can be describedby the way that information is ordered in them.Time-ordered displays record a sequence of events in relation totheir chronology. A simple example of this is a project programmegiving names, times and locations for tasks of different kinds. Thescale and precision of timing can be suited to the subject. Eventscan be of various types, e.g. tasks, critical events, experiences,stages in a programme, activities, decisions. Some examples oftypes of time-ordered displays are:Event listing. A chronological presentation of significant eventsin the various strands of a situation.Critical incident chart. Lays stress on the incidents that havebeen critical in the process.Activity record. A display that analyses steps in accomplishinga task.Decision modelling. Commonly used to analyse the steps in acourse of action or categorisation (see Diagram 7.8).Conceptually ordered displays. Concentrate on variables in theform of abstract concepts related to a theory and therelationships between them. Examples of such variables aremotives, attitudes, expertise, barriers and coping strategies.They can be shown as matrices or networks to illustratetaxonomies, content analysis, cognitive structures,relationships of cause and effect or influence. Here is aselection of different types:Cognitive map. Plots a person’s way of thinking about an issue.Effects matrix. Plots the observed effects of an action orintervention.Taxonomy tree diagram. Useful to break down concepts intotheir constituent parts or elements.Causal models. Used in theory building to provide a testableset of propositions about a complete set of variables withcausal and other relationships between them (see Diagram7.9).Role-ordered displays show people’s roles and their relationshipsin formal and informal organizations or groups. A role defines a
person’s standing and position by assessing their behaviour andexpectations within the group or organization. These may beconventionally recognized positions (e.g. judge, mother, machineoperator), or more abstract and situation dependent roles (e.g.motivator, objector). People in different roles tend to see situationsfrom different perspectives: a strike in a factory will be viewed verydifferently by the management and the workforce. A role-orderedmatrix will help to systematically display these differences or canbe used to investigate whether people in the same roles areunified in their views.Partially ordered displays are useful in analysing ‘messy’situations without trying to impose too much internal order onthem. For example, a context chart can be designed to show, inthe form of a network, the influences and pressures that bear onan individual from surrounding organizations and persons whenmaking a decision to act. This will help to understand why aparticular action was taken.Diagram 7.8 Network example: a decision tree model sorting outpublic road transport vehicles
Diagram 7.9 Network example: multiple cause diagram of rabbitcontrol systemExercise 7.41. Consider the following different types of data and state if youconsider them to be amenable to quantitative or qualitative analysis,or both.1. A collection of scripts taken from semi-structured interviewswith 150 prisoners, asking them about their attitudes to theregime of discipline in the prison.2. The results of a questionnaire sent to all UK solicitor practices,asking them about their fee income, insurance premiums,turnover and staffing levels.3. Observations noted of traffic movements at a busy trafficintersection during a period of one week.4. A pile of historic documents found in an attic, consisting ofletters, diaries and financial accounts, all to do with the buildingof an eighteenth-century stately country home.5. A survey of government statistics about homelessnessthroughout the twentieth century.2. What is the essential difference between parametric and non-parametric statistics?3. Could you use inferential statistics to prove that smoking causescancer? If you think you could, how; and if not, why not?
4. Qualitative analysis appears to be more of a creative art than ascience. Do you think that this is true? State your reasons foragreeing or disagreeing in two or three sentences. What are three orfour essential differences in method which distinguish it fromquantitative analysis?5. Describe which form of display may be suitable to usefully show thefollowing:1. The management structure of a government department.2. The attitudes of parents in different social classes to privateschooling and state schooling.3. To explain what one has to do to change the wheel of a car.4. The results of introducing into a factory a new pay systembased purely on productivity.Qualitative analysis of texts,documents and discourseSources in the form of texts and documents provide a great dealof data about society, both historically and in the present. There isa wide range of analytical methods that can be applied to analysisof the subtleties of text. Both quantitative and qualitative optionsare available. Checklist 7.18 provides a list of the main qualitativemethods, followed by a brief summary of their characteristics.Checklist 7.18 Types of qualitative analysis of texts etc.Interrogative insertionProblem–solution discourseMembership categorizationRhetorical analysisNarrative analysisSemioticsDiscourse analysisInterrogative insertionThis method consists of devising and inserting implied questionsinto a text for which the text provides the answers. In this way, you
can uncover the logic (or lack of it) of the discourse and thedirection and emphasis of the argument as made by the author.This helps to uncover the recipient design of the text – how thetext is written to appeal to a particular audience and how it tries tocommunicate a particular message.Problem–solution discourseThis is a further development of interrogative insertion whichinvestigates the implications of statements more closely. Moststatements can be read to have one of two implications. This isvery commonly found in advertising, e.g. ‘Feeling tired? Eat aMars Bar’. The same, but in more extended form, is found inreports, instruction manuals and books. The analysis aims touncover the sequence of the argument by following through firstthe situation, then the problem, followed by the response and theresult and evaluation. The absence of any of these steps in thesequence will lead to a sense of incompleteness and lack oflogical argument. Another way of presenting the analysis is todevise a network in the form of a decision tree that traces theproblems and the possible solutions with their implications.Membership categorizationThis technique analyses the way people, both writers and readers,perceive commonly held views on social organization, how peopleare expected to behave, how they relate to each other and whatthey do in different social situations – for example, therelationships between parents and their children, behaviour ofmembers of different classes of society, or the roles of differentpeople in formal situations. Most of these assumptions are notmade explicit in the text. By highlighting what is regarded asnormal, assumptions and pre-judgements may be revealed andan understanding of typical characterization can be gained.Rhetorical analysisRhetoric is the use of language and argument to persuade thelistener or reader to believe the author. For example, politicians try
to give the impression that they should be believed, and harnessthe vocabulary and structure of spoken and written language tobolster this impression. Rhetorical analysis uncovers thetechniques used in this kind of communication. Rhetoric is used toaim at a particular audience or readership. It may appeal to, andengender, belief in the target audience, but is likely to repel andundermine the confidence of others. For example, a racist diatribewill encourage certain elements on the far-right but repel others.The analysis detects credibility markers – signals that indicate thatthe ‘rightness’ of the author and the ‘wrongness’ of others, suchas assertions about the ‘correct’ moral position, claims ofprivileged understanding and dismissal of alternatives asunbelievable. Even in apparently non-partisan writing, such asscientific reports, where the author is de-personalized, rhetoricaltechniques are used to persuade the reader about the ‘rightness’of the conclusions. Here one should check for markers relating toobjectivity, logic and correct methodology.Narrative analysisThis form of analysis is aimed at extracting themes, structures,interactions and performances from stories or accounts thatpeople use to explain their past, their present situation or theirinterpretations of events. The data, which are primarily aural, arecollected by semi- or unstructured interviews, participantobservation or other undirected methods. The narrative isanalysed for different aspects, such as what is said rather thanhow, or conversely, the nature of the performance during thetelling, and perhaps how the storyteller reacted with thelistener(s). Alternatively, the structure of the story is inspected. Allthis is done in order to reveal the undercurrents that may liebeneath the simple narrative of the story.SemioticsThis is the term for the ‘science of signs’ which is used to examinevisual and other media as well as written texts. Semioticsattempts to gain a deep understanding of meanings by the
interpretation of single elements of text or visual units. Words areonly meaningful in their relationship with other words, e.g. we onlyknow the meaning of ‘horse’ if we can compare it with differentanimals with different features. Likewise, the meanings of objectsare related to their context, e.g. the meanings of a red traffic lightcan be seen as embedded in the system of traffic laws, colourpsychology, codes of conduct and convention etc. (which couldexplain why for a time in China a red traffic light meant ‘go’). Astrong distinction is therefore made between denotation (what weperceive) and connotation (what we read into it) when analysing asign.SemioticsA range of technical terms has been devised that indicate thedifferent aspects of signs, for example: ‘signifier’ – that whichperforms as a vehicle for the meaning; ‘signified’ – what thesignifier points to; ‘denotative meaning’ – the obvious functionalelement of the sign; ‘code’ or ‘sign system’ – the generalizedmean instilled in a sign; and many others. This can be explainedin a simple way by looking at a traffic sign and attributing theseterms to the various parts, as is shown in Diagram 7.10.Diagram 7.10 The semiotics of a traffic signDiscourse analysisThis studies the way that people communicate with each otherthrough language within a social setting. Language is not a neutralmedium for transmitting information; it is embedded in our social
situation and helps to create and recreate it. Language shapesour perception of the world, our attitudes and identities. Twocentral themes can be identified: the interpretive context in whichthe discourse is set, and the rhetorical organization of thediscourse. The former concentrates on analysing the socialcontext, for example the power relations between the speakers(perhaps due to age or seniority) or the type of occasion wherethe discourse takes place (a private meeting or at a party). Thelatter investigates the style and scheme of the argument in thediscourse, for example a sermon will aim to convince the listenerin a very different way to a lawyer’s presentation in court.Post-structuralist social theory, and particularly the work of theFrench theorist Michel Foucault, has been influential in thedevelopment of this analytical approach to language. According toFoucault, discourses are ‘practices that systematically form theobjects of which they speak’ (1972, p. 43). He could thusdemonstrate how discourse is used to make social regulation andcontrol appear natural.Mixed methods and interdisciplinaryresearchMixed methodsMany subjects, particularly those concerning human beings, canbenefit by being studied from both a quantitative and a qualitativeapproach. This is where a mixed method approach isrecommended.The strict distinction between quantitative/positivist andqualitative/relativist research methods is neat and simplistic, butrather goes against our normal experience of life, where quantityand quality are usually considered to be different butcomplementary aspects of the same phenomenon. That is why itis now quite common to use a variety of methods when doing aresearch project. It was not always so. Since the 1970s therehave been many objections to taking a mixed methods approach
due to the claimed incompatibility of the relationship betweenparadigm, methodology, epistemology and ethics relating toqualitative and quantitative research, and the issues are stillrumbling on in some areas (Denzin, 2010).It is now generally accepted that the distinction between themethods used for quantitative and qualitative research are not asclearly separated as previously believed, that many data collectionmethods and data types can be used for both approaches andthat the paradigms behind them can coexist in a single study(Symonds and Gorard, 2008). Using a combination of methods isseen as a third option to using either quantitative or qualitativeresearch. The combination of the two ways of collecting andanalysing data (using the concept of triangulation – despite beingonly two!) can produce more perspectives on the subject, whichenables you to make stronger arguments for your interpretation ofthe data, and allows greater depth of understanding and a truerpicture of the object of study. There are now numerous books,papers and conferences dedicated to exploring this third option(see ‘Further reading’ at the end of this chapter).Although there are numerous definitions of mixed methodsresearch, they all point out that using both quantitative andqualitative methods in a connected way to address a researchproblem will help compensate for the weaknesses of the singlemethods and increase the likelihood of producing richer, morereliable, more meaningful and more useful answers to theresearch questions.How to use mixed methods researchMixed methods research can be used for all the common typesof research investigations, such as explaining and interpreting asituation, exploring a phenomenon, assessing new systems orinstallations and testing theoretical ideas. But how do you selectfrom the whole range of research methods available for mixingand matching to produce a suitable combination for your projectthat makes sense in the context of your research? Here is achecklist, based on Yanchar and Williams (2006) on what youneed to consider when choosing your types of research.
Mixed methodsChecklist 7.19 What you need to consider when choosing your mixedmethodsTailor your choice of research methods to the demands of theparticular research questions and their context.Be imaginative in the construction of theories, formulation ofquestions and problem-solving, using a creative combination ofsuitable research methods rather than adopting some unsuitablestandardized formula.Acknowledge that all aspects of your work are based onphilosophical standpoints, that no research subject is historicallyunburdened and lacking in individual theoretical bases.Select methods that are compatible in order to producemeaningful and interpretable accounts, and which are based onan overall theoretical purpose that avoids contradiction and self-refutation.Identify and evaluate the consequences of assumptions andvalues that underlie the research questions and chosen researchmethods.How you mix the methods can be done in different ways. The firstquestion to ask is what the balance between the quantitative andqualitative elements of the project will be. Will the qualitativemethods dominate, with the addition of quantitative methodswhere appropriate? Or will equal status be given to the qualitativeand quantitative research, working in a complementary fashion?Or will the quantitative approach dominate, whilst recognizing thatsome qualitative aspects are usefully studied? The decision willdepend on the context and nature of the subject that you will bestudying and how familiar you are with using quantitative and/orqualitative data.Once you have decided the balance, you should then consider theorder in which you use the methods and how they interrelate withand inform each other. But it is not only the balance of these twoperspectives that you need to decide on, it is also the strategiesthat you will employ in their use. The following checklist based on
Clark and Creswell (2008) provides a useful guide to the differentpossible combinations and how they relate.Checklist 7.20 Six ways to combine the different research methods1. First collect and analyse quantitative data, and then usequalitative methods to further explain and interpret the resultsand promote generalizability.Example: Use a questionnaire survey, gather quantitativedata on your chosen subject, and on the basis of theanalysis, delve deeper into interesting aspects by holding aseries of interviews with respondents to obtain qualitativedata.2. Start with qualitative methods and use the results to set up thequantitative part of the study.Example: Hold a series of discussion groups to discuss aparticular issue, and on the basis of the qualitative findingsasses the prevalence of the views in a wider sample througha questionnaire survey.3. Use both quantitative and qualitative methods one after theother, and only when the results are achieved, integrate andinterpret in a complementary fashion, making use of thestrengths of each for different aspects of the study.Example: Carry out a questionnaire survey to collectquantitative data about several variables so as to analyse forcorrelations between them. Later, hold open-endedinterviews to examine the same variables to assess theopinions relating those variables and their perceivedconnections. A comparison can then be made between thestatistical outcomes and the perceptions of people in thesample.4. Deliberately use the two approaches equally at the same time inorder to cross-check (i.e. triangulate) the various findings, and tooptimize the data collection and analysis tasks.Example: Collect and analyse quantitative data collectedfrom a particular situation and at the same time explore thequalitative aspects to obtain additional insights into the samesituation in order to provide a rounded view and richer resultsfrom the investigation.5. Give priority to either the qualitative or quantitative approach,which then guides the project, and the remaining approach isembedded within the process in order to answer specificquestions.Example: Use quantitative data in order to solve a technicalissue and supplement this with some consumer preferences.
6. Use the quantitative and/or qualitative methods according to thenature of different aspects of the research project and then bringtogether the results to form overall conclusions.Example: Find answers to the sub-questions of a researchproject using the appropriate types of methods, and thencombine the findings to answer the main question.Data management and mixingUsing mixed methods is obviously going to be more complicatedand labour intensive than concentrating on only one method, butthe advantages gained from the richer resource for analysis andresultant greater understanding of the issues makes the extrawork worthwhile. More organizational skills will be required tomanage the complexities, but there is a virtue in you gaininggreater experience and competence in a wider range of researchmethods.You will have to clearly explain why you have chosen yourparticular mix and sequence and what advantages this has over amore simplistic approach. Each step needs to be described, andhow the data and results are integrated to provide the overallconclusions. You also need to assess any possible problems as aresult of your choice (both practical and theoretical) and how youwill overcome them. Your aim should be to convince the reader ofthe validity of your approach and the robustness of theconclusions you can draw from your planned work.Interdisciplinary researchInterdisciplinary research should not be confused with mixedmethods. In many subject areas, such as sustainability, the builtenvironment, art, healthcare, business and transport, there will beaspects that involve different disciplines, the most commonincluding economics, politics, geography, social science,technology and aesthetics. Each of these will require you to befamiliar with the theoretical background, intellectual and technicaldemands of the discipline. These can be radically different, whichposes problems for when you want to combine them in one study.
So you will have to work out how the different types of approachcan be linked, reconciled or even merged. However, note thatthere is a distinction between interdisciplinarity andmultidisciplinarity, where the former seeks to integrate and orsynthesize perspectives from several disciplines, and the latterworks through cooperation between the disciplines, but eachworking within their own standard frameworks (Barry, Born andWeszkalnys, 2008). It is no doubt easier to combine thedisciplines that lie closer together than those with completelydifferent skill sets.As an interdisciplinary researcher you will need to have an open-minded attitude that embraces new ideas, be flexible, open tochange and creative, and be good at team working, with awillingness to listen and ability to communicate. A high toleranceto ambiguity is also required so as to keep options open to allownew explorations across boundaries (Tait and Lyall, 2007).It is best to start interdisciplinary projects with an open-endedpreliminary research phase in order to explore the scope of theproject and to determine possible boundaries. This is the time todecide which disciplines are most relevant to the questions raised.Checklist 7.21, based on Tait and Lyall (2007), reminds you whatthe outcomes of this exploratory phase should be.Checklist 7.21 Outcomes of the explanatory phaseA specification of the range of issues that are central to theresearch problemHow they interact with one another to create or sustain theproblemHow these interactions can be modified to deliver animplementable, synergistic solutionIf you are part of a research team you will have to make extraefforts to promote cohesion by combining your expertise andovercoming communication problems. The divergent worldviewsof different disciplines will also need to be resolved, and effortsmade to coordinate institutional structures and procedures. This isbest achieved by devising a management structure that can cope
with these differences and to ensure that the project is a success.A description of this preliminary research phase should beincluded in the research proposal to provide evidence that acohesive strategy has been adopted.There are several advantages of doing interdisciplinary work, suchas crossing boundaries to find novel solutions, benefiting from thedifferent perspectives of the team members and using theircomplementary skills, and nurturing a more holistic view of theissues that can result in more rounded and effective results. It isalso personally stimulating to work with people from otherdisciplines and promotes academic understanding and creativity.It is quite obvious that you can learn a lot about the different methodsof data analysis, but you should ask yourself how much you reallyneed to know in detail in order to define, at this stage, what you will bedoing in your research project. The point of learning about them is todiscover the range of available methods and the situations where theyare appropriately used. At the proposal stage you rarely need to gointo great detail about how you will apply the methods. The importantexception to this is when you are applying to do funded research. Inthis case you will have to describe quite precisely how you will collectand analyse data, and how you will manage the process in terms ofboth staffing and timing. When you have chosen the appropriatemethods from reading this chapter, then you will have to go to moredetailed textbooks to learn about the specifics of those selected.ConclusionsThe three perspectives of research with which we began thisdiscussion, i.e. data collection/analysis, qualitative/quantitativeresearch and research strategies, interlace in a three-way matrix.This complex combination provides a large number of choices ofresearch methods from which you, as a prospective researcher,must make a selection appropriate to your research problem.It is evident that, even within one research project, there may beseveral research methods that could be justifiably applied to
different aspects of the same research problem. Each enables adifferent aspect of the problem to be investigated and analysed.Data collection and analysis may not just happen once: there isoften a repeated sequence of collection and analysis, for example:collection of historical data > analysis > hypothesis for casestudy > data collection> analysisThis chapter can only be a very brief introduction to some of themore common methods available. In the course of yourbackground research it will be necessary for you to investigatemore thoroughly the particular methods that have been used incontexts similar to your own study, and to evaluate them in orderto select those that are appropriate. Detailed study can then bemade of the selected methods.The next steps: which researchmethods will you use?The aims of this section are for you:to determine the type of data you should collect in order toinvestigate your research problemto decide on the appropriate methods for collecting the datato investigate appropriate methods of data analysisto examine what can be achieved by using these techniques.Checklist of activities that will progress yourresearchStep 1: Quantitative or qualitative researchor mixed methods?Review the sort of data you will be collecting and the type of analysis youwill want to subject them to. Do they fall into one or other category, or do
they contain elements of both? It is important for you to recognize thedifference between the two types of research because their fundamentalassumptions are different. How can you recognize the difference? Try thetests in the next steps.Step 2: Test for quantitative researchTwo features distinguish quantitative research: the use of precisemeasurement, and the use of quantities expressed as numbers capable ofbeing analysed by statistical methods or other computational methods.Once you start counting how many, measuring how much, you are doingquantitative research!Step 3: Test for qualitative researchQualitative research is any research that does not conform to thedescription in step 2 (well, almost). Qualitative research usually is to dowith people and their activities, and is concerned with variables thatcannot be quantified on an interval or ratio level. Data in the form of words,either written or spoken, are a common feature, but data could also be ofsounds or actions (e.g. music or dance).Step 4: Devise your research strategyReturn to Box 7.1 to help define your research strategy. Decide the typesof question you will be asking. You might decide that the list of strategies isnot sufficiently complete (see Chapter 1) or you might add your own,based on another source. Also, you might use more than one in order tocover your topic.Step 5: Link it all upMake a chart or a diagram of the different threads to your research,considering the different perspectives you will be using to study them. Usethe three perspectives shown in Diagram 7.1, i.e. strategies, datacollection/analysis, and quantitative/qualitative approaches. Note that insome instances you can collect qualitative data and analyse themquantitatively, and vice versa. This exercise will help you to consider thetype of actions you will need to take in your research project.Step 6: Collecting data
Make a list of all the types of data that you will need to collect in order toanswer your research questions. Try to order them in categories ofsecondary and primary data. Then decide where you might find these data– the data source. In the case of secondary data, will you be able to testtheir authenticity? In the case of primary data, will you need to select asample? In both cases, will it be possible for you to get access to therequired data?Step 7: What sort of sample?If you do not need to select a sample, you can leave out this step. First,consider what your population is and, if possible, what are itscharacteristics (e.g. homogeneous, stratified, etc.). Next, select anappropriate sampling method. As you can see, there are a lot of factorsapart from accurate representation that will influence your choice, such asmanageable numbers, locations, accessibility, practicality, timing, etc. Atthis stage you do not have to be too precise about your methods (unlessyou are doing a proposal for a PhD direct), but you will need to indicatethat you are aware of the issues involved.Step 8: Choose your data collectionmethodsReflect on your list of data required from step 6. For each item on the list,decide on the best method to use to acquire the data. You might end upwith a variety of methods. Consider the practical consequences. Now is agood time to assess the practicality of what you propose.Step 9: Choose your data analysis methodsThe methods of data analysis you select will depend on the nature of thedata and what you want to find out from them. Consider the list of reasonsin Checklist 7.13 for analysing data. Your background reading will help youto ascertain what methods have been used successfully in previous,similar research. Again, practicality will be an important factor, e.g. will youneed expensive equipment or special computer resources?Step 10: Combining data collection andanalysisPerhaps the research into your subject does not divide so neatly into twoseparate sequences of data collection and analysis. If so, formulate anappropriate strategy for combining these activities. Normally, the collection
and analysis still use different methods, but these are alternated so as toprovide mutual feedback at regular intervals. You will have to convinceyour supervisor that your planned activities are both appropriate andpracticable.Step 11: Draft your plan of workA reality check! Now that you are clearer about what you want to do andhow you will do it, the question of whether you can fit it into the timeavailable needs to be answered. A simple bar chart showing a list ofactivities with their time allowance set within the overall time available is allthat is required here. Do not forget to add in time taken for learning newtechniques, writing up, holidays, travel, conferences, etc. This will enableyou to check that you are not being unrealistic in your ambitions. Askyourself: is the time allocated to each task reasonable? If not, now is thetime to make adjustments. This might entail reducing the scope of theresearch, cutting down on case studies, limiting the size and complexity ofsurveys, simplifying experiments, and reducing the amount of analysis tothe bare essentials.Consolidation and assessmentWhen you are ready, arrange a tutorial with your tutor orsupervisor. The main point of this consultation is to discuss theresearch methods you have selected to use in your researchproject, and to explain why you think they are appropriate andwhat you hope to achieve by using them. Use the information youhave gained from this chapter on methods of data collection andanalysis and different research strategies to explain how yourresearch problem can be tackled in a realistic way.You should explore how the research methods might be practicallycarried out, their scope and level of complexity, while consideringwhat resources will be needed to produce and analyse the data.Consider particularly the relationships between collection andanalysis of data: can the former be completed before commencingthe latter, or will they have to be carried out in tandem?Contemplate whether you will need to select samples during yourresearch and, if so, what will be the important considerations inyour case. Surmise how using these particular methods will
reinforce your argument, and discuss the possible advantagesand disadvantages of alternative approaches.Further readingThere are hundreds of books about data collection methods andsampling techniques, and about all the ways to analyse data. Formore choice, look up key words such as sampling, questionnaires,survey research, etc. in your library catalogue. In addition, thereare examples of previous theses done in your subject, which youcan consult for the nature of the data collected and their use ofmethods.The first set of books listed here are general research methodsbooks that take what you have read in this chapter to a muchgreater depth of treatment. They are useful to learn more aboutthe methods that you have selected as being suitable for yourresearch project. If necessary, you can then go on to even morespecialist texts, a selection of which I add after the general books.Robson, C. and McCartan (2016) Real World Research: AResource for Social Scientists and Practitioner-Researchers,4th edn. Oxford: Blackwell.This is a first-class book written in a practical and accessible style,and is a valuable resource for anyone embarking on a researchproject. It covers all the main research methods in considerabledetail.Bryman, A. (2015) Social Research Methods, 5th edn.Oxford: Oxford University Press.One of the most popular standard handbooks on researchmethods for social research students. Comprehensive and clearlywritten.Cohen, L. and Manion, L. (2017) Research Methods inEducation, 8th edn. London: Routledge.
Superb range and detail, and of interest to more than justeducationalists.Gill, J. and Johnson, P. (2010) Research Methods forManagers, 4th edn. London: Sage.Provides a step-by-step guide to implementing particularmethodologies, while simultaneously encouraging a strongawareness of philosophical assumptions.And for the ultimate in research methods resources, go to SAGEResearch Methods Online (SRMO) (http://srmo.sagepub.com).SRMO contains over 100,000 pages of book, journal andreference content on research methods, as well as a speciallydesigned map and taxonomy of the methodological field. You willneed to register on the home page to get free access to the site.Still in development, SRMO is expected to be an essential tool forresearchers, academics and students.Secondary data collection and analysisLargan, C. and Morris, T. (2019) Qualitative SecondaryResearch: A Step-By-Step Guide. London: Sage.Clear and easy to understand, this book is a ready-made toolkitfor successfully using qualitative secondary data. It provides anaccessible introduction to the theory of secondary research andsets out the advantages and limitations of using this kind ofresearch. It walks you through each stage of the researchprocess, provides practical, ethical tools to help you with yourproject and shows you how to avoid the potential pitfalls of usingsecondary data.Stewart, D. and Kamins, M. (1993) Secondary ResearchInformation Sources and Methods, 2nd edn. ThousandOaks, CA: Sage.A useful guide to finding secondary sources, obtaining thesources, and evaluating and integrating the information to answer
specific research questions.Heaton, J. (2004) Reworking Qualitative Data: ThePossibility of Secondary Analysis. London: Sage.Provides an emphasis on the re-use of available qualitative data.Kiecolt, J. and Nathan, L. (1985) Secondary Analysis ofSurvey Data: A Sage University Paper. Newbury Park, CA:Sage.This presents strategies for locating survey data and provides acomprehensive guide to US social science data archives,describing several major data files. The book also reviewsresearch designs for secondary analysis.Sampling and data collectionI have pointed out the sections on data, primary data collection,sampling, etc. in some general methods books which are usefulbecause of their brevity. Look at the table of contents in the othersfor the relevant sections, which cover other aspects of theresearch process too.Seale, C. (ed.) (2012) Researching Society and Culture,4th edn. London: Sage.See Chapters 9 and 10 on ‘Sampling’ and ‘Questionnaires andInterviews’.Holliday, A. (2016) Doing and Writing Qualitative Research,3rd edn. London: Sage.Consult Chapters 4 and 5 for what counts as data and writingabout data.Fowler, F.J. (2009) Survey Research Methods, 4th edn.London: Sage.
This book goes into great detail about all aspects of the subject ofdoing surveys. Good on sampling, response rates, methods ofdata collection – particularly questionnaires and interviews. Use itselectively to find out more about the particular methods you want.This book will also be useful later for analysis, and has a sectionon ethics too.Aldridge, A. (2001) Surveying the Social World: Principlesand Practice in Survey Research. Buckingham: OpenUniversity Press.Another comprehensive book; find what you need by using thetable of contents and index.Greenfield, T. (2016) Research Methods forPostgraduates, 3rd edn. London: Arnold.A wide range of authors give tips on all aspects of postgraduateresearch. Easy to navigate as the chapters are nicely divided intodiscrete topics.Fink, A. (1995) The Survey Kit. London: Sage.Nine volumes covering all aspects of survey research! This mustbe the ultimate.Here are some books specifically on questionnaires, in order ofusefulness:Peterson, R.A. (2000) Constructing EffectiveQuestionnaires. London: Sage.Gillham, W.E. and William, E.C. (2000) Developing aQuestionnaire. London: Continuum.Dillman, D.A. (2000) Mail and Internet Surveys: TheTailored Design Method, 2nd edn. Chichester: Wiley.
Frazer, L. (2001) Questionnaire Design and Administration:A Practical Guide. Chichester: Wiley.And a few on interviewing, again in order of usefulness at yourstage of work:Keats, D.M. (2000) Interviewing: A Practical Guide forStudents and Professionals. Buckingham: OpenUniversity Press.Jaber, F. (ed.) (2002) Handbook of Interview Research:Context and Method. London: Sage.Wengraf, T. (2001) Qualitative Research Interviewing:Biographic, Narrative and Semi-Structured. London: Sage.And a couple on sampling:Schaeffer, R.L. (2005) Elementary Survey Sampling, 6thedn. Belmont, CA: Duxbury.Fink, A. (1995) How to Sample in Surveys (The SurveyKit, vol. 6). London: Sage.And a couple on case studies, simplest first:Nisbet, J.D. and Watt, J. (1982) Case Study. Rediguideno. 26. Oxford: TRC Rediguides.Yin, R.K. (2018) Case Study Research: Design andMethods, 6th edn. Thousand Oaks, CA: Sage.Quantitative researchFor a straightforward introduction to statistics, see:
Foster, L., Diamond, I. and Jeffries, J. (2014) BeginningStatistics: An Introduction for Social Scientists. London:Sage.This book emphasizes description, examples, graphs and displaysrather than statistical formulae. A good guide to understanding thebasic ideas of statistics.For a comprehensive review of the subject see the following. Ihave put these in order of complexity, simplest first. The list couldgo on for pages with ever-increasing abstruseness. You could alsohave a browse through what is available on your library shelves tosee if there are some simple guides there.Wright, D.B. and London, K. (2009) First Steps inStatistics, 2nd edn. London: Sage.Kerr, A., Hall, H. and Kozub, S. (2002) Doing Statisticswith SPSS. London: Sage.Byrne, D. (2002) Interpreting Quantitative Data. London:Sage.Bryman, A. and Cramer, D. (2011) Quantitative DataAnalysis with IBM SPSS 17, 18 and 19: A Guide for SocialScientists. London: Routledge.Siegel, S. and Castellan, N.J. (1988) NonparametricStatistics for the Behavioral Sciences, 2nd edn. New York:McGraw-Hill.Seale, C. (ed.) (2012) Researching Society and Culture,4nd edn. London: Sage.For a good guide on how to do historical and documentaryresearch, look at Chapter 16 by Ben Gidley.
Qualitative analysisAs you would expect with this big and complex subject, there is amyriad of books dedicated to explaining all aspects. In the listbelow I have tried to explain a bit about each book and how it maybe of use to you. I have ordered them in what I think is going fromsimplest to most sophisticated.Flick, U. (2019) An Introduction to Qualitative Research,6th edn. London: Sage.Useful sections on analysing verbal and visual data, with practicaladvice on documentation, coding, interpretation and analysis. Thewell-organized layout will enable you to be selective in picking outwhat is relevant to you.For a really comprehensive though incredibly dense and rathertechnical guide to qualitative data analysis, refer to:Miles, M.B. and Huberman, A.M. (2013) Qualitative DataAnalysis: An Expanded Sourcebook, 3rd edn. London:Sage.This has a lot of examples of displays that help to explain howthey work, but is technically sophisticated so you might find itdifficult initially to understand the terminology in the examples.Once you get into it, it could provide you with valuable techniquesfor managing and analysing text-based data.And a few more books if you don’t find what you want in theabove. Your library catalogue will list many more. Try a searchusing key words, such as data analysis, with management,education or whatever your particular subject is, to see if there arespecific books dedicated to your particular interest.Silverman, D. (2015) Interpreting Qualitative Data:Methods for Analysing Talk, Text and Interaction, 5th edn.London: Sage.
Holliday, A. (2016) Doing and Writing Qualitative Research,3rd edn. London: Sage.A general guide to writing qualitative research aimed at studentsof sociology, applied linguistics, management and education.Schwandt, T. (2015) The Sage Dictionary of QualitativeEnquiry, 4th edn. Thousand Oaks, CA: Sage.To help you understand all the technical jargon.Coffey, A. and Atkinson, P. (1996) Making Sense ofQualitative Data: Complementary Research Strategies.London: Sage.Modelling and experimentationAnother huge subject with a huge choice of books. I have dividedthe list below into different sections.Here are some books on modelling, a subject that soon getsimmersed in mathematical formulae, though there are computerpackages that take much of the sweat out of calculations. I haveordered them in levels of increasing detail and complexity. Look inyour library catalogue to see if there are any devoted to your ownsubject.Marker, D. (2002) Model Theory: An Introduction. London:Springer.Edwards, D. (2001) Guide to Mathematical Modelling, 2ndedn. Basingstoke: Palgrave.Edwards, D. (2000) Introduction to Graphical Modelling,2nd edn. New York: Springer.Examples of modelling applied to specific subjects:
Oppenheim, N. (1980) Applied Models in Urban andRegional Analysis. London: Prentice-Hall.Hedman, J. (2002) IT and Business Models: Concepts andTheories. Malmö: Liber Ekonomi.Samuel, A.E. (1999) Introduction to Engineering Design:Modelling, Synthesis and Problem Solving. Oxford:Butterworth–Heinemann.Some books dedicated to experimental methods. Again, check outwhat is available under your own subject headings.Lewis-Beck, M.S. (ed.) (1993) Experimental Design andMethods. London: Sage.Dean, A., Voss, D. and Draguljić, D (2017) Design andAnalysis of Experiments, 2nd edn. New York: Springer.Montgomery, Douglas C. (2017) Design and Analysis ofExperiments, 9th edn. New York: Wiley.Wheeler, A.J. and Ganji, A.R. (2010) Introduction toEngineering Experimentation, 3rd edn. London: Pearson.For a simple explanation about algorithms see the first bookbelow, but for more technical guides on computing, see theothers.Broadbent, G. (1988) Design in Architecture: Architectureand the Human Sciences. London: Fulton, p. 326.Bateman, R. and Paterson-Stephens, I. (2002) The DSPHandbook: Algorithms, Application and DesignTechniques. Harlow: Prentice-Hall.
Harel, D. and Feldman, Y. (2004) Algorithmics: The Spiritof Computing, 3rd edn. Wokingham: Addison-Wesley.InternetQuinton, S. and Reynolds, R. (2018) UnderstandingResearch in the Digital Age. London: Sage.This book is split into three main sections: considering how thedigital environment has changed and how we do social scienceresearch; the different elements of digital data and how socialscience researchers can evaluate these elements; and practicaladvice on how to conduct research in the digital age.Here are three more good books about using the Internet to doresearch, particularly for doing surveys. They discuss the viabilityof the Internet as a research tool, the equipment required, how todesign a survey, and what can go wrong. They also provide somecase studies as examples.Hewson, C., Yuyle, P., Laurent, D. and Vogel, C. (2016)Internet Research Methods: A Practical Guide for the Socialand Behavioural Sciences, 2nd edn. London: Sage.Fielding, N., Lee, R. and Blank, G. (2017) The SAGEHandbook of Online Research Methods, 2nd edn. London:Sage.Mann, C. and Stewart, F. (2000) Internet Communicationand Qualitative Research: A Handbook for ResearchingOnline. London: Sage.Mixed methods and interdisciplinary researchThere are a growing number of books dedicated to explainingmixed methods research. Here is a selection that will provide youwith more information.
DeCuir-Gunby, J. and Schutz, P. (2017) Developing aMixed Methods Proposal: A Practical Guide for BeginningResearchers, Volume 5 in Mixed Methods ResearchSeries. Thousand Oaks, CA: Sage.This is a really useful, hands-on guide that helps you to create amixed methods research proposal for your dissertation, grant orgeneral research studies. There are plenty of examples fromactual studies and a complete proposal at the end of the book. Ifyou complete the exercises in each chapter you will have anindividualized, detailed template for your own mixed methodsresearch proposal.Creswell, J.W. (2014) A Concise Introduction to MixedMethods Research. London: Sage.This book provides a brief overview of mixed methods researchthat takes you through the essential steps in planning anddesigning a study. It offers students and practitioners in the social,behavioural and health sciences a foundation for understandingmixed methods methodology.Tashakkori, A. and Teddlie, C. (eds) (2010) SageHandbook of Mixed Methods in Social & BehavioralResearch, 2nd edn. Thousand Oaks, CA: Sage.This is a big book that examines all aspects of mixed methodsresearch across a variety of disciplines. It covers all aspects ofenquiry from conceptualizing research to selecting methods,analysing and interpreting data, and reporting findings. It alsoprovides learning objectives, discussion questions and exercises,and extensive cross-referencing to help you in your studies.And here are some books aimed at specific disciplines:Watkins, D. and Gioia, D. (2015) Mixed MethodsResearch. Pocket Guides to Social Work ResearchMethods Series. New York: Oxford University Press.
This practical guide to mixed methods research has been writtenwith health and human services professionals in mind. It reviewsthe fundamentals of mixed methods research designs and thegeneral suppositions of mixed methods procedures, looks criticallyat mixed method studies and models that have already beenemployed in social work, and reflects on the contributions of thiswork to the field.Curry, L., and Nunez-Smith, M. (2014) Mixed Methods inHealth Sciences Research: A Practical Primer, Vol. 1.Thousand Oaks, CA: Sage.This book pulls together foundational mixed methods principles,synthesizes the knowledge base in the field, and translates it for ahealth science researchers. It presents key theories, concepts andapproaches in an accessible way and is packed with illustrationsfrom the health sciences literature.Mehdi Riazi, A. (2017) Mixed Methods Research inLanguage Teaching and Learning. Sheffield: EquinoxPublishing.This volume is the first to examine MMR in language teaching andlearning and how such a methodology works in practice.Roelen, K. and Camfield, L. (eds) (2015) Mixed MethodsResearch in Poverty and Vulnerability: Sharing Ideas andLearning Lessons. Basingstoke: Palgrave Macmillan.This volume shares experiences from research in developed anddeveloping country contexts on how mixed methods approachescan make research more credible, usable and responsive tocomplexity.And one on interdisciplinarity:Repko, A.F., Szostak, R. and Buchberger, M.P. (2017)Introduction to Interdisciplinary Studies, 2nd edn.Thousand Oaks, CA: Sage.
This book shows a different way of thinking that can lead to newsolutions for complex real-world problems. It breaks down theprocess of interdisciplinary studies into manageable andunderstandable parts and is illustrated with plenty of clearexamples, interesting analogies, and detailed deconstruction ofinterdisciplinarity.
8 The Research Proposal: Planningand Structuring Your IdeasAimsTo explain the necessary ingredients of a research proposalTo set up the definition of your research aims and actionsTo explain what makes a good proposalTo provide a step-by-step guide to writing your successfulproposalIntroductionIn this chapter, we will concentrate on writing your researchproposal, which will provide a foundation for your research project.You will need to use all the understanding you have acquired andthe skills you have developed in the previous chapters in order toaccomplish this task. You will have to clarify in your own mindexactly what your specific area of study will be, and explain it in acompact and logical form so that anyone who reads your proposalwill be able to understand the character and value of yourintended work. Because submitting your proposal is a part of theformal process of attaining a research degree or applying forproject funding, you will have to abide by the relevant university orfunding body’s regulations with regard to the suitability of yourproposed research and the form of the research proposal. Thischapter will help you to achieve this.You will notice that the exercises in this chapter aim to help you toexamine aspects of your own work in preparation for writing yourproposal. There are obviously no right or wrong answers to thesequestions, so the answers at the end of the book tend either toprovide a list of points for you to check against what you havewritten, or to urge you to consult with your tutor or supervisor todiscuss the outcome of the exercise.
Of course, gaining approval for your intended work as describedin your proposal is only the beginning of the process of research.The main effort will be in carrying out the research. Although thisbook is not intended to be a practical guide to help you to do this,it does conclude with some useful hints as to how to approach theproblem of writing and presenting your thesis or report. You will nodoubt be faced with many new situations during your subsequentwork, but the preparatory work which you have done duringreading this book should help you to beat a path through thecomplex and often frustrating process of completing your researchproject.Defining your research projectHow to formulate your overall research problem, or area ofresearch, has been discussed in Chapter 1. Deciding like this onthe general issue to be addressed is the first step in narrowingdown the range of literature that you need to review in order toinvestigate further. The steps to take in order to achieve athorough review are detailed in Chapter 5. But it is useful to knowsome techniques to help you find and formulate the precisesubject and components for your research project.Let’s look at the way that you can consider the broad picture andthen break this down into the elements of a practical researchproject. Consider Diagram 8.1.Diagram 8.1 Levels of abstractionI have drawn parallels between the different levels of abstractionof language and those of your research components. A theory is acombination of several concepts, linked to make some kind of
assertion, e.g. being wealthy makes you happy. This is parallel inabstraction to a main question, e.g. Does being wealthy make youhappy? You can use these concepts to formulate the title of yourresearch, e.g. Wealth and happiness. In practice, it is best to bemore precise in order to limit the scope of the research. So youcould add some limitations such as time, location, situation, e.g.Wealth and happiness of selected billionaires in 21st centuryLondon.From what you have discovered from the literature, and your owninterests and beliefs, you should be able to formulate several mainresearch questions on different hot topics at this level ofabstraction, as a subject for discussion with your supervisor, and arange of subjects from which you can make a choice of what youprefer to follow up.Once you have formulated your main research question you canadvance to the next step towards the practical aspects of doingthe research by deriving sub-questions from the main question.This is done by separating out the concepts contained in the mainquestion and formulating questions around each. For example,What are the theories about wealth? How can happiness bedefined? How would one find billionaires in London?The next step is to consider what data types can be used asindicators of the concepts involved. For example, what sort of datashould one collect that indicates that someone is happy? Thisshould be apparent from the ways that happiness is defined, e.g.an appraisal of life satisfaction (such as of quality of life),subjective well-being (such as the experience of joy, contentmentor positive well-being), combined with a sense that one’s life isgood, meaningful and worthwhile (Lyubomirsky, 2007).Staying with the indicators of the concept of happiness, the nextstep is to find out how the various indicators can be measured, i.e.how the different types of data such as positive well-being can bedetected and fitted to some kind of a scale. In this case you couldsimply ask respondents how much they experience positive well-being – an ordinal level of measurement, from not at all to verymuch.
Then, the final stage of the process would be the actual collectionof data so that you have a set of measurements for eachrespondent. Obviously this can only be done when you areactually carrying out the field research.In summary, here is a chart that shows the cascade of decisionsto be made following the selection of your main research question(see Diagram 8.2).In order to be fully aware of what is required of you if you aredoing a research degree, it is essential that you carefully readyour university research degree regulations, a copy of which youshould have received when you first registered. If you are doing amajor assignment or a project, follow the detailed guidelines inyour assignment or project instructions. If you are preparing aproposal for a funding application, then read the detailedinstructions prepared by the funding body given to you with theapplication forms.Diagram 8.2 Cascade of decisionsThe recipe for a successful researchproposalProposals for research tend to follow a defined pattern. They allhave these features in common: an explanation, in a compact andprecise fashion, of the nature of the research; why it is needed;how it will be done; the likely outcomes; and, in most cases,exactly what resources are required to carry it out. According toLocke, Spirduso and Silverman (2007, pp. 3–6), there are threemain functions of a proposal, as shown in Checklist 8.1.Checklist 8.1 Main functions of a research proposal
A means of communication from the researcher to those who willassess, approve and possibly fund the work.A plan for action to describe the scope, aims, step-by-stepprocedures and expected outcomes of the work.A contract that will form the basis of agreement between theparties involved, e.g. the researcher, supervisors, funders,institutions (e.g. university, college, workplace, funders). Thesigned agreement cannot be substantially altered without theagreement of all the parties.There are, of course, differences in the nature of research projectsand this will affect the exact nature of the proposal. However, keep inmind that any proposal is a type of promise, so it is advisable not to‘promise mountains and deliver molehills’!Types of research proposalThe types of proposal that this book will help you to prepare forare:Undergraduate research projectsMaster’s dissertationsMajor projects or dissertations that are the culmination ofmany postgraduate degrees and diplomasMPhil and PhD research degreesPractice-based researchBusiness researchFunded research projects of all kindsProposals vary in length and complexity according to the level ofstudy. It is worth pointing out at the outset the differences betweenthese types of research and how they are reflected in theproposal. For student academic research, the major emphasis ison the practicality of the scale and the relevance of the subject ofthe project according to the parameters of the course. Studentacademic research at any level lower than PhD is basically anexercise in critically investigating and evaluating an approvedtopic and demonstrating an understanding of research methods
appropriate to the chosen field, and to presenting the outcomes inthe form of a coherent argument to the satisfaction of theexaminers.For an undergraduate project, just a brief description of the aimsand objectives together with a basic depiction of the context and asimple explanation of how the objectives will be achieved willsuffice. For postgraduate projects, there must be greater detailboth in the exploration of the context, in the form of a literaturereview, and of the description of the research methods that will beemployed, together with greater precision in the definition of theresearch problem and the expected outcomes of the research.MPhil and PhD proposals make very similar demands, though atPhD level, there is the added factor that the research will result inan independent and original contribution to knowledge. Normally,students undertaking a PhD direct, as opposed to an MPhil/PhD,will have already completed an MPhil or a master’s degree, orhave demonstrated substantial expertise and previous research inthe chosen subject. Those doing a PhD by published worksshould demonstrate that they are already experts in the subjectand have a strong record of peer reviewed publication.For practitioner and other funded research, including academicresearch, these issues are also important but equally so is thequestion of how the research will be financed and managed. Inaddition, an argument has to be made to convince the funder ofthe need for the research and its likely contribution to practice orknowledge in the field. Description of the forms of disseminationand estimates of the impact of the research should also beprovided. There is usually a requirement to supply the CVs ofthose involved in the research and to demonstrate theircompetence to do the work. A detailed plan of action with aprogramme of deliverables should also be included. Furthermore,most larger funded research projects will require the particulars ofa team of collaborators from other organizations and expressionsof support from influential sources. For practice-based research,the situation in the workplace will probably have a big influence onthe nature of the proposal according to the specific requirementsof the organization.
In all cases, ethical issues are important, but can be critically so infunded research where the project involves vulnerable peoplesuch as children, the elderly and the sick.The main ingredients and sequenceStudent academic research proposals are usually composed ofthe elements listed in Checklist 8.2. This sequence is usual, butsmall variations are possible. For funded research the addition ofa cost plan, portrayal of personnel, timetable and specifications ofoutputs and description of methods of dissemination are required.Checklist 8.2 Sections of a student research proposalThe titleAims of the researchBackground, literature reviewResearch problemDefinition of research objectivesOutline of methodsPossible outcomesList of referencesIt is a good idea, at the first attempt, to write a preliminary draft ofthe proposal while concentrating your attention exclusively on theorderly sequence of statements and ignoring grammatical andstylistic considerations. Once you have achieved a soundstructure you can spend time refining your composition withoutdanger of a subsequent major reconstruction.It helps to look at proposals written by other students orresearchers in the past, but be careful to analyse them for theirform rather than content so as not to be influenced by theirparticular subject or problem. Some examples are given later inthis chapter.The title
The function of the title is to encapsulate in a few words theessence of the research. Although, in academic proposals, youare not committed to using the same title in the thesis (theemphasis of the study may alter during the work), it shouldaccurately reflect your intentions at this stage.As one of the main purposes of the title is an aid to retrieval, ittypically contains all the essential key words that someone mightuse in an attempt to locate the kind of study you are proposing.These words are likely to include the main variables, the type ofresearch tasks, the theoretical basis and the purpose of the study.In the title ‘The implications of the 1998 Labour and Conservativepolicies on the economics and management of day-care centres’,the independent variables are Labour and Conservative policies,and the dependent variables are economics and management;the type of research task is implied as a comparative study; thetheoretical basis is one of prediction of effects of policies; and thepurpose of the study is to predict their effects on day-care centres.As important as words that indicate the subject are those thatdelineate it, for example in time, place and type. A good title oftenhas a structure that progresses from the general to the specific,for example ‘Access and servicing of city centre shops inregeneration areas in the UK’. A shorter title, such as ‘City centreshopping problems’, which might describe a similar study, soundsfar too vague and lacks the key words. However, unnecessarilylong titles should be avoided: two lines should be enough. Youcan leave out such phrases as ‘an investigation into’, ‘a study of’,‘aspects of’, as these are obvious attributes of a research project.Exercise 8.1Write a list of key words that summarize the main aspects of your intendedwork. Don’t be afraid at first of writing too many. In order to reduce your listyou can eliminate those that are repeated or non-essential, or possiblycombine words to form more general concepts. When you have reducedthe list to the essential words, form a sentence with them to create a titlethat appears to satisfy both technical and aesthetic standards. Check inthe answers at the end of the chapter to see if you have included all theessential relevant points.
Aims of the researchThis serves as an introduction to the heart of the project. It shouldbe possible to state the aims of the research in two or threesentences. If you cannot do this, you have probably not thoughtsufficiently about what you are actually going to do. It should beevident that the aims are delineated sufficiently so that it ispossible to achieve them with the available resources and time.Exercise 8.2Write a list of the aims of your research. Think of as many as you can thatyou believe are relevant to your project. When you have done this,consider each one carefully by asking the questions:1. How will the aim be achieved – methods, resources, skills, time? Is itrealistically possible to achieve it?2. What results are required for it to be achieved?3. Is the aim central to your study?4. Are there any overlaps between the aims? Keep them discrete.5. Is there a sequence or hierarchy which links one aim to another? Ifso, are they in the correct order?6. Are there too many aims to be realistically achievable?Make notes answering questions 1, 2 and 3 about each aim on your list.The background and literature reviewIt is necessary to explain, to the reader of your proposal, thecontext from which the research problem emerges. You should beable to demonstrate that you are aware of the major factors thatsurround your problem, and of any significant literature relating toit. It is also the function of this introduction to capture the interestof the reader.Some of the factors that make up the context might be of aphysical nature, such as building types (e.g. schools, prisons),materials (e.g. historical artefacts, household products), situations(e.g. the playground, the street, the countryside) or organizations(e.g. local government, secret society, local community), while
others might be more conceptual, such as the economy,legislation, development policy, etc. More abstract are theoreticalconcepts such as power, poverty and Marxism. In order to keepthe description as short as possible, use references to relevantliterature which, if known or followed up by the reader, give a fullaccount of the concept or situation mentioned. It is quite difficult topitch the level of your text so that any intelligent readerunderstands the factors from which your problem emerges, whileat the same time persuading the expert that you are conversantwith details of the principal issues. Do not assume that the readerknows anything about your subject.The efficient use of references is crucial in keeping the proposalshort. By quoting the references to publications you can refer towhole theoretical positions, approaches and movements in just afew words. In addition to this, the inclusion of relevant literaturedemonstrates your knowledge of the significant personalities andpositions in your subject.There should be a flow in the text of the background section whichreads in the form of a progressive account, where every issueraised can be seen to contribute to an argument which leadsinexorably to the research problem. This is the case in all types ofresearch proposal. However, for MPhil and MPhil with latertransfer to PhD, and in initial studies and exploratory projects, it isrecognized that, subsequent to the approval of the proposal, muchof the initial research work consists of a deeper investigation intothe relevant literature and nature of the research problem. But inapplications for limited dissertations, PhD direct, transfer fromMPhil to PhD and many funded research projects, it is expectedthat the background has already been thoroughly researched andit is from the conclusions of this work that the research problemhas been precisely formulated.The research problemThe research problem must be the focus of the proposal and, infact, of the whole research project. The nature of researchproblems has already been covered in some detail (seeparticularly Chapter 1). It is the culmination of the background
work and the initiator of the specific research tasks. It must bevery clear from the text of the proposal what the nature of theproblem is, how it was identified, and why it is a significantproblem which needs to be researched It is also essential that youcheck carefully that the description of your problem complies withthe required university regulations, course requirements orconditions of the funding body, and with the relevant form whichdefines how your application should be set out.The problem may be expressed in abstract terms initially, butthrough the statement of sub-problems you should indicate how itcan be investigated practically. You might, particularly whenapplying to do PhD direct, transferring from MPhil to PhD stagesor when applying to do funded research state the exacthypotheses or questions which will direct the original research.Although the problem may be expressed in abstract terms, it is amistake to express it in general terms. To limit the scope of theresearch to what is practicable, the problem must be veryprecisely defined and delimited. It will help in this respect if youcarefully define your terms and state the assumptions you aremaking.How do you express your research problem?There is often a quandary about how to express the researchproblem in an appropriate and precise fashion. As the researchproblem is the sole focus of the research on which everythinghangs, it is important to express it in a way that makes theresultant planned work flow naturally from it. There is a range ofoptions, as shown in Checklist 8.3.Checklist 8.3 Ways to express a research problemA question or questionsA hypothesisPropositionsA statement of intent
The foundation of the hypothetico-deductive approach is thehypothesis, so it is important to know what makes goodhypotheses and how they can be formulated. You might find, afteryou understand what exactly a hypothesis is, that it is notappropriate in your research topic to use one (or several): it mightnot be possible to summarize your research problem sosuccinctly. There are less precise methods of expressing theresearch problem in a researchable form, so these are describedsubsequently.Hypotheses are nothing unusual; we make them all the time. Theyare hunches or reasonable guesses made in the form ofstatements about a cause or situation. If something happens inour everyday life, we tend to suggest a reason for its occurrenceby making rational guesses. For example, if the car does not startin the morning, we might hypothesize that the petrol tank wasempty, or that the battery was flat. For each hypothesis, aparticular action taken could support or reject it. If the petrol gaugeindicated ‘full’, then the hypothesis of an empty petrol tank couldbe rejected, and so on. When a particular hypothesis is found tobe supported, we have got a good chance that we can take theright action to remedy the situation. If, for example, wehypothesized that a wire to the starter motor had become loose,and then find such a loose wire, fixing the wire back might result inthe car starting again. If this was not the result, further hypotheseswould be needed to suggest additional faults. Although theseexamples may seem banal, many of the greatest discoveries inscience were based on hypotheses: Newton’s theory of gravity,Einstein’s general theory of relativity and a host of others.In your research, you will encounter hypotheses in yourbackground reading, sometimes overt and clearly stated, and atother times, in less scholarly documents, hidden in the text or onlyhinted at. If you use one in your own research study, a hypothesisshould arise naturally from the research problem, and shouldappear to the reader to be reasonable and sound. There are twogrounds on which a hypothesis may be justified: logical andempirical. Logical justification is developed from arguments basedon concepts and theories and premises relating directly to the
research problem; empirical justification is based on reference toother research found in the literature.Question or questionsThe method of investigating the problem may be expressedthrough asking a question or a series of questions, the answers towhich require scrutiny of the problem from one or more directions.This is a very direct and open-ended way of formulating yourinvestigations. Your aim is to provide some answers to thequestions. It is your judgement, and that of the examiner’swhether your answers are sufficient and based on enoughevidence. Box 8.1 shows an example of this form of presentation.Box 8.1 Research question and sub-questionsThe subject of this dissertation is ‘Representation of contemporary artin the media’.The main research question is ‘How is contemporary art representedin the media?’Three interrelated research sub-questions are raised. Thesequestions are:1. What are the characteristics of the representation ofcontemporary art issues in the media?2. How does this representation differ in coverage presented indifferent types of media, e.g. television, magazines,newspapers?3. What role do specialist journalists, and specifically artcorrespondents play in shaping this representation?Obviously, the question or questions should be derived directly fromthe research problem, and give a clear indication of the subject to beinvestigated and imply the methods that will be used. As above, theform of the questions can be a main question, divided into sub-questions that explore aspects of the main question. The mainquestion is very general – you could probably devise other sub-questions to explore different aspects of this question. But by being sospecific in your choice, you can limit your research to only thoseissues that you think are important, or that you have interest inpursuing.
HypothesisThe use of hypotheses is the foundation of the hypothetico-deductive approach to research, so it is important to know whatmakes good hypotheses and how they can be formulated. Whenused in a rigorous scientific fashion, there are quite strict rules tofollow. According to Kerlinger (1999), there are important qualitiesof hypotheses which distinguish them from other forms ofstatement; these are as shown in Checklist 8.4. Note howKerlinger stresses the need for utmost precision in the use oflanguage and a strict delineation of the statements.Checklist 8.4 Important qualities of hypothesesHypotheses:are assertions (not suggestions)are limited in scopeare statements about the relationships between certain variablescontain clear implications for testing the relationshipsare compatible with current knowledgeare expressed as economically as possible using correctterminology.The objective of this method is either to reject the hypothesis byfinding evidence that contradicts it, or to support it (you will not beable to prove it) by presenting evidence that underlines it. It mightalso be possible to modify the hypothesis in the light of what youhave found out.In order to formulate a useful researchable hypothesis, you needto have a good knowledge of the background to the subject andthe nature of the problem or issue that you are addressing. Agood hypothesis is a very useful aid to organizing the researcheffort. It specifically limits the enquiry to the interaction of certainvariables; it suggests the methods appropriate for collecting,analysing and interpreting the data; and the resultant confirmationor rejection of the hypothesis through empirical or experimentaltesting gives a clear indication of the extent of knowledge gained.
You need to formulate the general hypothesis on a conceptuallevel, in order to enable the results of the research to begeneralized beyond the specific conditions of the particular study.This is equivalent to the general research question. Then, younormally need to break down the main hypothesis into two ormore sub-hypotheses. These represent components or aspects ofthe main hypothesis and together should add up to its totality andare equivalent to the sub-questions. It is one of the fundamentalcriteria of a hypothesis that it is testable. However, a hypothesisformulated on a conceptual level cannot be directly tested; it is tooabstract. It is therefore necessary to convert it to an operationallevel. This is called operationalization. Each sub-hypothesis willsuggest a different method of testing and therefore impliesdifferent research methods that might be appropriate. Theoperationalization of the sub-hypotheses follows four steps in theprogression from the most abstract to the most concreteexpressions by defining in turn the concepts, indicators, variablesand values (see Diagram 8.1):Concepts: the building blocks of the hypothesis, which areusually abstract and cannot be directly measuredIndicators: phenomena that point to the existence of theconceptsVariables: the components of the indicators that can bemeasuredValues: the actual units or methods of measurement of thevariablesNote that each concept may have several indicators, eachindicator several variables and each variable several values. Toclarify these terms, consider this operationalization, which givesonly one example of each expression:Concept: povertyIndicator: poor living conditionsVariable: provision of sanitary facilitiesValues: numbers of people per bathroom, WCEach sub-hypothesis will intimate a different method of testing andtherefore implies different research methods that might be
appropriate.Although the term hypothesis is used with many different meanings ineveryday and even academic situations, it is advisable to use it in yourresearch only in its strictest scientific sense. This will avoid you beingcriticized for sloppy, imprecise use of terminology. If your researchproblem does not lend itself to being formulated in a hypothesis, donot worry, there are plenty of alternatives, many of which involve acompletely different research approach to that of the hypothetical-deductive method.PropositionFocusing a research study on a proposition, rather than on ahypothesis, allows the study to concentrate on particularrelationships between events, without having to comply with therigorous characteristics required of hypotheses. Consider thefollowing example.The title of the research was ‘Public sector housing for youngsingle people’. The main research problem was formulated in theform of three interrelated propositions:1. Specifically designed public sector housing provided foryoung single people to rent has been, and continues to be,designed according to the recommendations and standards inthe design guidance for single persons’ housing.2. The relevant design guidance is not based on accurateperceptions of the characteristics of young single people.From these two propositions the third one should follow:3. There is a mismatch between the specifically designed publicsector housing provided for single young people and theiraccommodation requirements.Statement of intent
Not all research needs to answer a question or to test ahypothesis. Especially at undergraduate dissertation or in smallerresearch studies, a more exploratory approach may be used.You can express the subject and scope of the exploration in astatement of intent. Again, this must be derived from the researchproblem, and imply a method of approach and indicate theoutcome. Here are four examples of this form of researchdefinition:1. The intention of this study is to identify the main aspects ofrecent developments in government organization andprocedures for local taxation in Switzerland and Bulgaria, andthen to assess the extent to which, or whether, features of theSwiss organization and implementation of local taxation canbe adopted to improve the system of the administration oflocal taxation in Bulgaria.2. This study examines the problems in career development ofwomen lawyers in the British legal establishment. It focuseson the identification of specific barriers (establishedconventions, prejudices, procedures, career paths) andexplores the effectiveness of specific initiatives that havebeen aimed at breaking down these barriers.3. In this study it is intended to consider whether relevant USstandards for food safety as applied to hotels and restaurantscould be transferred to hotels and restaurants in Nigeria.4. This thesis provides a reassessment of EngelbertHumperdinck’s early career and travels up to 1878. It aims toexplore the musical foundation of Humperdinck’s career andanalyse his early compositions in order to provide a basis fora more objective reassessment of his music.Exercise 8.3To prepare the argument which leads up to the statement of your researchproblem, make a list of the issues you will address in your account of thecontext. The issues should be summarized in a few words, as you areconcerned only with the skeletal structure of your proposal at this stage.Put them in an ordered sequence so that it is possible to progress fromone issue to the next in a logical fashion. Keep the list brief, but make sure
that it is complete (i.e. covers all the essential issues). You will probablynot get the order right at the first attempt, so do not hesitate to experimentwith different sequences with the help of the word processor or by writingeach issue on a separate slip of paper and then rearranging the slips.Write out your research problem in one sentence at the end of the list (youmay elaborate on it later if necessary). Check that in your sequence theissues all relate to the problem and lead logically towards it. If there aregaps in the logic of the argument, add linking issues. If the argument leadsto a different problem, either change the problem or re-examine the issuesyou have listed, and their order.When you think that the argument is cogent, you can put some ‘flesh’ onthe ‘bones’ in the text by making full sentences and adding references.Definition of research objectivesWhen a research problem has been identified, in order to indicatewhat measures will be taken to investigate the problem or providemeans of overcoming it, it is necessary to formulate a definition ofthe research objectives. This should be accompanied by someindication of how the research objectives will be achieved. Anexample is given in Box 8.2.Box 8.2 Research objectives: exampleIt is proposed to provide an adequate assessment of the relationshipbetween the design of security systems in computer networks and theresulting restrictions on the wide accessibility of these systems to thegeneral public. The research problem previously highlighted a lack ofsuch methods of assessment. To overcome this problem it isnecessary to:1. Propose a method of measurement by which the extent ofincorporation of security systems can be assessed. This willenable an objective comparison to be made between alternativedesign proposals in terms of the extent of incorporation ofsecurity features. There is a need to identify and categorize themain security systems advocated in past studies of publiclyaccessible computer networks in order to establish the generalapplicability of the methods of measurement proposed.2. Propose a method of measurement by which the extent of publicaccessibility to computer networks can be assessed. This willenable an objective comparison to be made between computer
networks in terms of the extent of their accessibility in use. Inorder to arrive at a method of measurement, a morecomprehensive interpretation of accessibility needs to bedeveloped so that the measures proposed will not be confined toany one particular computer system type.3. Assess the extent of accessibility achieved after theincorporation of security systems, by a study of actual computernetworks in use. To achieve this a number of publicly accessiblenetworks need to be examined.Outline of methodsThis part of the proposal explains briefly what you are going to doin order to carry out your research, based on your chosenresearch approach. In proposals for funded research this mayneed to be described in great detail, so that the funder can knowexactly how you will gather the data and analyse it in order toproduce the required outputs. The practicality of the methods willthen also become apparent. Every proposal will be different in itsdescription of methods. However, one common feature is likely tobe a review of the literature relevant to the research topic. Otherreview activities may be through consultation with experts andspecialists in the subject, and perhaps with people involved inother ways. It is generally relevant to distinguish between themethods of data collection and data analysis, although in somecases, particularly in qualitative research, these may go hand inhand.It may be important that you indicate how you will be able toaccess certain types of information. If there are obvious problemsin access, describe how you will go about solving them. If youhave privileged access to obscure or restricted information, thisshould be reported, as should the fact that you have resources tocover the cost of research in your own country, or if relevant,abroad.It is the convention to explain the methods to be used in the formof a list, without numbering the items. This enables you to be veryconcise. A very detailed account of methodology is not generallyrequired: you should limit yourself to an indication of the range of
methods required and the general scale of the procedures andwhy you are using those methods. Always answer the questions:which method, how will you carry it out, and why? The answer tothe question ‘why’ is often left out, with disastrous results. Use thelist in Checklist 8.5 to check for the inclusion of relevant tasks.Checklist 8.5 Tasks for outline of methodsLiterature search and critical analysisConsultation with expertsIdentification of research population(s) or situationsSampling (if relevant) – size of sample(s), location of sample(s),number of case studiesData collection methods – questionnaires, interviews, study ofdocuments, observations, etc.Analytical methods – quantitative, qualitative and combination ofbothEvaluation of results of analysisIn the case of an MPhil/PhD application, some of these methodswill be used in both the MPhil and the PhD stages of the work.With the PhD direct and most funded research projects, you mayhave to describe separately the methods used to investigate eachsub-problem, or sub-hypothesis, as these are likely to varyconsiderably.Possible outcomes of the researchSince the proposal is a type of contract to deliver certain results, itis a mistake to ‘promise mountains and deliver molehills’.Although you cannot predict exactly what the outcomes will be (ifyou could, there would be little point in carrying out the research),you should try to be quite precise as to the nature and scope ofthe outcomes and as to who might benefit from the information.Make sure that the outcomes relate directly to the aims of theresearch you described at the beginning of the proposal. You canbe more general when describing the overall significance of theoutcomes, when you make the point quite directly about theimportance of what you are proposing to do in your chosensubject. Note that in PhD proposals and most funded research
there is a need to indicate what will be the original contribution toknowledge. These outcomes may be a contribution at a practicaland/or a theoretical level.Figure 8.1 If you have privileged access to obscure or restrictedinformation, this should be reportedExercise 8.4Prepare a list of methods suitable for your own research, going from thegeneral to the particular, i.e. from background research to detailedresearch and evaluation. Check that they read in a logical order and that,although brief, they are informative enough to indicate types of methodappropriate to the subject researched.Culminate your list of methods with a statement that lists the possibleoutcomes of the research and another that specifies why, and how, theintended research is important.
Additional information for proposals for fundedresearchIn addition to the information required in a research proposal foran academic degree, a proposal to do funded research requiresfurther information covering costs, personnel, timetabling, outputsand other issues. It is important, when preparing such a proposal,that you argue a case for awarding the money and demonstratethat the researchers are likely to make good use of the funds. Youshould take care to apply only to funders who are likely to beinterested in your work, or to organizations whose terms ofreference specifically cover your research topic. Nearly allresearch funders provide clear information on their specific areasof interest. It is a waste of your time to attempt to gain theirbacking for research into subjects that they are unable or unwillingto support.CostsWhen seeking funding, a clear and comprehensive account of allthe costs that will be incurred is obviously of vital importance.There are many categories of costs; these are often listed in theapplication forms issued by funding bodies. The main categoriesare: personnel remuneration, travel and subsistence, capitalequipment (e.g. computers, printers, measuring instruments andexperimental equipment), overheads (e.g. universities charge foruse of their facilities), consumables (e.g. stationery, telephone).Your own university or organization will probably have clearbudgeting guidelines and will be able to provide help in setting upa spreadsheet if necessary.PersonnelThe funders are interested in the competence of those proposingto do the research, and of those whom they wish to employ tohelp them, in order to be assured that the team will be capable ofcarrying out what they propose. CVs of the main members of theresearch team will usually be required. It is important to get theright balance of managers and workers, and the requisite skills
and capabilities. Some funders insist on a partnership betweenacademic and industrial/commercial organizations. In this casethe roles of the various members of the partnership must be speltout.Successful project management and organization are becomingincreasingly important issues in the selection process. Indicate theintended management structure, the relevant experience of theresponsible managers, research quality assurance proceduresand other relevant budgetary and monitoring devices.TimetablingThis shows when the various operations will be carried out and inwhich sequence, and how long the project will take overall. A barchart showing the various tasks and their timing is a useful way tosummarize the programme, together with a written description ofwhat is involved in each task. A sequence of milestones with adescription of what will be produced at each will give a reliablemonitoring guide. Take into account any risk factors and build insome fallback solutions, particularly if the risks are overt and aninherent part of the project.OutputsThere is little point in carrying out funded (or, for that matter, anyother) research if the outcomes are not disseminated to those whomight benefit from them. The proposed outputs could be in theform of articles, reports, papers, books, pamphlets, guides,conferences, computer programs, etc. How the outputs will bemade available to potential readers/users will also be of interest.Your intellectual property rights should also be considered here.Your aim should be to retain as many rights as possible. However,in more commercial work this can become a complicated issueand you will probably need to gain expert advice and be involvedin discussions with the other participants and the promoter.Finalizing your proposal
By now you have certainly thought a lot about what your researchproblem is, and how you are going to do your research, and youhave probably written several drafts describing your intentions.Now you must write your research proposal in full, following therequired structure and including all the information required on theappropriate form.Checklist 8.6 summarizes what to do and how to do it, withreference to the parts of this book that deal with the differentaspects. You might not have to start at the beginning if you arealready clear about the area of your research, but you will have toclearly define the research problem, even if you are pretty sureabout the subject of your research. If you are doing a proposal fora PhD direct, it is assumed that you are already veryknowledgeable about your specific research topic and that you willbe able to go almost straight into the original research work. To beable to do this you will need to give a much more detailedexplanation of your research methods and a more comprehensivetimetable.Checklist 8.6 Finalizing your proposal: what to do, how to do it andwhere to find the guidance
The cogency of the argument is of paramount importance. A goodway to start is to make a series of short statements in which yousummarize the main aspects of your research project, and whichfollow each other to form an argument. Do not concentrate ongrammatical perfection or formal elegance at this stage. A seriesof concise notes will do. You have already done this as far as theexposition of the research problem in Exercise 8.3, so build onwhat you have achieved already. You will have to add whatapproach you will take to research the problem and discussalternative methodologies which might appropriately be used.Keep conferring with your tutor or supervisor as you refine thestructure of the argument and develop your successive drafts ofthe proposal. Make sure that other members of staff and/orcolleagues also see your work; different comments are essential.Remember that it is you who must decide what you want to do in theresearch, so be prepared to discuss it further with your tutor if you donot agree with some of the comments or recommendations made byothers.The completed proposal must be fitted onto the official registrationform for the university’s research degrees committee or fundingbody, or must conform to the requirements of the course or projectin which you are taking part. Normally, there is limited space onthe forms, and often you must ensure that your proposal is notlonger than can be comfortably fitted onto two sides of A4. This isnot easy, but is a very good discipline in writing a denseargument. Where possible, use the references as a shorthand todescribe theoretical approaches.Though there can be no standard or total guidance on howapplications should be completed, given the wide range of typesof research and students’ backgrounds that have to beencompassed, nevertheless you may find the list of typicalquestions in Checklist 8.7 of use in completing a form devised for
application to register for a research degree. Almost all of thepoints on the list will be relevant to more limited academicresearch projects as well as funded projects. Carefulconsideration of the questions should assist in both completingthe form and achieving the successful approvals.Whichever way the form is designed and in whichever sequencethe various sections follow each other (e.g. the aims of theinvestigation, the background, relationship to previous work,proposed plan of work, etc.), when preparing to complete the formyou must think of the overall structure in terms of:background (relation to previous work) > problem definition >problem solution (aim) > methodology (plan of work)After a logically structured research approach has been devised, itmust then be re-sequenced to fit into whichever format the formstipulates.Checklist 8.7 Questions by the research committee about yourresearch proposalThe following questions are those that the research committee mightask themselves when they are reviewing your proposal. Use them asa checklist when devising your proposal to make sure that you answerthem all as appropriate.1. Do the overall layout, structure and terms used demonstrate theintellect and training of a highly educated mature candidate whohas recently completed a lengthy course (with its extensivebackground of reading) in research methods, theories andpractice?2. Do the applicant’s qualifications, training, experience andprevious published research papers (if any) provide an adequateand sound basis for the research proposal?3. Does the title of the proposed research exactly identify anddelineate the area of the investigation/study proposed?4. Are the aim(s) of the investigation/study clearly related to theproblem(s) outlined in the background to the research?5. Does the research clearly identify the problem(s) it undertakes toinvestigate/study?
6. Does the application show:1. an adequate knowledge of the general background to theproblem – giving references to the mainliterature/publications on the subject2. an adequate knowledge of the specific problem area –giving references to the specific literature/publications onthe subject3. an adequate and up-to-date knowledge, by reference, ofother research in the same subject area that either hasbeen completed or is in progress4. an adequate basis for any tentativehypotheses/propositions that are used to structure themethodology (plan of work)?7. Does the application give adequate reasons/arguments to showthat the proposed specific area of research constitutes aworthwhile study as:1. a contribution to knowledge, or2. a contribution to development of theory in the subject, or3. a contribution to either theoretical or practicalmethodology?8. Does the proposed plan of work:1. show how the research will be structured2. show how the research will be phased3. clearly differentiate between deskwork and fieldwork4. show how the research will be carried out5. show the techniques and methods that will be used and thereasons why these techniques and methods were chosen6. show that the work proposed is practicable and realisticwithin the time available – including the period for writingup the thesis?9. If the research involves gaining access to the data and facilitiesof government departments or other organizations or institutions:1. What evidence is there that such access/cooperation willbe available?2. Are the result(s) of the research expected to have anygeneral or specific practical applications, e.g. designguidance or government policy?10. Does the research involve any problems of ethics/confidentialityand, if so, how will they be overcome?11. If the research involves investigation overseas, does theapplication:1. outline the knowledge/techniques/methods that arerequired to be learnt/understood in the home country priorto the investigation abroad2. illustrate an understanding of the part that different culturalvalues/factors/variables may play in the research and its
findings?12. If the application includes the possibility of transfer subsequentlyto a PhD, is the potential extension of the MPhil part of theresearch into the original research work required of a PhDlogically structured and connected?13. Are the facilities detailed in the investigation adequate and dothey include:1. computing2. libraries – both general and specialist?14. Collaborating institution:1. Does the application involve collaborating with anotherinstitution? If not, do the experience, expertise andbackground of the supervisor(s) as well as the type ofresearch involved render the use of a collaboratinginstitution unnecessary?2. Where a collaborating institution is proposed, is a letter ofacceptance from the institution concerned attached?15. Does the programme of related activities include reference to:1. completion of the research methods course2. courses/lectures that are being attended or will be attended3. computer course(s) completed4. any other relevant courses/studies?16. Are adequate details of the second supervisor(s) and adviser(s)listed in the application? Where the second supervisor(s) has notpreviously supervised a university student, is a CV (clearlyillustrating the expertise etc. necessary to the subject area)attached to the application?Successful proposalsThere is increasing competition in the research and consultancyenvironment. The process has become more subject to the time-limited, programmatic priorities of funding bodies, and toincreasing emphasis on the relevance to industry and theeconomy. The ways in which applicants convey their interest andexpertise, register and enter the competition are important criteriaused by promoters in the selection of participants and allocation offunds.It is essential that researchers are aware of how the researchagendas are set. To be best informed, get yourself on relevantcirculation lists that will supply you with information, calls for
expressions of interest, research conferences and otherdissemination events. Consult with the sponsoring orcommissioning bodies for details and opinions. For insideinformation, find out if any of the staff in your organization aremembers of an advisory or steering research group of thecommissioning body: they should know more details about theresearch-relevant agenda, although they might be bound byconfidentiality restrictions. Consult also with research directors inyour organization; they will have had plenty of experience inmaking applications. And do not hesitate to show the drafts ofyour proposal to colleagues in order to get opinions about theclarity and strength of your argument. Strict submission deadlinesare usually enforced, so do give yourself enough time to developthe proposal and to get support and approval from all partiesinvolved in order to avoid last minute panics: it inevitably takeslonger than you think at first!Submitted proposals are generally reviewed and evaluated blind(that is, without knowing who the submitter is) by a selected panelor committee, often with the help of specialist advisers drawn in toensure that this type of peer review is thorough. These specialistsmay be competitors or colleagues in your field of interest. Theywill be looking for the bids of the highest quality, based on solidand dependable research approaches that break new ground inareas of current concern, and that are feasible within the projectedframework, using the personnel proposed.Judith Margolin (1983, pp. 233–4) noted that a large proportion ofapplications for funded research are rejected. She devised a list often reasons why proposals are turned down, which provides auseful tool for a critical appraisal of your own proposal and isshown in Checklist 8.8. The list has been adjusted to show whichitems apply to academic as well as to funded research proposals,and which apply only to proposals for funded research. You willnotice how she stressed the need for clarity throughout as well asa demonstration of the practicability of the project. This is alsoreflected in efficient use of the money in funded research.Checklist 8.8 Reasons that proposals fail
Reasons relevant to all research proposals:1. There is an inadequately presented statement of need. It isperceived by the assessor either as not significant or as one ofsuch magnitude that it is clearly impossible for the loneresearcher to come to any useful conclusions in view of his/herlimited resources (or, in the case of funded research, a few grantdollars would barely make a dent in the problem).2. The objectives are ill-defined and are put forward as vague goalsor personal aims.3. The procedures are confused with the objectives.4. There is a lack of integration within the text among componentsof the proposal (e.g. aims and objectives do not match).5. The individual has adopted a poor approach and appealed on anemotional or a political rather than a factual or a theoreticalbasis.6. Not enough information is provided about the details of theproject.Reasons that apply only to applications for funded research:7. The funder does not accept proposals from unaffiliatedindividuals.8. The funder knows that the proposed idea has already been triedand failed.9. The funder approves of the concept but believes that theapplicant is not the proper individual to conduct the project orthat the institution with which the applicant is affiliated is notsuitable.10. The idea costs too much.Boxes 8.3 and 8.4 give two examples of successful proposals thatclearly explain the intended research, making good use ofsubheadings for each aspect of the project. The first is for a PhD,the second for an MPhil research degree.Box 8.3 Sample proposal 1Gender, work and poverty in IndiaJustification and thematic priorities
The ESRC thematic priority on ‘social inclusion and exclusion’ raisesquestions about the relations between economic growth andprocesses of social inclusion which are increasingly framed in a worldof global interdependence. This research will investigate a set ofresearch questions at the heart of the growth/social-inclusionrelationship in the context of India, since South Asia is where globalpoverty is most concentrated. The population of India will render theresearch results relevant to a large population, and there areimportant policy lessons to be learned from the experiences of theIndian state in poverty reduction. The Department for InternationalDevelopment White Paper on development (1997) put povertyreduction at the centre of all UK development policy, and DFID iscommitted to gender equity in development. The second thematicpriority which this research addresses is ‘economic performance anddevelopment’, in particular ‘the relationship between economic growthand development and patterns of deprivation and inequality’.The current thematic focus on social inclusion and exclusion aims todeepen our understanding of the nature of contemporary poverty andprocesses of marginalization. Work is pivotal to many understandingsof poverty, well-being and social exclusion, as well as to prescriptionsfor poverty reduction, as exemplified in both the New Poverty Agendaand many Women in Development policies, which emphasize labour-intensive growth and greater participation by women in employment(World Bank, 1990; UNDP, 1995). Social exclusion approaches alsofocus on employment-based inclusion for vulnerable or excludedgroups (Gore et al., 1995), and assume that well-being will beenhanced by employment, since work leads to income, socialapproval and self-esteem. Gender analysts have, however,questioned the pathways through which work leads to well-being, forexample, through ‘time-famine’ experienced by poor women due tolong working days, cultural disapproval of women in employment, andgender conflict within households. Furthermore, questions about theeffects on the poor of labour-intensive work that is also effort-intensivehave been raised by Breman (1996). It is important for developmentpolicy to unpack assumptions about work and social exclusion, andthe proposed research would provide an in-depth case studyinvestigation that would contribute to this.Research contextThe most appropriate case study within India, which would offer adepth of previous studies, is the Employment Guarantee Scheme(EGS) in the state of Maharashtra, which is the state programme thatbest exemplifies the labour-intensive approach of the proponents ofthe New Poverty Agenda. The EGS was adopted by the state
government of Maharashtra in 1974 and now comprises the singlelargest poverty reduction scheme of any state in India (Ravallion,1990). It offers a guarantee of employment to any rural adult at awage rate aimed to be low enough in order to self-target the poor. Thework typically consists of unskilled manual labour on small scale ruralpublic works, such as roads, irrigation facilities and reforestation, andthe explicit labour-intensive nature of the scheme is maintained via alabour–capital ratio in expenditures of at least 60:40. I am currentlyemployed on a research project that involves a field visit to the EGS,my supervisors have extensive contacts in India and I expect to beattached to a research institute in Mumbai during the fieldwork for thisresearch.Usefulness of the research and usergroupsThis research will be useful to two sets of users: first, to otheracademics and development analysts who are concerned withtheorizing well-being, work, gender and development; and second, todevelopment policy makers at a range of levels from project staff topolicy specialists in bilateral and multilateral aid agencies, and NGOs.I am visiting the US at Easter 1999 to collect materials and discussthis research with policy analysts in the UNDP, World Bank and IFPRI,and will be taking account of their comments in the research design.DFID has a core interest in poverty and gender research to informpolicy and I will consult widely within DFID.The research will have direct and indirect implications for policy, forexample, ascertaining which forms of employment-based interventionmight best enhance well-being, which interventions might threaten thewell-being of poor individuals, how the burdens and benefits ofinterventions are distributed, and how interventions need todisaggregate by gender in employment-based policies for socialinclusion and poverty reduction. The conceptual refinements incapability and embodiment theories will also aid academics workingon the conceptual underpinnings to contemporary labour-basedpoverty reduction policies.Research objectives and conceptualframeworkThe broad research objective is to explore the social processes bywhich labour-intensive work translates (or does not) into personalwell-being, and social inclusion, for poor women and men (more
detailed research questions are set out below). The research will begrounded in two conceptual approaches: the capabilities approach towell-being, and notions of embodied subjectivities that have beendeveloped in gender analysis. Combining these two approaches isnovel and will hopefully result in fresh insights.The capabilities approach (Sen, 1983) has increasingly come toreplace traditional poverty assessments that were in terms ofhousehold income, production or calorie levels. The capabilitiesapproach suggests the need to trace through the links from anindividual’s ‘endowments’ to their command over commodities,consumption and ‘decision-making’ which thereby transformsentitlements into personal well-being, in other words a focus on thesocial processes which generate well-being, rather than on poverty asa condition.The concept of embodiment, which grows out of the recent socialtheories of the body, is broadly taken to mean the display,maintenance and management of the body (Turner, 1996) and hasarisen due to an increasing recognition that our life experiences,subjectivities and agency emerge from our bodies’ habitual relationsto the world (Annandale, 1998). Research on embodiment in the westhas tended to focus on ‘abnormal’ bodies, such as the anorexic, andon sexualities. However, a concept of embodiment seems a veryuseful approach to poverty analysis as it attends to the endowmentwhich poor rural people rely on more than any other – their bodies;that is their physical endowments, their labour entitlements, theirextended entitlements to the labour of others (or their products), theircapabilities and functionings and their achievement of well-being(Jackson and Palmer-Jones, 1998).Research questionsThe proposed research will address the following linked researchquestions:1. How do the income/goods earned in the EGS connect toimproved economic well-being for poor women and men? Whatgender differences are there in these connections? Here I willlocate the EGS work as one element of a livelihood portfolio,which includes other elements, and look at how the food andincome gains from the EGS contribute directly and indirectlytowards poverty reduction, and how these effects are mediatedby gender.2. How does the effort intensity of EGS work affect the bodily well-being of gendered subjects? Does labour-intensive work, which
is also effort-intensive, such as in the EGS, have unrecognizednegative effects on the bodily condition of the poor? How arethese effects gender differentiated? Here I will study the socialbackgrounds (of caste, class, gender and age) of EGSparticipants, and the everyday realities of participation ofdifferently gendered bodies in the EGS will be explored tounderstand possible health effects and hazards for poor peoplewith stunted or wasted bodies. I will also consider questions ofagency: e.g. do poor people, whose bodies are frequentlystunted or wasted, nevertheless manage their bodily condition tolimit the negative effects of high energy-intensive work? Whatpractical and discursive strategies are used for this end, andhow do they vary by gender? I plan to investigate how farparticipants exercise choice in relation to participation, andagency in how they manage the effects of heavy labour on theirbodies, through working practices and social relations bothwithin the scheme and outside it.3. How does the experience of employment in the EGS producesocial inclusion of the poor, and how are these effects genderdifferentiated? Further research questions will focus on thesocio-cultural evaluations of work, the deployments ofdiscourses about work in intra-household bargaining andresource allocation, and the wider symbolic significance ofmoney/goods in changing gender relations, and move betweenlevels of analysis from household to community.I will also examine the ways the scheme is implemented tounderstand who obtains access and who is excluded, and what theterms and conditions of social inclusion are.The empirical study of this major poverty reduction policy will be usedto develop a broader gendered analysis of the policy implications ofemployment-based social inclusion initiatives.TimetableOctober 1999 to Summer 2000: UK, literature reviewing, researchtraining courses, language training. Summer 2000 to December2000: first fieldwork period in India.December 2000 to January 2001: UK, analysis of initial data andsupervision. February 2001 to Summer 2001: second fieldworkperiod in India to refine and complete data collection.Summer 2001 to Summer 2002: UK, writing-up period.Sources and methods
During the first year of research I will focus on analysis of secondarydata and materials. Relevant literature available in the UK will bereviewed to refine the research design and questions and to learnfrom existing studies. Visits to IFPRI and UNDP (April–May 1999)necessary for my current research job will also be a useful source ofinformation from international institutions. Other secondary data usedwill be EGS statistics, Maharashtra employment figures, the Indiancensus and Indian-based literature. The predominant source ofinformation for the research will come, however, from the primary datacollected during fieldwork.Attempting to explore people’s embodied experiences of work andwell-being is best approached through qualitative methodologieswhich will include life histories, participant observation, semi-structured interviews and key informant interviews. I plan to interviewEGS personnel, and observe the detail of its implementation, as wellas labour contractors and unions. A major part of the study will beinterviews with a number of male and female EGS participants andwork with them to understand how the EGS is situated within theirlivelihoods, social relations and life courses, and how they perceivewell-being. This work will be supplemented with studies of non-EGSworkers, local labour markets and livelihood systems. I will try toensure rigour in the qualitative work by triangulation, and a thoroughcontextualization of respondents’ voices. These methods will requireme to build trust between respondents and myself, which will involvelanguage study, a period for integration into the community uponwhich the research is focused, and collaboration with localresearchers.Ethical considerationsThis research will be based on studying a group of vulnerable poorpeople, and awareness of their rights to confidentiality and privacy willbe important, as will a sensitivity to making demands on their time. Iintend to approach these issues through informed consent, thoroughexplanations of the research, guarantees of confidentiality andanonymity in the thesis, acceptance of reluctance or refusals ininterviews, and a serious attention to building cooperation over time. Ialso think it important to recognize the assistance of local assistantsand scholars and be mindful of my own positionality. I will ensure thatcopies of research outputs are distributed within India and that Indianpolicy makers are consulted on the research.References
Annandale, E. (1998) The Sociology of Health and Medicine: a CriticalIntroduction. Cambridge: Polity.Breman, J. (1996) Footloose Labour. Cambridge: CambridgeUniversity Press.Gore, C., Rogers, G. and Figueiredo, J. (1995) Social Exclusion:Rhetoric, Reality and Responses. Geneva: International Institute forLabour Studies.Jackson, C. and Palmer-Jones, R. (1998) ‘Work intensity, gender andwell-being’. DP 96. Geneva: UNRISD.Ravallion, M. (1990) ‘Reaching the poor through rural publicemployment: a survey of theory and evidence’. Discussion Paper 94.Washington, DC: World Bank.Sen, A. (1983) ‘Economics and the family’. Asian DevelopmentReview, 1 (2): 14–26.Turner, B. (1996) The Body and Society, 2nd edn. London: Sage.UNDP (1995) Human Development Report, 1995. Oxford: OxfordUniversity Press.World Bank (1990) World Development Report, 1990. Oxford: OxfordUniversity Press.[This was a successful PhD proposal by Louise Waite, NorwichUniversity.]Box 8.4 Sample proposal 2The impact of introducing cognitivebehavioural therapy on nursing practicein acute areas of a mental healthcare trustAims of the investigation1. To introduce a psychological model to nurses working on acutein-patient units of a Mental Healthcare Trust.2. To ascertain and explore nurses’ perceptions of their role andclinical competencies in an in-patient area using a CBTframework.
3. To evaluate the impact of CBT techniques on nurses’ perceivedand actual clinical effectiveness in their work areas.Proposed plan of work, including itsrelationship to previous work, with keyreferencesIntroductionSeveral nurses with recognized specialist training and skills wereemployed to help develop clinical nursing practice within the Trust(OMHT). The main aims of the role as a Clinical Nurse Specialist(CNS) is to provide specialist post-basic education and clinicalsupervision to other nurses. Two of the outcomes sought are: (1)continual improvement of nursing practice and clinical effectiveness,and (2) personal development of the nurse practitioner. In the processof developing the role of CNS in Cognitive Behavioural Therapy (CBT)for the OMHT, one of the challenges identified was establishing aneffective way to introduce speciality skills into nursing practice withdifferent cultural characteristics to each of the five units offering acutein-patient care. While each unit has different ideologies and values intheir approach to nursing care, they share the same function in thepatients they treat. The first action taken by the author from 1990 to1993 was to have a nurse seconded from an acute unit of OMHTworking with the Clinical Nurse Specialist group for a period of oneyear. The nurse was introduced to all specialist theories and hadexperience working with patients using the differing models oftreatment. The second clinical practice development was a CBTtraining group. The membership comprised a staff nurse from each ofthe five acute units. During the CBT training the members were alsoencouraged to identify research questions within their practice withrelevance to all five units and suited to the CBT process and elementsof an action research paradigm.From reflective journals based on open-ended questions on topicssuch as attitude to level of functioning in their role and with patientsetc., there were a number of outcomes (Dodd, 1995). Theseoutcomes were not formally investigated. Both of the clinicaldevelopments produced similarities. The main areas this researchseeks to investigate based on the results are: (a) present nursingpractice does not use a standard nursing model, (b) nurses initiallybelieved themselves to be less effective with patients, (c) thebehaviour in some of the patients was experienced as threatening andinsoluble, (d) nurses believed they were not always skilled to deal with
certain groups of patients even though their skills were evident, (e)negative assumptions (irrational beliefs) and thoughts were triggeredwhich made them question their role, believe they were deskilled andinadequate to be clinically effective, (f) nurses were not identifyingtheir own distress, comparing themselves negatively to theircolleagues, (g) their own self-evaluation was negative, affecting theirability to procure job satisfaction, and they felt on the fringes of theteam, fearing the team would identify them as inadequate in their roleand be critical of them as nurses, (h) their behavioural response wasto contain patient behaviour functioning in a custodial manner ratherthan work in a psychological method. A preliminary studydemonstrated and confirmed that nurses believed the existing nursingmodels were not always relevant in psychiatry.In relation to research, none of the interviewed nurses read or usedresearch material and they were unaware if any colleaguesparticipated in this exercise. No nurse interviewed had or was likely toinitiate research within their practice. The main reasons were lack oftime and insufficient knowledge of research. These findings are similarto those in studies elsewhere (Bostrom et al., 1989; McKenna, 1990;Marsh and Brown, 1992; McKenna, 1994).There is a lack of theory underpinning nursing practice in the acuteunits of the Trust. The outcomes of the clinical developmentsdemonstrate a lack of confidence and a negative self-evaluation ofnursing staff who were participating in the projects. The secondclinical development integrated a theoretical understanding of apsychological treatment into an established nursing practice.Recognizing their own perceptions of themselves whilst in their role asa nurse was an integral component of the CBT training. The nurseslearnt how to challenge and correct their perceptions if they werenegative and interfered with their confidence and self-esteem. Theimpact improved understanding of the patients, the nurses becamemore efficient in their role, they supported their colleagues, and theyfelt more potent and skilled to deal with most difficulties that acuteareas offer.Cognitive behavioural therapyOver the past three decades, cognitive behavioural therapy (CBT) hasbecome one of the predominant forces in psychotherapeutic practice(e.g. Mahoney, 1988; Arkowitz and Hannah, 1989). The publication ofa book specifically devoted to in-patient applications of Beck’s modelof cognitive therapy is further evidence of the growth of this approach(Thase and Wright, 1991; Beck, 1993; Wright et al., 1993). Theconcerns world-wide appear to seek cost containment and more
efficient treatments for in-patients. Indeed, in the UK the governmenthas moved toward community care and Trust status confirming thesechanges in attitudes and treatment (HMSO, 1989). Some of theadvantages of cognitive therapy that have fuelled its development isshort-term format, established efficacy, compatibility with biologicalpsychiatry, and evidence of relapse prevention effects (Wright et al.,1993). Nurses may be introduced to basic knowledge of CBT in theirtraining or through a locally based postgraduate course. Unfortunatelythe practice of CBT by the majority of CBT trained nurses is inspecialist areas usually attached to Departments of Psychology orPsychotherapy or working in Primary Care. There is little opportunityor facilities for in-patient staff to develop such skills (Friedberg andFidaleo, 1992). Cognitive behavioural therapy is widely researched asan effective psychological treatment for patients with mental healthproblems (Hackman, 1993). Some of the principles governing thistherapy are: collaborative alliance with the patient; the patient, whereappropriate, setting their own targets or goals in treatment; treatmentsare specified in operational terms and evaluated with a variety ofobjective, valid and reliable measures, providing the practitioner andthe patients with empirical data on the effectiveness of theinterventions utilized. This results in a continuous dialogue betweenclinical practice and research, ensuring that theoretical research andclinical issues interact in a unique way (Hackman, 1993). Meyer(1993) and Titchen and Binnie (1993) describe how action researchsupports the systematic development of knowledge born of practicewhich complements the CBT model and approach.Theoretical frameworkBuilding on the informal information derived in clinical developments,this study will further investigate the impact of CBT on nursing practicein mental healthcare. The study will be developed through the CBTdevelopmental model (Fennell, 1989). This model is similarlydescribed elsewhere for research as a ‘multi component view ofattitude’ in Fishbein and Ajzen (1975) based on a conception byRosenberg and Hovland (1960). The developmental model suggeststhat attitudes are predispositions to early life events that are operatedfrom the unconscious by stimuli pertinent to those attitudes. Theattitude will trigger cognitions which are the perceptual evaluation ofthe stimulus (Fennell, 1989). Beck et al. (1979) suggest thesecognitions influence how we feel emotionally and physiologically. Ourbehaviour is then based upon the cognitive evaluation and subjectiveattribution of belief and the emotional impact on the person. Thisprocess is well described in the outcomes of the pilot project. It wasapparent that the influences of CBT training had an impact on some of
the group which influenced their attitudes and beliefs and thereforetheir clinical behaviour. The therapeutic model is deemed appropriatefor this study, as it is introducing the construct for self-understandingand the therapeutic process and it links together theory and practiceas well as the areas this study seeks to examine. The study willexplore the impact of the CBT tools and techniques on the nurses’perceptions within their role and clinical behaviour, i.e. what they dowith patients while working in the acute areas, and ascertain if thenurse is more clinically effective.Operational definitionsClinical effectiveness. This is defined as patients’ perceptions ofthe impact of nurse interventions on their psychological, emotionaland physical health. These will be rated by a patient/nurse ratingscale.Competencies. This is defined as clinical decision making asportrayed in the case notes, care plans and clinical supervision.Irrational beliefs (IB). This is the ‘cognitive’ term that describesattitudes, assumptions, basic beliefs that are not in the consciousawareness, yet govern our responses to outside stimuli. Ifirrational, then one’s response may be dysfunctional.DesignThis study will use a combination of qualitative and quantitativemethods to identify nurses’ attitudes and perceived role. Twoquantitative instruments will be modified for this study. There will befour components involved in the study design: (1) exploratory, in whichreflective journals will record uncensored thoughts, emotions andbehaviour of subjects and provide themes to formulate semi-structured interviews; (2) modification of the IB instrument based onthe literature and the exploratory phase; (3) test IB instrument forreliability and validity; (4) a pilot study on five subjects and fivecomparison subjects, using the IB and second quantitative instrument.Phase 4 includes the CBT training intervention. The didactic periodwill be six weekly workshops followed by ongoing weekly supervisionuntil the group terminates. Key areas that will be taught are:assessment, conceptualization (Persons, 1990), Socratic questioning,and the application of CBT tools and measures to practice.Sample
The purposive sample for this study will be nurses working on in-patient acute mental health areas of a Mental Healthcare NHS Trust.The five participants will be chosen for their interest in using CBT andrepresent each of the five acute in-patient areas. A comparison groupof five will be randomly selected from nurses of similar grade/positionworking on the in-patient areas and not undertaking the CBT trainingintervention. The sample for this study is small owing to the nature ofclinical practice development and introducing theory into practice. Theprocess is prolonged for the subjects because the nature of trainingand supervision required is to ensure safe and effective practice.Finally, the research is undertaken within the context of the CNS roleand not as an independent researcher.InstrumentsThe study will use a range of instruments:1. Reflective journals will allow subjects to free associate andrecord their thoughts, feelings and behaviour within their workarea pertaining to their role and clinical practice.2. Semi-structured interviews using Socratic (open-ended)questioning based on the themes contained in the reflectivejournals, e.g. thoughts of being deskilled and inadequate, beingnegatively evaluated by colleagues etc.3. An IB of the nurses’ questionnaire modified from the IrrationalBelief of Therapists (Wright, 1985). This instrument consists of anumber of items that subjects will score on a 10-point scale of 1–10. The questionnaire will be tested for reliability and validity aspart of this research. The IB scale is used in therapist training toaid in recognition of assumptions that may interfere with thetherapeutic alliance and clinical behaviour.4. A patient/nurse scale which will be developed from thepatient/therapist scale, also a tool used in therapist trainingwhich supplies feedback from patients, giving the opportunity ofcollaboration to patients not assertive in session and aiding thetherapist in refining their clinical skills (Beck et al., 1979). Thisscale will be completed by the patients for each formal one-to-one session. The subjects will give it to the patient(s) who willrate their perceptions on a scale of 1–10 based on a number ofitems relating to the clinical behaviour of the nurse. The scale isorganized into key domains: appropriate use of CBT tools; thenurse being clear and rational about interventions; and thecharacteristics and behaviour of the nurses in session. It isanticipated that the scale will be introduced periodicallythroughout the training and clinical intervention. This instrument
will be used to collect data in relation to clinical behaviour andefficacy.Data collection1. Reflective journals will be completed at the subjects’convenience. This may take up to 30 minutes per entry at most,depending on the events of the shift.2. Following completion of the journals over a one-month period,they will be examined and help to form the semi-structuredinterviews. The subjects will be seen in their area of practice forthe interviews.3. The IB scale will be administered before and after the CBTtraining intervention.4. The patient/nurse scale will be administered for a one-weekperiod at differing stages throughout the training and supervisedclinical practice. It is anticipated that each subject will havetwelve to fifteen ratings per week. Collection of the scales will beduring the group meeting where the participants discuss,compare and share their experiences and values with theresearcher.5. The IB and patient/nurse scales will be given to the fiverandomly selected groups of nurses of similar grade/positionworking on the in-patient areas at the same time as the subjects.AnalysisThe reflective journals will be examined for themes that reflectattitudes/beliefs, e.g. perfectionism, expectations, responsibility andfears of failure. The journals will be compared to identify similaritiesamong the subjects. Semi-structured interviews will be conducted toclarify the themes from the reflective journals into statements (beliefs)and consequences, e.g. ‘If I make a mistake then I will damage thepatient’. These will be used to modify the IB scale. The IB scale is notrecognized as a formal research instrument; therefore as part of thisstudy the instrument will be tested for reliability by test–retest, internalconsistency and interrater reliability, and for validity through criterion-related validity and concurrent validity. The subject scores from the IBwill be compared pre and post CBT training intervention. The IBscores will be compared to the IB scores of the comparison group preand post CBT intervention. The patient/nurse scale scores will beexamined to ascertain any improvement as the subjects progressthrough the training and clinical practice and will be correlated withany significant changes in the subject IB scores. These scores willalso be compared to those of the comparison group.
EthicsThe study will be forwarded to Nursing Research Committee by March1996. Subjects will be assured of confidentiality. Recorded datanecessary for reports will be given anonymity. Subjects will beassured that the purpose of this study is to enhance their clinicalpractice and development and is not an assessment for anymanagerial appraisal or to their detriment. A full information sheet willbe given to outline these assurances.Plan1. Registration, form group and consent of participation, March1996.2. Reflective journals and semi-structured interviews, April/May1996.3. Modify and test IB, June 1996.4. Administer IB, initiate CBT training intervention and patient/nursescale, weekly supervision, assess adequate CBT workingknowledge of project group, September 1996.5. Clinical behaviour for appropriate period, apply patient/nursescale periodically, begin September 1996.6. Data analysis, case report, publish findings, November 1997.BudgetThis project has the financial support and clinical collaboration of theOxfordshire Mental Health NHS Trust. This is demonstrated by thecovering letter attached.ReferencesArkowitz, H. and Hannah, M. T. (1989) ‘Cognitive, behavioural, andpsychodynamic therapies’, in A. Freeman, K. M. Simon, L. E. Beutlerand H. Arkowitz (eds), Comprehensive Handbook of CognitiveTherapy. New York: Plenum.Beck, A. T. (1993) Cognitive Therapy: Past, Present, Future. NewYork: Guilford Press.Beck, A. T., Rush, J. and Emery, G. (1979) Cognitive Therapy forDepression. New York: Guilford Press.
Bostrom, A. C., Malnight, M., MacDougal, J. and Hargis, D. (1989)‘Staff nurses’ attitudes toward nursing research: a descriptive survey’,Journal of Advanced Nursing, 14, pp. 915–22.Dodd, H. (1995) Staff Nurse Development Project Annual Report.Published for OMHT internally.Fennell, M. J. V. (1989) ‘Depression’, in K. Hawton, P. M. Salkovskis,J. Kirk and D. M. Clarke (eds), Cognitive Therapy for PsychiatricProblems: a Practical Guide. Oxford: Oxford University Press.Fishbein, M. and Ajzen, I. (1975) Belief, Attitude, Intention andBehaviour. Philippines: Addison–Wesley.Friedberg, R. D. and Fidaleo, R. A. (1992) ‘Training inpatient staff incognitive therapy’, Journal of Psychotherapy: an InternationalQuarterly, 6 (2), pp. 105–12.Hackman, A. (1993) ‘Behavioural and cognitive psychotherapies: pasthistory, current applications and future registration issues’, Journal ofthe British Association for Behavioural and CognitivePsychotherapies, Supplement 1, pp. 2–72.HMSO (1989) Working for Patients. London: HMSO.Mahoney, M. J. (1988) ‘The cognitive sciences and psychotherapy:patterns in a developing relationship’, in I. Dobson (ed.), Handbook ofCognitive Behavioural Therapies. New York: Guilford Press.Marsh, J. M. and Brown, T. L. (1992) ‘The measurement of nurses’attitudes towards nursing research and the research environment inclinical settings’, Journal of Clinical Nursing, 1, pp. 315–22.McKenna, H. P. (1990) ‘The perception of psychiatric-hospital wardsisters/charge nurses towards nursing models’, Journal of AdvancedNursing, 15, pp. 1319–25.McKenna, H. P. (1994) ‘The attitudes of traditional and undergraduatenursing students toward nursing models: a comparative survey’,Journal of Advanced Nursing, 19, pp. 527–36.Meyer, Julienne E. (1993) ‘New paradigm research in practice: thetrials and tribulations of action research’, Journal of AdvancedNursing, 18, pp. 1066–72.Persons, J. B. (1990) Cognitive Therapy: a Case Conceptualization.London: Norton.
Rosenberg, M. J. and Hovland, C. I. (1960) ‘Cognitive, affective, andbehavioural components of attitudes’, in C. I. Hovland and M. J.Rosenberg (eds), Attitude, Organization and Change. New Haven, CT:Yale University Press.Thase, M. E. and Wright, J. H. (1991) ‘Cognitive behaviour therapymanual for depressed in-patients: a treatment protocol outline’,Behaviour Therapy, 22, pp. 579–95.Titchen, A. and Binnie, A. (1993) ‘What am I meant to be doing?Putting practice into theory and back again in new nursing roles’,Journal of Advanced Nursing, 18, pp. 1054–65.Wright, J. H. (1985) ‘Irrational belief of therapists’. Unpublisheddocument.Wright, J. H., Thase, M. E., Beck, A. T. and Ludgate, J. W. (1993)Cognitive Therapy with In-patients: Developing a Cognitive Milieu.New York: Guilford Press.[This was a successful MPhil proposal by Harry Dodd at OxfordBrookes University]Exercise 8.5Read Boxes 8.3 and 8.4 analytically, paying particular attention to thestructure of the proposal, and locating the following features of each:Title of the researchAimsLiterature review and explanation of the theory and history to supportthe researchMethod of citationExposition of the research questions or problemsResearch methods used to investigate the questions or problemsEthical issuesProgramme of workOrganization of list of referencesCompare these with what you have written by making the same analysisof your proposal. How does it match up?
So now you should be at the point where you can submit yourproposal and are really clear about what you will be doing over thenext few months. In response to the common question asked whenpeople find out that you are a researcher – ‘Oh yes, so what is yourresearch about?’ – you will be able to respond with the question, ‘Doyou want the five second answer, the five minute answer or the fivehour answer?’ConclusionsThe research proposal is the most important document you haveto prepare before writing the thesis or report itself. Although itdoes not irrevocably commit you to a particular detailed course ofaction, the effort and time which you expend in its preparation,and the vigorous thought which is required to formulate yourintended project, make it unlikely that you will want to drasticallychange the direction of your research after your proposal hasbeen approved. In the case of funded research, it will probably bedifficult to make any significant changes anyway. This is not tosuggest that no alteration in your course will take place.Particularly with very long courses of study, such as theMPhil/PhD course, it is impossible at the proposal stage to predictin detail how the outcome of the MPhil work will affect theresearch to be carried out at PhD level. You have the opportunitywhen submitting the transfer document to provide a review of thefindings of the MPhil stage of the research and to outline theproposed PhD research on the basis of your conclusions.Apart from the proposal serving as an essential guide to setting upyour plan of work and starting your writing, you will find that it isuseful to refer back to it at regular intervals during the researchwork. When you are deeply involved in the minutiae of detailedpracticalities, it will help to refresh your memory about the overallobjectives of the research, and to see how your present task is astage in the context of the whole project.The next steps: your researchproposal
The culmination of this book is the production of your researchproposal. If you have understood the issues raised in thepreceding chapters, and have related them to your research intothe background and important current issues of your topic, youshould be quite clear about the nature of your research problemand the methods you will use to undertake the research. The realchallenge facing you now is to form a tight and compact argumentthat justifies the importance of your project and demonstrates itspracticability within your resources, while at the same timeconforming with the regulations relating to the research degrees,course objectives or project description.The aim of this section is the completion of a sound researchproposal.Checklist of activities that will progress yourresearchStep 1: Specific course-related guidanceBefore you finalize your proposal, make sure that you are aware of anyspecific course-related guidance or instructions. This will save you makingchanges later. For funded research, carefully read the conditions thatcome with the application forms and guidance given by the fundinginstitution.Step 2: Type of research degreeYou should be aware by now about the requirements of the type of degreethat you are undertaking, e.g. master’s, MPhil, MPhil with transfer to PhD,or PhD direct. Remember that your proposal should respond to theserequirements. If you are in any doubt, consult with your tutor or supervisor.Step 3: Get the right forms‘If you are alive, you have to fill in forms’ noted my son when he went touniversity. There is a form for everything, including for research proposals.There are likely to be standard forms on which to submit your proposal.You will be required to include not only your proposal text, but alsopersonal details and administrative information (such as your sponsors,
supervisors, advisers, etc.). Find out who must sign the forms, thesubmission procedure and timing.Step 4: Finalize the title of your studyRemember that the title is the shortest summary of your work, and themost used for searching in databases. Ensure that it contains the mainconcepts of your study and any limitations to its scope, but remainssufficiently precise to distinguish it from other works.Step 5: Your list of aimsThis should not be more than four or five long, so the aims need to befocused and sufficiently varied to cover the range of objectives you areaiming at. Check that the research problem and expected outcomescorrespond to the aims: it is best to use the same terminology to ensurethis.Step 6: Your background and literaturereviewIt is easy to write too much here, as a result of your extensive backgroundresearch! Concentrate on forming a succinct argument that persuadesabout the need for the investigation based on the context of existingknowledge (or lack of it) and on previous research. Provide enoughinformation for any intelligent person to understand the main factorsdiscussed and the state of the debate: the research committee are notnecessarily academics in your field. For PhD direct, make sure that youinclude your own published and unpublished work on the subject and anyrelevant experience you may have: you are expected to have completedthe MPhil-type research as preparation for this project.Step 7: Your research problemThe research problem is the linchpin of the proposal, which is generatedby your discussion of the background and in turn generates the outline ofresearch methods. It is essential that its formulation is tightly constructedso as both to focus the work and to limit it in scope. There is no room forvagaries.Step 8: Outline of methods
You can do this in list form if you like. If you are doing an MPhil withtransfer to PhD, divide the sequence into two parts, MPhil stage and PhDstage. You will use the MPhil stage to clarify and further define theresearch problem and select the specific research methods for the PhDstage. For a PhD direct, you will be expected to be absolutely clear aboutthe problem and which research methods you will employ to investigate it.Particularly important are the details about data analysis and how youintend to achieve the aims of the research.Step 9: TimetableA timetable showing the succession of stages and the time they will take isan essential tool for any research project, however small. Use one ofthese to demonstrate the practicality of what you are proposing. If you aredoing a funded research programme, you will also have to include‘milestones’. These are intermediate points in the project where progressis reviewed and staged payments are made.Step 10: The rest of the proposal applicationformRead carefully all the sections that you need to fill in and make sure theyare complete. You may need to cooperate with the course administrator tocomplete details about supervisors and to get the appropriate signatures.Make sure you leave enough time for this; it might take several days toliaise with the people concerned. For funded research, quite detailedparticulars of collaborators will be required and it can be a lengthy processto compile all the necessary information.The research committees will have regular dates for their meetings,research funders will have deadlines for submissions, and master’scourses will also stipulate when the proposals have to be submitted. Makeadvance preparations so as to make sure that you do not have a panic atthe last moment.Step 11: Ethics approvalIf there are ethical issues concerning research involving humanparticipants, make sure that you comply with the procedures of yourinstitution and any other that might be involved. More form filling!Consolidation and assessment
You will need to arrange a whole series of tutorials with yourallotted tutor or supervisor and also other members of staff and/orcolleagues as you work from one draft of the proposal to the next.Ten to twenty drafts are not unusual for an extended project, somake sure that the correction and development of each draft arequick (if possible on a daily basis) so that the issues remain freshin your mind, and you do not get bogged down in minutiae.When you have completed the proposal to an acceptablestandard, write and paste in all the information required into theelectronic copy of the proposal application form provided by yourcollege, university or research funder. Be careful to follow all theinstructions which come with the form to maximize your chancesof success.After having submitted your proposal, and while waiting for theresult, you can take comfort in the fact that the skills you havelearned whilst reading this book will be a good preparation foryour research project which, because of your lucid and thoroughproposal, is about to be approved!Further readingThere are books that are solely dedicated to writing academicproposals of all kinds. The principles are the same for all of them;it is the extent and detail that vary. All are reasoned arguments tosupport a plan of action. If you want to read more, or find differentapproaches to proposal writing, you can explore some of thesebooks; you will undoubtedly find something useful. I have putthem in order of complexity, simplest first, though you may want tolook at a more subject-specific book first. Every book on how to dodissertations and theses will also have a section on writing aproposal; you will undoubtedly have seen some of these by now.Andrews, G. (2017) Research Proposal: Academic WritingGuide for Graduate Students (Essay and Thesis WritingBook 3) Kindle Edition.
http://www.writeyourthesis.comviewed 20 March2019.This short guide will give you the tools you need write a researchproposal, no matter what your field of study is.Terrell, S. (2015) Writing a Proposal for Your Dissertation:Guidelines and Examples. New York: Guilford Press.This user-friendly guide helps students get started on andcomplete a successful doctoral dissertation.Cape Town: Juta.A practical, application-focused guide to writing a proposal forbasic and advanced research projects.Trowler, P. (2015) Writing Doctoral Project Proposals, 2ndedn. CreateSpace Independent Publishing Platform.Designed to be as helpful as possible for anyone who needs towrite a doctoral research proposal in the area of higher educationresearch.Locke, L.F., Spirduso, W.W. and Silverman, S. J.(2007) Proposals That Work: A Guide for PlanningDissertations and Grant Proposals, 5th edn. London:Sage.Covers all aspects of the proposal process, from the most basicquestions about form and style to the task of seeking funding.Silyn-Roberts, H. (2016) Writing for Science andEngineering. Papers, Projects and Proposals: A PracticalGuide, 2nd edn. London: Elsevier.Explains in plain English the steps to writing abstracts, theses,journal papers, funding bids, literature reviews and more. Thebook also examines preparing seminar and conference
presentations. Written in a practical and easy to follow stylespecifically for postgraduate students in Engineering andSciences.Coley, S.M. and Scheinberg, C.A. (2013) ProposalWriting: Effective Grandmanship, 4th edn. ThousandOaks, CA: Sage.Offers a step-by-step guide to writing a successful grant proposalto meet community needs. Throughout the book, the authorsprovide a guided process to assist the new grantwriter inunderstanding how to find grant opportunities, how to develop aviable project and evaluate outcomes, and how to prepare anapplication for funding.Meador, R. (1984) Guidelines for Preparing Proposals.Chelsea, MI: Lewis.A bit old, but this one is good on motivation. Read this to get thefull story.Jay, R. (2003) How to Write Proposals and Reports That GetResults. Harlow: Pearson.Really aimed at managers, but we all have to manage! Goodpractical advice on all aspects of writing.
9 Writing Strategies: Getting Startedand Maintaining Momentumthroughout Your ProjectAimsTo assist you to start planning and writing your thesis ordissertation or research reportTo remind you of the need for correct citation and referencingTo provide guidelines for preparing other sorts of other writing,such as presentations, progress reports posters, articles andconference papers, and possibly even research papers foracademic journalsIntroductionAfter your proposal has been accepted by your course leader,your university’s research degree committee, or the funding bodyto which you applied – what comes next? It is very easy to relax atthis point, having exerted great efforts to sort out and describeyour intended research. After a short pause for breath, you shouldnow regard your proposal as your springboard to doing theresearch proper, and take the first steps in carrying it out.It is worth at this stage considering your ideal working style. It iswell known that some people work best early in the morning whileothers late in the evening, and others in between. You will knowby now to which group you belong, and can work accordingly. Aword of warning, though, if you are an evening person: Roeser etal. (2012) found that late workers tended to be more stressed thanthose working during the day and suffered more psychologicalproblems.As you make headway with your research and get on with yourwriting, you may be asked to present your work in progress in
seminars or student research conferences. This entailsencapsulating your work in a short talk supported by apresentation, quite a challenge when you are full of all the detailsand issues that you are grappling with! If you think it might be ofpublic interest, you may also want to describe what you are doingin a local paper or magazine.Attending conferences is a brilliant way to find out what is thelatest and ongoing research in your subject, and meeting theexperts in your field. Better still is to be able to contribute a paper,presentation or poster to the conference in order to flag up whatyou are doing to the research community. When you are moreadvanced in your research, getting a paper published in a peerrefereed research journal is a pinnacle of achievement. In somePhD courses this is mandatory before you are allowed to submityour thesis.This chapter will provide you with some simple guidelines andexercises on how to achieve interesting and successful resultsfrom the different types of writing you will have to do.Where to begin on your dissertationor thesisThe description of methods provides a succinct action list whichyou can now start to implement. What you should do first reallydepends on what is on the list. However, there are two preliminarytasks that will help you to organize your work:1. Devise a structure plan of your thesis, dissertation or report(or at least of the introductory chapters).2. Devise a detailed timetable of your actions for the next yearor so if you are doing a research thesis, or for the duration ofa shorter research project.The most usual item at the head of the plan of work is to continuebackground reading and consultations and critical analysis of theliterature. When you submitted your proposal, you should havecompleted much of the necessary reading and preliminary
information collection, so you should be able to start writing a draftof your introductory chapters. It is an acknowledged fact thatcommitting yourself to beginning your thesis or report is not easyor comfortable. Preece has described the psychological problemsof starting the task of writing:One common reaction is that the student becomesoverwhelmed, not to say mentally paralysed, by the seeminglydaunting prospect of writing so many words. (1994, p. 210)However, there are two good reasons why you should begin yourresearch project in this way. First, the process of writing forcesyou to review your notes and to assess their suitability for use inthe argument you will be developing in your chapters, and mayalso reveal gaps in the information you have collected. Secondly,it tests the adequacy of the argument that must form thefoundation of your research.It is difficult for your supervisor to comment in any detail on yourwork if it is not written down. However, the writing process is noteasy and requires considerable discipline and organization. It hasoften been noted that professional authors invariably work regularoffice-type hours and do not sit around waiting for inspiration!Inspiration comes through active engagement in the writingprocess, and feeds on the efforts required to organize and createwritten work. However, do remember to take appropriate breaks toavoid the stresses and strains caused by using the computer forlong periods. You can use these breaks to give yourself time tothink quietly about your work. Once you have successfully writtena few sections of text, the task will no longer seem to be sodaunting.Figure 9.1 Professional authors invariably work regular office-typehours
The writing processAccording to Freeman and Meed (1993, p. 53), all writing involvesa number of stages, as shown in Checklist 9.1. Note how onestage is shown to follow the other, as is inevitable when presentedas a list, but in reality, the first stages tend to be iterative, in thatyou need to do reading throughout in order to gain knowledge andrefine your topic. No writing is done in a vacuum – it is alwaysdone within a context of other texts and situations.Checklist 9.1 Stages of writing1. Identify what kind of writing you must do (the task), and what it isfor (the reader).2. Decide how to tackle it: this can involve choosing a topic oranalysing a question.
3. Collect information you will need: this may involve doingresearch or reading.4. Prepare an outline for what you will write.5. Write a first draft and submit it to others for comment.6. Edit and redraft to refine it.7. Give a final check before handing it or sending it to the reader.Whenever you want to produce a piece of writing, you have to askyourself whether to write something that merely describes asituation or an event, or to argue about a particular point of viewor a research problem, saying, for example, that there is arelationship or not between X and Y. So you will be faced witheither a descriptive or an argumentative approach. Smith andSmith (1990, p. 116) define a descriptive approach as one inwhich you have to list important issues. Your own point of view isoften not required at all, although you may wish to comment onand compare the issues. On the other hand, an argumentativeapproach is one in which you have to state your own point of view,and to defend it by giving supporting arguments.Forming the structure and preparing an outlineThere is no single right way of setting about your writing task.Before you start writing, however, it is essential to devise astructure for the work, and then to prepare an outline of what youwant to write. The best approach depends on the nature of thepiece (a short article, a chapter of an extended thesis, a denselyargued conference paper, etc.). Consider the following adviceoffered by various authors, and select and adapt it to help you toorganize the ‘skeleton’ of the particular writing task you are facing.Many reports are written with the intention of answering aquestion or a series of questions. When formulating the structureof your report, you can start by setting out the question orquestions you are trying to answer. You may start off byformulating a large number of questions and in this way breakdown the major problem posed by the subject into a series ofquestions or sub-problems. You have to select the significantissues and reject those of only peripheral importance. The muchreduced list of questions can then be arranged to form a
preliminary plan of your report, and consideration of how you willanswer these questions will prompt you to devise a logicalarrangement for the components of your report. According toNewman:The final questions must clearly be of such a nature that theywill fit into a pattern or structure which can form the outline ofthe essay. From the answers to these questions your aimmust be to construct a coherent, logical pattern of ideas andarguments that will form the structure or backbone of youressay or paper. Into this structure the data which you havediscovered can then be fitted to ‘flesh out’ and support yourtheme, argument or thesis. (1989, pp. 20–1)There are other ways of planning a piece of written work. A planwas defined by Freeman and Meed (1993, p. 59) as a series ofheadings with an idea or two under each heading. The headingsshould do no more than cover the points you intend to write about.The priority in the plan is logical order. This does not mean thatthere is one and only one order which you can consider but itdoes mean that your final order must justify itself and be clear tothe reader. One way of doing this is to look at the material youhave collected and decide which are the main points you wish tocover. Your plan would then look like Checklist 9.2.Checklist 9.2 Plan of written workAn introduction: to define your terms and indicate how youintend to tackle the topicA list of main sections, summarized in the form of topic headingsA summary or conclusion: to recall the issues raised in theintroduction, draw together the points you made in the mainsections and explain the overall significance of your conclusionsFairbairn and Winch (2011) elaborated on the above points. Theyadvised that in preparing an outline you should consider carefullywhat you want to say about the topic. How should you bestintroduce what you want to say? How will you conclude? What are
the central points of your argument? Each word should addsomething to the sense of the sentence and each sentenceshould add something to the thrust of the paragraph of which it isa part. Each paragraph should make a coherent contribution towhat you write about your topic. They also advise you to pursueyour argument as in Checklist 9.3 (2011, pp. 220–33). It isinteresting to note that they state you should be aware of yourconclusions even before you start. This means that you will havedone enough preparatory work to have reached your conclusions,and that you can describe the stages that lead up to them. Thistype of writing is not like the working process of some novelists,which involves being led by the development of characters andletting them take you on a journey of discovery.Checklist 9.3 Setting up and pursuing your argument1. Begin by thinking about what it is you want to say, including whatconclusions you wish to reach. Write down a brief statement ofthe gist of what you intend to communicate, perhaps listing themain points to be made.2. Work out the best order for presenting your main points. Make askeleton of your essay.3. Check that the ideas you want to present follow on from oneanother. If they don’t, add linking ideas.4. Check whether any examples you have listed are adequate toillustrate the main points you wish to make.5. Decide on how you will introduce your piece. Write the draftintroduction. Avoid long-windedness: introductions should do nomore than engage the reader by giving some kind of ‘trailer’ forwhat is to come.6. Write a draft of the remainder of the essay.7. Write the conclusion or concluding paragraphs. At this point youmay wish to emphasize some points or recapitulate the mainthreads in your arguments; or you may wish to indicate areasinto which you have not entered because to do so would havetaken more space than you had available. Paradoxically, it willoften be worth thinking about what you wish to conclude beforeeven beginning to set pen to paper in preparing your initial plan,since this might influence what you decide to include andexclude from the main body of your essay.
Bear in mind the following useful points also made by Fairbairn andWinch (2011, p. 28). When you are planning a piece of writing,particularly if it is to be an important one such as an article, an essayor an extended thesis, your text can be re-read and analysed overand over again. On each occasion different aspects andinterpretations may occur to the reader. Although your reader will beable to formulate ideas in his/her own mind about what you are sayingor meaning, he/she will usually not be able to ask you any questionsat the time of reading. You will therefore need to ask yourself thequestions your reader might be inclined to ask, in order to savehim/her the fruitless task of asking and not being able to get ananswer. Do not be afraid of patronizing your readers by supplyinginformation, references or arguments that you think are ‘obvious’,because they may not be obvious to him/her.Retrieving and organizing notesIn order to write, you will need to refer to the information you haveassimilated from the literature over the previous months. By nowyou will have collected a substantial amount of relevantinformation that you have ordered in some way so that it remainscoherent, manageable and controllable. Assuming that you haveused a reference management program you will find it easy to sortout and select notes through the search facility, which allows youto retrieve notes that contain particular key words, or that refer toa particular subject or author. Again, there is much to recommendthe practice of keeping the notes short and devoted to only onesubject. The use of paper based notes on cards, sheets of paperor notebooks is unlikely to be manageable for anything but thesimplest research project.Figure 9.2 The use of paper based notes is unlikely to bemanageable for anything but the simplest research project
In drafting an outline of your report or chapter, begin by retrievingthe relevant notes already saved. This should not be very difficult;gradually start feeding the information into the planned sections ofyour work, which is quite simple if you have initially well organizedyour information notes into sections related to the relevant subjectareas, and if the overall plan of your work has been broken downinto a number of small and manageable subject divisions.The information collected as notes and organized under specificsubject areas has been assembled piecemeal over a period oftime. In addition, during information collection you will have writtenout and recorded your own ideas as they occurred to you, andinserted these as notes into the appropriate subdivision of yourmaterial by allocation of key words. By reviewing all the relevantnotes and gaining an understanding of the contents, you candevelop a clear idea of the overall structure of your argument. Youshould then take out the notes from each section (or print out therelevant records) and sift and rearrange them in an order that
provides the basic facts and ideas needed for each section of yourargument. You might in fact discard some notes as no longerrelevant to your work. Now it is time to think and write.Drafting and redraftingOnce you have identified your arguments, formed the structure,prepared an outline and collected and ordered the information youneed, you are ready to start writing. Do not worry if you do not getit right the first time. Remember that you start with a first draftwhich you then revise (often many times). Freeman and Meed(1993, p. 61), when outlining the advantages of drafting as inChecklist 9.4, stress that it is not a linear process, but rather apatchwork that can be adjusted and rearranged as required. It is amultifaceted progression towards refinement.Checklist 9.4 Advantages of draftingYou do not have to get it right first time, and this sometimesmakes it easier to start writing.You do not have to start at the beginning: the introduction can bedifficult to write, so it might make sense to start in the middle.You can change the order if it does not work in practice.You do not have to worry about getting your style and spellingright first time: you can come back and correct later.In addition, according to Newman (1989, p. 31), the main assetoffered by a first draft is that it gives an opportunity to revise thetext. It does this in two ways. First, you can circulate the draft textamongst your tutors or supervisors with a view to gaining theirexpert comment on the draft as a whole and on particular details.But, most important of all, it gives you the opportunity to thinkagain. Having written the first draft you should leave it alone for afew days and get on with some other work. When you return to itlater you may well be amazed at the faults in structure and indetail that leap to the eye as you examine it.You should regard redrafting as a normal activity, not anadmission of failure. Writing and developing ideas always takestime, and until you have made an initial draft, you and others
cannot review your work and change and improve it. There areseveral reasons why you may wish to redraft your work: the lengthmay be wrong, the arguments may not be really clear, there arespelling and grammatical mistakes, or you have developed orchanged your ideas since you originally wrote them. You may alsohave spotted repetitions or omissions, or you want to adjust thetext to fit better with new material that you have subsequentlywritten.Writing can be seen as an iterative process, particularly when thework is long and complicated. How often should you redraft? Thisdepends on the length, complexity and duration of the researchwork. Short reports can sometimes be drawn up in one go andrevised as a whole. With longer texts, a chapter by chapterapproach is normally used initially, and then the whole work islooked at and adjusted to ensure continuity and completeness.For a very condensed piece of work, such as a research proposalor an academic paper, inexperienced writers may have to redraftfive, ten or even twenty times to reach the required standard. Formore experienced writers writing less crucial texts, perhaps onlyone or two redrafts will be necessary.Always keep a saved copy of the original text in case you areunhappy with changes you make and subsequently want to revertto the earlier version. It is essential to remember to save yourwork regularly to avoid losing hours of work should the power beinterrupted or you make a silly mistake on the keyboard.If you are working on your own computer, do not rely solely onsaving to the computer’s hard drive. You should also saveregularly onto another removable storage device such as amemory stick or flash drive or onto the ‘cloud’ so that you stillhave access to your work should your computer develop a fault or,in the case of a laptop, be lost or stolen. Do make the effort tolearn how to use the functions built into the word processingpackage, e.g. cut and paste, spell check, search, replace,formatting, footnoting, insertion of page numbers. Also, make useof the links to your reference management program to speed upcitations and lists of references. Use the ‘online help’ facility if you
get stuck, and don’t hesitate to ask your more experiencedcolleagues for advice.ParagraphsI suggest that at the first draft stage you should aim to write oneparagraph (or perhaps more if necessary) for each of theheadings in your outline. But you need to be careful aboutparagraph content and construction (see Checklist 9.5).Checklist 9.5 Paragraph constructionEach paragraph should contain one main idea from your outline.The first sentence should, if possible, introduce this idea.Other sentences should support the main idea by explaining itmore, giving examples or linking it to other paragraphs.It is very dangerous to introduce a second idea as this willalmost certainly confuse the reader.Ideas and evidenceWhen you are writing your report, you will often wish to putforward other people’s ideas as well as your own. It is thereforeessential that you express these ideas clearly and concisely andgive credit where it is due; for example, if you quote from a bookor an article, make reference to the title and author. Exactly howyou should do this is described later in this chapter. Ideas shouldalways be backed up with evidence, whether from your ownresearch or as quotations from other sources from your reading.Bald statements of fact will not be accepted unless they aregeneral currency (e.g. that the world orbits the sun).Introduction and conclusionAccording to Newman (1989, p. 33), if you are writing a self-contained report, the introductory section should comply with thepoints in Checklist 9.6.
Checklist 9.6 Introduction to a report1. State precisely the area of study. This is very important; you canhardly be criticized for failing to cover something outside thearea of research you have defined.2. Explain why the research is being carried out.3. Draw attention to the originality and the conclusions.4. Point to the inadequacy of the work so far carried out in this areaof research.5. Describe the methods of research.6. State what special difficulties have been encountered.When you feel that you have reached a reasonable structure foryour report, try to draft an introductory section, which will becontinually improved and gradually rewritten as the researchprogresses. Writing the conclusion is more complex. Newman(1989, pp. 33–4) suggests that a clear distinction should be madeat an early stage of the research between, on the one hand, thepossible conclusions that may emerge or the hypotheses that arebeing tested (which must necessarily be delineated as a part ofthe structure) and, on the other hand, the actual research-basedconclusions which can be written only when the research hasactually been carried out. The conclusion should also include anassessment of the soundness of the whole written report, withperhaps suggestions of how further research in the same subjectarea might be carried forward.When writing an individual chapter in a thesis or a report, similarconsiderations apply to the introduction and conclusions, butthese must be tailored to the specific needs and content of thechapter, while referring to, and making connections with, the otherchapters. It is difficult to make all the connections before otherchapters are written, so be prepared to review the work as awhole when the full draft has been completed in order to forgethese connections.IllustrationsAt the first draft stage, you should also think about whether, andhow, you should illustrate what you write. Depending on what you
are writing, you may wish to draw your own pictures – diagrams,charts, maps, graphs or tables – or to include copies of picturesfrom other sources. Checklist 9.7 gives criteria for illustrations.Checklist 9.7 Use of illustrationsAny illustrations should:be relevant to what you are sayingbe clear and simple to understandadd usefully to what is already therebe cross-referenced and explained in the textbe clearly numbered and captioned and, if from other sources,fully referenced.QuotationsWhen you are drafting your research report, make sure that youutilize only necessary quotations; over-weighting your report withthem might possibly diminish its originality, and consequently itsvalue. Interlarding a report indiscriminately with quotations andfootnotes is a cultural affectation that some inexperienced writersassume to conceal shoddy workmanship. Pasting numerousquotations into an authority-laden mosaic does not create anacceptable research report. Strings of these passages reveal thatyou are little beyond the note-taking stage of your work. Aresearch report is a creative effort – a synthesis of what you haveread, observed and mentally ordered into a new pattern – ratherthan a mere compilation of other people’s work. You may usequotations, but use them sparingly and purposefully.If you wish to refer to another person’s work by citing his/herwords directly, you should give a precise location for them, i.e.include the page number of the book/article. This will test thequality of your information storage and retrieval system: many afrustrating hour or even day has been spent looking for theprecise location of an essential quotation months or years after itwas noted! If you are citing less than a certain number of words(say up to 30) you should put the quotation within your text, usingquotation marks to show to the reader that you are borrowing
these words from another person. If you have longer quotations,begin a new line, perhaps use single spacing and perhaps adifferent font, and indent the quoted passage slightly from bothsides. The use of quotation marks is not then necessary. Thereference should follow the quotation, strictly following therequirements of the chosen citing system (more about this later inthe chapter).PunctuationProper use of punctuation helps the reader to understand themessage you intend to convey. Punctuation is an indication of theway in which you should speak the text, so it is often a good testto read it out aloud, which will check that the punctuation helps toindicate meanings conveyed by tone of voice, gesture, stress onparticular words or syllables, pauses and so on. On the otherhand, overuse of punctuation can also cloud meaning.Figure 9.3 The use of punctuation can also help to indicatemeanings that in speech are conveyed by tone of voice, gesture,stress on particular words or syllables, pauses and so on
You should make sure that you know how to make correct use ofthe following: full stop, capital letters, comma, semicolon, colon,question mark, exclamation mark, apostrophe, brackets, invertedcommas and hyphen. You should also learn to paragraphappropriately and to use indentation to make certain sections ofyour work stand out. The use of the dash – for example, in makingan aside – is also worth learning. For further information, includingdescription, function, use and examples of the above-mentionedtypes of punctuation, you can refer to Fairbairn and Winch (2011,pp. 83–116) and other books treating the use of punctuation intexts.You will notice that there is sometimes conflicting advice on theuse of punctuation, so which approach should you take? Youdecide! The main issue is that you provide clarity and consistency,and ensure that the style you adopt is appropriate for your type ofreader. As and example of advice, the short summary in Checklist
9.8, based on guidance notes prepared by Ackrill (1994) forarchitectural research students at Oxford Brookes University, liststhe main points.Checklist 9.8 Use of punctuation1. Commas are the smallest interruption to continuity, enabling thereader to pause. They mark off one thing from anotherpreventing ambiguity and misreading. They are not neededbefore the last item of a simple list which is usually preceded by‘and’. Try reading the following without a comma, to see why oneis necessary: ‘After the trumpets blew the walls of Jericho fell.’2. Semicolons give a greater break, and usually separate twopoints of equal significance, e.g. ‘The buses were late; the trainsusually ran on time.’3. Colons give an even bigger gap and precede a clause or phrasewhich explains or expands what has just been said, e.g. ‘Theweather was unusual for April: never since the beginning ofrecords had it been so cold and sunless.’4. Exclamation marks, unless part of a quotation, are notappropriate for a thesis.5. Quotation marks are needed around all short (say less than 30-word) direct quotations. Single marks are used, except for aquotation within a quotation, when double ones surround theinner quotation. If the end of the quotation is the end of yoursentence, a full stop should be used outside the final quotationmark if the quotation itself is not a full sentence, but inside if it is.Omissions within a quotation should be shown by three dots(ellipsis): when omissions spread over a full stop, some writersshow the full stop before or after the three dots as appropriate.Long quotations, i.e. those spreading over say more than 30words, should be indented, possibly typed in single-space type,and not surrounded by quotation marks.6. Brackets set off something in parenthesis, something said in anundertone, e.g. ‘The building (considering its mode ofconstruction) was in a good state.’ Round brackets should beused unless the parenthesis is something you have insertedwithin a quotation, when square brackets [ ] should be used.SpellingWhen drafting and revising your work you should, according toFairbairn and Winch (2011, p. 165), learn to read it in two ways.
First, you should be able to read it as if you have never read itbefore at the level of content. Does it make the points it issupposed to make? Does it make them succinctly and clearly?Secondly, you should read your work in order to attend to format –to spelling, punctuation, grammar and so on.Spelling is very important. It is common courtesy now to use thespell checker on your draft before submitting it to someone toread, to avoid them spending unnecessary time and effortcorrecting obvious spelling errors. You must take the time to checkyour work thoroughly. Even when the spelling errors are pointedout after reading a draft, it is not unusual for many of the sameerrors to reappear in a subsequent draft. It is worthwhile gettingsomebody to read your writing with the intention of pointing outspelling errors and even places where he/she is unsure whetheryou have spelled a word correctly; you can check the queriedspelling in a dictionary or the spell checker in your wordprocessing program.Some of the most common spelling mistakes arise fromconfusions between words that sound the same but are spelleddifferently and mean different things. A little thought should helpyou to avoid these confusions in spelling. Words that sound thesame but have different meanings are called homophones (e.g.‘their’ and ‘there’). If you use a computer, look out for spellingmistakes caused by typing too fast. All word processing packagesinclude a facility for spell checking; however, because theseperform a merely mechanical function they have their limitations.For example, you might type a word that is correctly spelled but,because it is not the right word, does not give any meaning to thesentence – say you have typed ‘of’ instead of ‘for’. When you useyour spell checker it will not reveal your mistake because, for it,‘of’ is a correctly spelled word. Mistakes such as these mayremain in your final draft unless you habitually read your workcarefully for spelling errors before deciding it is finished.As your drafts become more refined, you need to start thinking abouthow the text and illustrations will be presented.
Tables and figuresThere are conventions that will help you to keep presentationclear and consistent. Here are some of them as summarized inAckrill (1994).TablesA table should follow the first reference to it in the thesis. If it willnot fit into the remaining space on a page, place it at the end ofthe paragraph in which it is first mentioned. All tables should havea title and be numbered within each chapter. The third table inChapter 2, for instance, should be shown as Table 2.3.A separate list of tables should appear after the list of contents atthe beginning of the thesis.Where the table is not your own work, the source should be givenat the foot of the table, in the same form as the references.A two-line gap must be left before the table number and title, anda similar gap at the end of the table. If the source is cited, this gapis after the named source. Where tables are unwieldy, they go intoan appendix. Consult your tutor or supervisor about how theyshould be numbered.The number and title of a table may appear at the side or in thecentre of the page. Wording may be in capitals, or with capitals forimportant words, or with no capitals. But all positioning andlettering must be consistent throughout the thesis or report.FiguresFigures (i.e. small graphs or charts) are presented in the sameway as tables, except that they must have their own set ofnumbers, e.g. the third figure in Chapter 2 will be Figure 2.3.Numbering
All chapters must be numbered, and sections within chapters mayalso be numbered, e.g. the first section in Chapter 2 can benumbered 2.1, and subsections within it 2.1.1, 2.1.2, etc. Youshould think carefully about going beyond the third level ofnumbering, as it becomes rather unwieldy and akin to a legaldocument.The items in any lists should be consecutively numbered orotherwise indicated, e.g. by bullets. They should be in the samegrammatical form, e.g. all beginning with a noun, or a participle, oran infinitive. Initial capital letters and punctuation should follow aconsistent system. The lists themselves do not need to benumbered by chapter in the same way as the figures and tables.Bibliographies, references and footnotesWhen reviewing the literature relating to your research project,you should, if at all possible, include recent information. This byno means implies that older information is not relevant, but thereview should be up to date. A review of the literature that had noentries less than two years old, for example, would be suspect,and could show a serious gap or deficiency (Wiersma and Jurs,2008).When information is reported from a source, it must be adequatelyreferenced. It is important that whenever you cite others, youmake available sufficient information to enable the reader to lookup such work for him/herself. Below is a brief description of twowidely acceptable methods, either of which you can safely adopt.Neither, in my experience, has ever been questioned by anexaminer or a reviewer.Note, though, that a number of alternative formats are used inacademic publishing. Different journals and other publicationsadopt their own particular systems, and insist that their authorscomply with them. It is possible to convert your bibliographiesfrom one format to another very easily if you use suitablereference management software (such as RefWorks or EndNote)when you compile your initial bibliography.
List of referencesAll reports and theses must include a list of references. This is alist of all written materials found useful in writing the report orthesis. It must include everything from which direct quotations aremade, or to which the text refers, or which is the source ofreasonably specialized information. It does not normally includeentries in encyclopaedias.Whenever you read something useful, make a note of the authors’names, and indicating whether the first and any other namesappear on the title page in full, or are referred to only by initials.Then record, if an article, the journal in which it was published,together with its volume and issue number and the date ofpublication (including the month of an issue, or the date of anewspaper article). If a book, record the title, the location andname of the publisher and the date of publication. The page orpages of an article should also be noted, e.g. pp. 34–48. Thepresentation of this information in the list of references shouldtake one of the two forms explained below.QuotationsReferences should be given for all direct or indirect quotations,and in acknowledgement of someone’s opinions, or of a source offactual information that is not general knowledge. The first methodillustrated here is the British Standard (numeric) system, useful forshorter texts. The second is the Harvard (author–date) system,which is ideal for longer works containing several chapters.Whichever method is adopted, it should be followed in detail.Examples of the two methods are given in Box 9.1.British Standard (numeric) systemA reference is cited in the text by a number in either round orsquare brackets or as a superscript numeral. The referencenumber always precedes the punctuation marks, e.g. ‘… of thislearned work (27)’. and not ‘… of this learned work. (27)’.
The numbers should run consecutively through the text for eachnew reference introduced. References appearing in tables shouldbe numbered in sequence at the point at which the table is firstcited in the text. You can, of course, repeat a number later in thetext if required. Because of the need to keep the numberingconsistent, this system becomes unwieldy in longer texts that aresubject to several revisions. That is why in a multi-chapter reportsuch as a dissertation or thesis, a separate set of numbersbeginning with (1) should be used in each chapter.At the end of the chapter (or report) the works to which thenumbers refer are listed in numerical order. Each entry beginswith the author’s last name, followed by a comma and by his/herinitials as they appear on the title page of the work in question,with full stops. Titles of books are in italics; titles of articles andjournal papers, or sections of a book edited by others are innormal script. The title of an article is followed by the name of thejournal, and this name is in italics, followed by the volume and (inbrackets) the issue number, the date of publication, and the pageor pages on which the information is found, all separated bycommas, with a final full stop. After the title of a book, its place ofpublication is given, followed by a colon, the name of the publisherand date of publication, followed by a full stop.References to other types of material, such as newspaper articles,videos, websites and conference papers follow a similar patternas can be seen in Box 9.1.Harvard (author–date) systemCitations within the text consist of a brief description of thereference enclosed in round brackets, and full details of thereferences are given in the list of references at the end of thework. In a multi-chapter work there are no lists of references at theend of chapters.A citation should be inserted at the exact point in the documentwhere you refer to someone else’s work. It consists of the lastname of the writer, followed by the year of publication and, ifrelevant, the page number, all separated by commas. Should the
bibliography include more than one work written in a given year bya given author, the works are lettered a, b, c, etc., and this letterfollows the year of publication, e.g. (Huxtable, 1976b, p. 43). If theauthor’s name is mentioned in the text, then it can be omitted fromthe citation. The page number is essential when you reference adirect quotation, and desirable if you are discussing a specificissue that appears on a particular page or pages in thepublication.The entry in the bibliography begins with the author’s last namefollowed by a comma, then his/her initials followed by full stops. Ifthere are several authors, they are similarly listed, divided bycommas, and with ‘and’ added before the last author. The year ofpublication and, if necessary, a, b, etc. follow, enclosed in roundbrackets. The title of the book or article is then given, with capitalsfor only the first letter of the first word and all other words thatnormally take a capital (e.g. place names). The title of a book is initalics, but not that of an article. The name of the journal in whichan article appeared then follows, and is in italics; volume and (inbrackets) issue number or month of publication, and the page orpages an article covers are then given, shortened to the fewestdigits, e.g. pp. 21–3 not 21–23. The entry ends with a full stop.The entry for a book ends with the place of publication followed bya colon and then the name of the publisher, followed by a full stop.Again, this pattern is largely followed for other publications, as canbe seen in Box 9.1.Box 9.1 Examples of referencing methodsBritish standard (numeric) systemBook reference:1. Norris, A.B. Why it doesn’t fall down: a study of single spanarchitecture. London: Lothbury Press, 2017.2. Herring, G.R., Kirkner, L.A., eds. Monetary easing. 2nd ed.Oxford: Oxford University Press, 2010.Article reference:
1. Blenkinsop, P.J. Nothing is for nothing: the cost of charities.Journal of Social Economics 14(3), 2014, pp. 46–51.Video reference:6. Gregory, J. Lost cities. Produced and directed by HenrySilversmith. 25 min, 2008. Video.Harvard (author–date) systemBook reference:Herring, G.R. and Kirkner, L.A., eds (2010) Monetary Easing, 2nd ed.Oxford: Oxford University Press.Norris, A.B. (2017) Why It Doesn’t Fall Down: A Study of Single SpanArchitecture. London: Lothbury Press.Article reference:Blenkinsop, P. J. (2014) Nothing is for nothing; the cost of charities.Journal of Social Economics 14 (3), pp. 46–51.Web page:Conservative Party (2017) Manifesto. Available at:www.conservatives.com/manifesto (Accessed: 15 March 2019).Conference paper:Hamilton, W.S. (2009) Reformulation of indices for future expansion.In: Advances in Economic Theory: Proceedings of the 23rdInternational Conference on Economic Affairs. University of Durham,15–19 August. Durham: Department of Economics, DurhamUniversity, pp. 454–67.Box 9.1 gives only a few examples of types of references. For fullexamples of every type you should refer to the books suggestedat the end of this chapter. Your library will probably be able toprovide you with a short guide, and if you search on the Internet,you will find that most universities publish online guides to theirpreferred style. You will notice from your reading that publisherstend to have their own sets of style rules with regard to things likequotation marks, full stops and capital letters. Look at the way
references are listed in this book and see what small points ofdifference you can spot from the way the details are given here.FootnotesFootnotes, in the literal sense of notes at the foot of the page,should be avoided if possible. Where a non-English word is used,explain its meaning in the text when first mention is made of it. Ifthe definition (e.g. of a technical or professional term) is so longthat it interrupts the flow of the thesis, it may be put in a note atthe bottom of the page. A superscript (1) after the term in the textand preceding the explanation at the bottom of the page is all thatis needed.Motivation and maintainingmomentumResearch projects, whether a PhD, a master’s dissertation or afunded project tend to be stretched over a long time, months oryears. How do you keep going over so long a period?The main advice is to break the project down into tasks which canbe accomplished in turn so that your deadlines are much closertogether. Admittedly, research is seldom so straightforward thatthe tasks can be cleanly divided, but each can be progressed to acertain identifiable point, e.g. to prepare a draft of a surveyquestionnaire for comment and a pilot study. Combining this list oftasks with a timetable will provide you with a framework with thesequence of tasks and the timings, ensuring that you can checkon your progress within the given time-frame of the project. Insertregular milestones indicating the major stages in the research.When you do a PhD these will be determined by the academicapproval stages and progress reports required by your university,and funded research projects are also divided into stages foradministration and funding purposes.Do keep in touch with your supervisors. They are there to helpand motivate you, but you will only get the best out of them if you
make the effort to engage them in the progress of your project.They can help you get over problems that are blocking yourprogress by using their experience of other research projects.Some supervisors are very busy with their other work, in whichcase it needs a certain amount of insistence on your part to claimsome of their time (to which you are entitled).It is not only your supervisor who will be very busy with their otherwork – this will probably be the case with you too! Life has to goon beside your research work, and family, occupation,relationships and other commitments can take great chunks ofyour time. One thing about research work is that it is very difficultjust to do a few minutes here and there; it takes time to get backinto the groove, to assess where you left off and to immerseyourself again into the current tasks. It is therefore best if you canarrange for regular periods during the week when you can workundisturbed for several hours. Let everyone know that you areunavailable during those times, and resist the temptation to doyour general emails and get caught by other distractions.Everyone gets bored with doing a repetitive job, and some of the tasksin research are repetitive, such as sorting out large amounts of data.You can break up these tasks by doing something more creative inbetween, such as writing your draft chapters or organizing yourfieldwork. But don’t think that time dealing with your collected data iswasted. The more you work with your data, the more you ‘immerse’yourself in it, the more you will be able to spot trends, inconsistencies,similarities, etc. which will guide your analysis.The most difficult time when doing a PhD is near the end whenyou draw up the overall conclusions and recommendations. Thisis where you need to demonstrate that you have made acontribution to knowledge and is the ultimate test of work at thislevel. The final deadline is looming and you are tired and there stillseems so much to do. Unfortunately this is the reality and a test ofyour determination to complete. Support from your supervisors isessential here, so make sure that you discuss your conclusionsand submit drafts of your final chapter for comments. Your
supervisors should be familiar with your previous chapters and willbe able to make informed judgements about the quality andconsistency of your conclusions and recommendations. It oftendifficult to stop and say ‘this is enough’ – one can always improveon the work. Again, your supervisors will be able to judge whetheryou have done enough to satisfy your examiners, and can adviseyou when to wind up your research.Other types of writing you mightneed to doPresentationsYou will often be asked to do a presentation when you take part ina student conference, a seminar or other internal event, anddefinitely if you are successful in getting your abstract approved ata national or international research conference. If your researchproject is of general public interest, you might also be asked toprovide an account of your project at a public event.There were 266 million results from Googling ‘how to dopresentations’, so there is no lack of advice on how to compileand present them! To save you the bother of looking these all up,here are some of the major nuggets of advice, emanating fromsome of the most famous experts at communication, such asSteve Jobs, co-founder of Apple; Guy Kawasaki, a Silicon Valleyventure capitalist; Tony Robbins, a life and business strategist;Garr Reynolds, a communication expert and Chris Anderson,curator of TED talks.First, the visual content of the presentation (using PowerPoint,Keynote, Google Slides, Prezi, Adobe Spark, etc. seehttps://zapier.com/blog/best-powerpoint-alternatives/ for moredetail). Follow the advice given in Checklist 9.9.Checklist 9.9 Advice on designing effective presentations
The 10/20/30 rule: a good presentation should have no morethan ten slides, should go on for no more than 20 minutes, andthe font size should be no less than 30 pointStructure it like an essay: introduction, body and conclusion.This is also known as the ‘what, why and how’.Don’t try to cover too much ground: concentrate on your uniquecontribution rather than describing the whole subject.Keep the text short: Seth Godin insists on no more than sixwords on a slide. EVER! This is hard to follow, especially if youlimit yourself to ten slides. But the principle is sound.Combine pictures and text: this is one of the best ways to helppeople engage with and retain information.Talking comes first: use the presentation to support yournarrative and illustrate the points you want to make.Only half of the job is done when you have designed an effectivepresentation. The other half is how you do the presenting whenyou are standing in front of your audience. Checklist 9.10 providesyou with some useful guidelines on how to put on the right act.Checklist 9.10 Advice on presenting your presentationBeing nervous is normal, but it helps if you are reallyenthusiastic and well prepared. Try the presentation out on somesympathetic colleagues beforehand if you can.Take your audience on a journey, work out where you want tostart and where to end. Know the level of knowledge of youraudience to make the introduction compelling.If you cannot memorize your talk, use small prompt cards toremind you of the structure. Avoid just reading what is on thescreen – let the audience do that for themselves!Make eye contact with the audience and smile (not all the time!).Avoid turning your back to the audience, and use a remotecontrol to change the pages of the presentation to avoid beingrooted to the spot.Don’t gesticulate and pace around too much – use yourmovements to provide accents to your talk.It is well worth doing some practice beforehand, especially if youare not used to doing presentations. Time yourself to ensure thatyou do not overrun your allotted slot, and make sure that youleave enough time for any questions from your audience. If you
can get some friends or relatives to listen to your practice session,all the better, as they will notice any shortcomings or oddities inthe content or your style of presentation.PostersPoster displays are a common feature of conferences andexhibitions. Presenting your research in the form of a posterrequires a good sense of design and skill at concise description.You will not be able to describe every detail of your project, so youneed to judge what are the most interesting aspects of theresearch and findings. Posters are a form of visualcommunication, so make good use of pictures and displays ofvarious kinds – graphs, charts and diagrams. Keep the text to aminimum (maximum 1,000 words total) and make the font largeenough to read from a distance of about three metres. Keepplenty of white space around the elements of the poster. A goodtest for legibility is to print the poster out at A4 size and see if youcan read it all easily.The conventional poster will have the following sections:TitleAuthor(s) and address(es) and logo of organization youbelong toIntroduction to the subjectMaterials and research methods usedResultsConclusionsLiterature citedAcknowledgementsFurther informationA typical layout is shown in Diagram 9.1.This is assuming that you have completed a research project andhave results to report. However, student conferences aregenerally about work in progress, so the emphasis is more on theprocess rather than the results. Diagram 9.2 is an example of onesuch poster prepared by Catalina Morales-Maya after nine months
of her PhD study, using a more artistic approach than the simplerectangular text box design.Diagram 9.1 Standard layout for a posterYou can use Microsoft PowerPoint to set up your poster. Beforeyou start, check with the guidance that is usually given by theorganization running the event.
It is best to work out several drafts of the layout options, sketchingby hand what and where the text and illustrations will be and howmuch space you will allocate to each. Apart from the title andauthors and addresses, make a box for each of the sections andarrange them in the order that you want people to read them –either horizontally, vertically or following a route around the poster.There are freely available poster templates on the Internet, onegood example can be found onwww.posterpresentations.com/free-poster-templates.html. Checkyour own university or organization website to see if it provides itsown templates. It is also a good idea to look at poster produced byother people – you will soon get the idea of what works best –search for ‘poster presentation examples’ on Google.Diagram 9.2 Example of a ‘work in progress’ poster
Source: Catalina Morales–Maya
You may also be asked to give a short talk to present and explainyour poster. This requires you to decide on the issues you want tohighlight as you will not have the time to present the whole projectin detail. Leave time for any questions.Conference papersA paper prepared for a conference is not intended to be read outat the conference itself, but is added to a collection of all thesubmissions that is published as a record of the proceedings ofthe conference. The method of conveying your project during theconference is by doing a presentation or displaying a poster. Thismakes it possible for many research projects to be covered in theday or days available.If you want to take part in a conference and have your paperincluded in the proceedings publication, you will first have tosubmit an abstract of your paper so that the organizers can checkwhether the subject is suitable for the conference themes and tojudge the quality of your writing. Make sure that the abstractcontains a summary of the subject matter, the research methodsand the conclusions.Once approved, you will need to check on the requirementsrelating to the papers to be submitted. This will relate to thelength, structure and format of the paper, including the style ofreferencing and type of illustrations. Normally, the paper yousubmit will be subjected to review, though not as stringent a peerreview process as for a paper published in a refereed academicjournal (see next section). Follow the normal format of a researchinformation publication as detailed for a poster in the previoussection.If your paper is judged to be of outstanding quality, it might beselected for publication in a relevant refereed academic journal. Ifthis is the case, you will have to revise the paper to be in line withthe requirements of the journal, and it will be subjected to peerreview of two or three experts in the subject. See the next sectionfor more details in what is involved.
Research papers for refereed academicjournalsGetting your paper published in a refereed academic journal is thepinnacle of research dissemination, particularly if it is a highlyrespected publication. You need to have some convincing resultsfrom your research in order to be able to prepare a suitable paper,so this inevitably means that you will be at or nearing the end ofyour project. The best way to ensure that your submission will beconsidered for publication is to select a journal whose title isnearest to the subject of your research. You can check on thejournal Web page what subject areas are covered, and in pasteditions what sort of papers have been accepted. The relevant‘instructions for authors’ will provide you with details of what isrequired, and a template is often provided to ensure consistency.To make sure that you are going in the right direction, send theabstract of your planned paper to the editor of the chosen journaland ask if it is in line with their specializations.There are usually three different types of papers that can beaccepted: full-length original articles with typically 5,500–7,500words (including references) which give details of originalresearch; review articles of the same length which review thehistory and literature of a certain subject to provide an overview ofthe ‘state of the art’; and shorter articles of opinions, viewpoints orcomments on particular relevant issues of about 1,000–2,000words. The first is the most prestigious, the second a good use foryour literature review, the third really only for fun.Do make sure that you follow the instructions to the letter, andparticularly adjust your referencing style to that required. Do getadvice and comments on your draft paper from more experiencedresearchers before you submit. After submitting, usuallyelectronically, you will have to wait to hear whether the paper hasbeen selected for review, and if so, then for the feedback from thereviewers. Unless your paper is absolutely outstanding, you willhave to make some minor, or perhaps major, amendments beforebeing accepted for publication. At worst, your paper will berejected as unsuitable for publication in that journal, eitherbecause the subject is not in line with the journal’s aims, or the
quality of research or writing is not sufficiently high. Don’t be toodisappointed with this; you can always, after taking account of thecomments in the reviews, try again with a different journal.Although it is very satisfying to get your paper published (and veryuseful for your CV) it is only really of value if it gets cited in otherpeople’s publications.You don’t succeed as a scientist by getting papers published.You succeed as a scientist by getting them cited. (Schimel,2011, p.3)There are several ways you can improve your chances of gettingyour paper(s) cited. These are listed in Checklist 9.11Checklist 9.11 How to help get your papers citedMake sure that your paper is accessible for free (open access).Cite your own papers in your subsequent ones.Present your research at conferences.Network within your subject area to become known as an expert.Add your most recent publications as a strapline to your emailmessages.Join online research communities such as ResearchGate andMendeley, and get a Google Scholar Citation Profile.Use social media to announce your publications.Carefully choose your key words.Be consistent in using your same name format in all yourpublications to ensure they all appear when searching by yourname.Progress reportsDuring your research project you will inevitably be asked, everynow and again, to produce a progress report to record what youhave done and how well you are keeping up with your programmeof work. This a straightforward document and a useful pause forthought for yourself as well as for others, as it forces you to take
time to review your work and progress, as well as reflect on theway forward.When doing a PhD, there is generally an annual review that alsoinvolves your supervisors, and may even include an interview withthe head of research. For example, at Oxford Brookes University,research students have to produce a 1,000-word progress reportthat must be approved by their supervisors before submission tothe Postgraduate Research Tutor (faculty head of studentresearch) together with a sample of the student’s written academicwork, and a report from Turnitin (a plagiarism checker program)following analysis of it. Questions are also asked about thetraining undertaken by the student and involvement inconferences and presentations. A timetable for completing theremaining tasks must also be provided. The supervisors are alsoasked to submit a report, commenting on the activities andprogress of their student, whether he/she should be allowed tocarry on and, if nearing completion, an estimated time beforesubmission.Sponsored research projects, especially those lasting for severalyears, also require regular progress reports to gauge whether themilestones have been achieved and what future tasks will beundertaken (e.g. EU international projects require updates oneach work package every three months).ArticlesThere are three other ways you might be asked to explain inwriting, articles, conference papers and refereed journal papers.The most informal is an article in a magazine, newspaper,newsletter or trade journal. Here, the strict rules of academicwriting do not apply. You are acting more like a journalist than aresearcher. The main feature of this kind of writing is that you willbe addressing the general public so you must assume that theyknow nothing about your subject.Newspaper articles are generally rather short and follow a certainpattern. The headline is an important feature for attracting thereader. Your suggested headline should be short and inform the
reader about ‘what’, ‘where’ and ‘when’. It is usually best to decidethis once you have written the article so that you pick up on themost interesting aspect. The first paragraph is a couple ofsentences is called the ‘lead’, and this is its function – to lead thereader on to persist with reading the rest of the article. It sets thetone and style of the article and provides the most importantinformation on the subject to capture the interest of the reader.The rest of the article fills in the details. Back up assertions withreferences to authoritative sources to gain credibility. Keep thewriting style informative and direct. Round off with a conclusionbased on the information presented.InformativeMagazine articles are longer, so you have more time to elaborateand perhaps discuss and compare alternative views. Do check onany advice or instructions provided by the magazine that you wantto publish in. If these are not provided, look at articles in backcopies to analyse the nature and form of previous submissions.As with any piece written for the enjoyment of the reader, youneed to highlight an interesting angle on the subject. Magazinesare aimed at a certain readership, so make sure that you havechosen the appropriate publication for your subject. Use the sameformat – introduction, main topic, discussion and conclusions.ConclusionsThe main motivation for doing a research project is the urge toknow and understand more about the chosen subject. That is whyit is of prime importance that you choose a subject that fascinatesyou, one for which you want to make a contribution that will be ofuse to other workers in the field. The satisfaction of making thisprogress and becoming a world expert in your chosen subject willdrive you on to successfully complete your project.No one pretends that doing research is either straightforward oreasy. But with careful planning, flexibility and imagination when
faced with difficulties, and an organized work regime, you cansurmount the challenges as they occur.Different forms of writing and presentations of your work requiredifferent skills, all of which will be valuable in your ‘life after yourproject’. So do not believe that you will be wasting time when youneed to divert from your research work to concentrate oncommunicating about it. Contacts made through disseminationcan be valuable for your future career.The next steps: getting to write yourthesis or dissertation or researchreportYou will have done plenty of note-taking and writing as preparationfor your proposal, but writing your thesis etc. takes this to anotherlevel. Here, I will describe the steps to take for writing a PhDthesis; the same will apply for dissertations or research reports,but on a smaller scale.Checklist: How to start writing your thesis or dissertationThe first step is to plan out the structure of your thesis. PhD theses insocial sciences and many other areas of research tend to follow afairly standard format in the arrangement of chapters:Chapter 1. Introduction to the subject, identification of the mainresearch questions or hypotheses, and a guide for the reader tothe structure of the thesis.Chapters 2, 3 and possibly 4. Literature review that explores thehistory and latest thinking in the various relevant aspects of thechosen subject. The aspects might be political, economic, social,cultural or technical. The culmination of these chapters is theidentification of the specific research questions or hypothesesthat will be the focus of your own research project.Chapter 4 or 5. The research methods you used (notice the pasttense – this is a report of what you have done, unlike theproposal where you are describing what you are planning to do).They will have to be in enough detail to allow someone else toreproduce your study. You will find that different research
methods for collecting and analysing data will be requireddepending on the aspects you are investigating. I have foundthat students are good at describing how they collected data, butfind it more difficult to describe the analytical systems they used.This second aspect is of great interest for your examiners, soconcentrate on providing a detailed account of what you havedone.The next two or three chapters are dedicated to describing yourresearch activities – data collection, data analysis and drawingconclusions. Each chapter deals with a different aspect of theresearch – usually responding to the different research questionsetc. Alternatively they may be organized to describe the differentstages of the work, e.g. quantitative research to get the overallpicture, then qualitative research to delve deeper into the mostinteresting aspects revealed.The last chapter draws together the conclusions from theprevious chapters and provides the answer to the main researchquestion (as reflected in the title of the thesis) or assessment ofthe verification of the main hypothesis. It is rounded off with anaccount of the limitations of the research and suggestions forfurther research.The next step is to set up a list of chapter headings and, as far as youcan, subheadings. This will provide you with a framework for your textand figures. If you have carefully devised a system for storing yournotes they will be easy to retrieve using key words, and you will bewell placed to start creating your first rough draft of your literaturereview chapters. Simply copy and paste your notes under the relevantsubheadings to provide a very rough draft of the chapters – a goodway to get over the daunting empty page situation! Then work towardsdeveloping your argument for the reasons for the research anddefinition of the research questions or hypotheses. Ignore Chapter 1 –best to leave this until the end when you have written your otherchapters and know what you are introducing.The research methods chapter will depend on the nature of yourresearch questions and you can only draft this once you are clearexactly what you need to find out. Refer to the recommendations inresearch methods books to substantiate your choice of data collectionand analysis methods.The remaining chapters rely on the actions and outcomes of yourpersonal research efforts. Once you know what you have done,devise a sequence of subheadings to provide a structure to thechapters and a framework for inserting the description of youractivities and how these lead to your conclusions.
You can only write the last chapter after you have completed theprevious ones, as it relies on the results of those in order to wind upthe whole project report.Keep referring to your timetable to check on your progress through thedifferent stages of your research. This will help to indicate when youshould stop working on the various tasks (you could always do more!).Consolidation and assessmentYour supervisors are there to check your work and guide youthrough the process – do make use of them! Discuss with themthe structure of your thesis and send drafts of your chapters tothem as soon as you have substantially completed each one. Beprepared to edit and change your text in agreement with theadvice you receive; several redrafts may be needed until you geta strong and consistent argument, based on evidence, throughoutthe thesis.Your supervisors will also be of invaluable help when you preparethe reports for the various stages of your research, e.g. progressreports, registration, transfer from MPhil to PhD, and for when youprepare conference papers or posters or a paper for a refereedjournal.Further readingWriting good English is not that easy, particularly if English is notyour first language. I have suggested a few books that will helpyou to check that you present your work as clearly and correctlyas possible.Here is lots of advice on writing. I have provided short notes onthe contents and style of the books.Oliver, P. (2013) Writing Your Thesis, 3rd edn. London:Sage.
Designed to help postgraduate and research students with theprocess, preparation, writing and examination of their theses.Murray, R. (2011) How to Write a Thesis, 3rd edn.Buckingham: Open University Press.Moving beyond the basics of thesis writing, this book introduceswriting techniques such as free writing, generative writing andsnack writing. Issues such as working out the criteria for yourthesis, writer’s block, writing a literature review and making notesinto a draft are also covered.Berry, R. (2004) The Research Project: How to Write It, 5thedn. London: Routledge.Guides the reader right through from preliminary stages tocompletion and sets out in clear and concise terms the main tasksinvolved in doing a research project.Woods, P. (2005) Successful Writing for QualitativeResearchers, 2nd edn. London: Routledge.Really thorough on all aspects of writing: getting started andkeeping going, organization, alternative forms of writing, style,editing, etc.Mounsey, C. (2013) Essays and Dissertations, 2nd edn.‘One Step Ahead’ series. Oxford: Oxford UniversityPress.Provides students of all levels with essential and easy-to-followguidance on how to plan, research and write essays, dissertationsand exams. Other titles in the series might be useful: Editing andRevising by Jo Billingham, Punctuation and Spelling by RobertAllen, and Words and Writing Reports by John Seely.The following books are about writing papers and reports:
Hall, G.M. (ed.) (2012) How to Write a Paper, 5th edn.London: BMJ Publishing.Aimed at medical writers, but the same lessons should be learnedfor any scientific writing.Schimel, J. (2011) Writing Science: How to Write PapersThat Get Cited and Proposals That Get Funded. Oxford:Oxford University Press.Although aimed at scientists, it gives good advice on the writingand editing process, and the importance of getting your papercited in other publications.Bowden, J. (2011) Writing a Report, 11th edn. Oxford:How To Books.This book is comprehensive in its detail and contents, coveringtype style, fonts, writing style and even advice on binding andillustrations.The following are two examples of books that specialize in writingabout particular subjects. I suggest you do a search for similarbooks in your own subject, which is easily done using key words(e.g. ‘writing’ and ‘geography’).Hamilton, A. (1989) Writing Matters. London: RIBAPublications.Part 1 deals with writing in general; after that it gets more intowriting for architects.Fabb, N. and Durant, A. (1993) How to Write Essays,Dissertations and Theses in Literary Studies. Harlow:Longman.Especially for those who specialize in writing about literature.
And here are two books to help you sort out the finer points ofwriting in English:Evans, H. (2000) Essential English, 2nd edn. London:Pimlico.Get your spelling and grammar and construction right with this.Aimed at journalists, editors and writers.Trask, R.L. (2002) Mind the Gaffe: The Penguin Guide toCommon Errors in English. London: Penguin.See whether you make common mistakes. Very useful too ifEnglish is not your first language.For citing and referencing, first consult your own library about anyguidance they might produce. They should be ideally suited toyour needs. If you really need more information, here are somebooks:Bosworth, D. (2004) Citing Your References: A Guide forAuthors of Journal Articles and Students Writing Theses orDissertations. Thirsk: Underhill.British Standards Institution (1990) Recommendationsfor Citing and Referencing Published Materials. BS 5605.London: BSI.University of Chicago Press (2017) The Chicago Manualof Style, 17th edn. Chicago: University of ChicagoPress.Li, X. and Crane, N. (1996) Electronic Styles: A Handbookfor Citing Electronic Information, 2nd edn. Medford:Information Today.And this one is for some light relief!
Fine, A. (1996) How to Write Really Badly. London:Egmont.
Answers to ExercisesChapter 1Exercise 1.1The pattern of widening and narrowing continues right throughuntil completion of the research, with unsuitable or irrelevantproblems, theories, concepts, etc. rejected at each stage.Exercise 1.21. No. Plenty of information may be collected on this subject, butthere is no indication of any planned analysis. The sentenceneeds to continue: ‘in order to …’.2. No. There are two main defects. What will be done with thecomparison? No indication is given of why the comparison ismade and what is to be learned from it. The scope of thestudy is enormous. How can ten years of exam informationfrom all the schools in Europe be collected by one person inabout six months?3. Possibly. This is better, but still rather vague. Is this aworldwide study, or will only a sample of families beinvestigated?4. Possibly. This has probably already been investigated, andthe form of the problem is too simplistic. A short study wouldprobably reveal a set of relevant equations showing thecorrelation coefficient between the sets of data.5. No. A seemingly straightforward technical study. But what arethe ‘effects’ that will be studied? They could be on anything –on the environment, on glass production, on individualtemperature. More precision about which ‘effects’ are to bestudied is required.6. Yes. This has the vital ingredients of a researchable problem.You should know by now what the ingredients are!
7. Possibly. Although this has probably been investigatedseveral times over, the feasibility of the research problemdepends on the number of large country houses built inBritain in the 18th century, or on a selection of representativecase studies.8. No. The solution to the research problem could be a simpleyes or no. So what? Advantages to whom? Over what time-span?9. No. Nicely delineated, but where is the problem? This is pureinformation collection.10. Possibly. The difficulty is that this is two problems, involvingtwo areas of study.11. Possibly. Well formulated, but a large-scale undertaking. A lotof research into this subject has already been undertaken,and some aspect that has not yet been investigated or someconflicts in existing research would have to be identified.12. Possibly. This is too general as it stands. Is this a historicalstudy or is it investigating the current situation? Moredelineation is needed. Does it require sociological as well asfinancial data?13. Possibly. There is no indication of which programme is meant.You would need good Chinese contacts for this one.‘Economic implications’, a very general term, could do withsome delineation.14. Yes. Quite usefully formulated.Exercise 1.31. Though the research problem is not labelled as such (e.g. thisresearch project addresses the following problem etc.), it isclearly and succinctly stated in the fifth paragraph. The mainresearch problem is the lack of knowledge of theeffectiveness of YDUs, particularly in meeting therequirements for independent living of young disabled people.The ‘aims of the study’ in the first paragraph describe whatthe researcher aims to do in response to the problem.2. There is clear progression from paragraph two, culminating inthe problem in paragraph five and a statement of the focus ofthe study in paragraph six. This is achieved in the followingway. The general structure of the argument is based on a
historical account of the development of socially providedcare for people with disabilities, with particular emphasis onthe concepts of segregation and independence. The accountshows progression from provision that segregated disabledpeople from society to forms of accommodation and care thatwere intended to allow more integration of residents,culminating in the YDUs. The point is added that, despite ashortage of research, existing evidence shows that therequirements of younger disabled people are not adequatelymet.3. The sub-problems are implied rather than stated. It ispossible to detect from the text that the researcher hasidentified two unknowns that require examination in order toinvestigate the main research problem:The influences on different YDU built forms. This is rathertoo general a statement to mean very much. Whichinfluences need to be studied in order to investigate themain problem – economic, political, design, managerial,etc.? This vagueness is possibly a result of not definingthis aspect as a specific sub-problem.The way that independent living can be measured (hencethe need to establish indicators). This is important to themain problem. Independent living is an abstract concept,and it is not initially obvious how it can be measured.4. There are many limitations that narrow down the scope of theresearch, including the following:1. Country: implied, but rather surprisingly not stated in thetext, limitation of the study to conditions in the UK orprobably England.2. Time: current, i.e. not a historical study.3. People: young physically disabled in YDUs.4. Place: YDUs, purpose-designed to wheelchairparameters. Though there are over 320 of these, thestudy is limited to survey different types or built forms ofYDU group-living schemes. This implies that only asample of the 320 will be surveyed.5. Subjects: independent living, aspirations and needs ofyoung people with severe disabilities.6. Scope of surveys: detailed multi-method surveys arelisted including appraisal of plans, measurement of
buildings, observation of buildings in use, structuredinterviews with residents. Providing a list like this impliesthat these are the principal survey methods to be used,but does not preclude other types of survey activity, e.g.study of the history of the building, interviews withofficials and designers.7. Criteria: to measure the effectiveness in meeting theindependent-living needs and aspirations of youngpeople with severe disabilities. ‘Effectiveness’ is a verygeneral term, but it is stated in paragraph seven that thestudy will be limited to the indicators of independentliving devised early in the research.5. By investigating different YDU built forms, a comparative typeof research will be carried out. It is pretty obvious that it willbe a qualitative rather than a quantitative study. The researchwill also take place in a natural setting – investigating whatthe situation is as it exists. There is a specific intentionexpressed in the last paragraph that the research shouldincrease knowledge of the subject.6. In the last paragraph, the text stresses the originality of thecontribution to research, and the practical value of therecommendations. The importance of the study lies in how itwill help to improve the design of independent-living schemesto the benefit of the young disabled occupants.7. As you see, no one’s particular theory or study is referred toin the text. The main point of the proposal is to stress thepoint that there is a lack of research in the subject. Thispaucity is the reason for proposing the study. In the originalversion of this proposal, each major statement was backedup by a numbered reference to the relevant literature, listed atthe end.Exercise 1.41. Obviously it is impossible to give a direct answer to thisquestion here; it is up to you to draw your own conclusions.However, have you considered the following?1. Is the process shown as a linear progression or are datacollection and analysis concurrent? Which will beappropriate in your study? Some qualitative approaches
require a more evolutionary strategy with short feedbackloops that connect data collection and analysis.2. Do the proposals intend to test a hypothesis or look for adeeper understanding of a situation? The plan of actionand type of output are greatly influenced by the aims ofthe research.2. Only list those that might be relevant to your research plan,not those that are indicated in the various examples. Use thislist for the tasks in the final section of this chapter.Chapter 2Exercise 2.11. You only have to observe for one sunny Saturday to falsifythis one, so it can be used as a hypothesis.2. This statement cannot be falsified. A square can only becalled thus if it has got four sides of equal length. This is anaxiomatic statement.3. It might be difficult to falsify, but in theory it should bepossible, e.g. if you are in orbit around the earth, i.e. at a verygreat height, a thrown object might ‘fall’ away from the earth!This is an obvious but still valid hypothesis.4. This is a hypothesis that is capable of being readily falsified,and subsequently refined by a few experiments.5. This statement excludes the possibility of being falsifiedbecause it covers all possible states, and is therefore not ahypothesis.6. Even if the deflection, or lack of it, cannot be seen, it can beaccurately measured. This is a hypothesis.7. This statement would be difficult to falsify. Even if you onlyknew about business enterprises which made losses, it wouldnot rule out the possibility that profits could be made. It reallybegs the question of the meaning of the concept ‘businessenterprise’, which is, after all, a system aimed at makingprofits, amongst other things. For these reasons, thisstatement should not be called a hypothesis.8. An axiomatic statement again, and not a hypothesis.
9. This is more like it. I can think of many round objects whichare not circles. It is falsifiable, and therefore a hypothesis.10. Again, all possibilities are covered by this statementregarding objects and motion. It cannot be falsified.Exercise 2.21. This is rather a difficult question to answer, in view of thenumber of lengthy books that each of these writers haswritten about the nature of scientific enquiry! However, fromthe very short summary given in the section, you should beable to extract some of the main views that were described.How do your answers compare with these?Popper maintained that scientific progress was a rationalprocess. Although scientific theories were not provable,they must be falsifiable, and the only criterion to judge atheory was by comparison with others, the better theorytaking precedence over others.Kuhn disagreed that the progress of science wassmooth, and maintained that periodic and radicalreassessments were made of scientific understandingwhen persistent findings incompatible with acceptedknowledge resulted in a new paradigm being formulated,radically different from the old.Feyerabend challenged the position of scientific enquiryas the only reliable method of advancing knowledge byproducing useful results. If judged on results, many otherdisciplines which were regarded as ‘unscientific’ alsosuccessfully produced useful results, despite them notbeing able to be explained by scientific theories.2. Probably Popper and Feyerabend might have opposing viewsabout the value of interpretivism as a valid approach togaining and analysing knowledge. Popper would questionwhether the theories put forward as a result of theinterpretivist approach were falsifiable; or, if so, whether thetests were repeatable. After all, the subjective element will bedifferent in every case, leading inevitably to different results.Feyerabend, following on from his refutation of scientificmethod as the only valid ideology, would probably welcomeinterpretivism as a valuable additional approach in order to
gain knowledge, especially if it was shown to produce usefulresults.3. The central issue of this research project is to explore therelationship between drawings and product design. It ispossible to perceive that both the objectivist and interpretivistapproaches could lead to useful insights into aspects of thisrelationship.1. This is an abstract and theoretical relationship whichcould be investigated without reference to individualsubjective attitudes and social reality.2. The mention of ‘how coded information is perceived’indicates that a subjective element is certain to bepresent. How and what percepts are formed byinterpreting coded information on drawings must be, atleast partly, influenced by individual background andexperience. This will be a qualitative appraisal usingdescriptive methods of analysis.3. This aspect of the study does not mention people, butconcentrates on the relationships between two concepts,i.e. the forms of drawings and design as a method ofcommunication. A positivist approach seems to be themost appropriate in this case, though the concepts willbe difficult to measure.4. Here, people will be asked about drawings and in whatway they influence design. There is obviously room herefor an interpretivist approach to the research, though thewording ‘empirically assess’ makes one suspect that theresearcher intends to make a rational assessment basedon observation from a neutral standpoint. Isn’t this justwhat Weber was advocating? First to critically assess thevalue-laden situation, and then to see what actual effectsthis has on outcomes.It is probably difficult to combine the two approaches in one study,unless the intention is to provide two completely differentperspectives of the research problem, or the research problemdivides itself into very different sub-problems each of whichdemands a different approach. In this project, the two aspects ofdrawing, i.e. communication and influence on design, might beconsidered sufficiently diverse.
Chapter 3Exercise 3.11. Abstract.2. Concrete – specific to time.3. Abstract.4. Concrete – specific to person.5. Abstract, if the two words are taken to mean a completeconcept; or, if the words are separated, the concept could beargued to be concrete in that the system is specific to type,i.e. social type.6. The same can be said for this concept as for 5, except thatthe capital letters indicate that this is a complete term, andnot a description of an association.7. Abstract.8. Concrete – specific to time of the day and place.Exercise 3.21. Nominal.2. Nominal.3. Ratio.4. Interval. Note that although these are also numbers, 0 has nomeaning in this context, i.e. no one has an IQ of 0 – theywould probably be dead if they had!5. Ordinal, though they are often thought of as interval whenregarded as being equally spaced out along the line ofachievement.6. Ordinal.7. Ratio (0.0 bar would be a vacuum). The measurement takesits values from 1.0 bar being approximately atmosphericpressure at sea level.8. Ordinal.9. Ordinal.10. Ratio.11. Interval (0 °C has no particular significance in this context).12. Ordinal.13. Nominal.
14. Cannot all be quantified in entirety. One could quantify allnominally by colour or by shape, but not both together. Onecould quantify some (while rejecting others) by, say, coloursof cube or shapes in red.Exercise 3.31. Argument. Premise: increased taxation on petrol. Conclusion:less traffic on the roads. Premise indicator: ‘since’.2. Argument. Premise: diminished social responsibility ofcriminal through deprived upbringing. Conclusion: criminal isnot to blame. No logical indicators present, but premiseindicator ‘because’ is implied.3. Statement – relational.4. This could be seen as an associational statement, withpositive correlation inferred. It could also be read as anargument if we take the premise indicator ‘look at’ to mean‘because of’. Premise: unequalled variety of drama etc. inLondon. Conclusion: my thinking that London is the bestplace in the world. The question invites you to seek a premisefor the conclusion.5. Argument. Premise: insufficient pay rise offer. Conclusion:train drivers going on strike. Premise indicator: ‘because’.6. Statement – obviously associational.7. Argument. Premise: strike threat. Conclusion: insufficient payrise offer. Premise indicator: ‘as’ (indicated by the fact that).8. Statement – associational with negative correlation.9. Argument. Premise: Members of Parliament have misled ustoo often. Conclusion: I distrust Parliament. Premise indicator:‘my reason being’.10. Argument. Premise: continued ban on fishing lobsters.Conclusion: lobsters will soon be unobtainable. Conclusionindicator: ‘so it follows that’.11. Existence statements. There is no link indicated between thetwo statements.Exercise 3.4
1. Giving birth to a healthy baby, whatever the circumstances,outweighs all possible disadvantages.2. The price of everything would go up to cover the cost ofreducing traffic.3. The facts listed should persuade authorities to provide morefacilities for cyclists.4. The greatest medical danger lies in reducing the infantmortality rate amongst poor people.5. It is difficult to disagree with the opinions of environmentalexperts that nuclear power is not the answer to our energyproblems.Exercise 3.51. Inductive.2. Deductive.3. Inductive.4. Inductive.5. Deductive.6. Deductive.7. Deductive.8. Deductive.9. Inductive.10. Inductive.Exercise 3.61. Inconsistent. Expertise in anything takes more than twominutes!2. Inconsistent. My knowledge was obviously insufficient.3. Inconsistent. The rapidly increasing national debt could be astrong sign of a financial crisis.4. Consistent.5. Consistent.6. Consistent. I probably want many other practices to stop too.Exercise 3.7
1. Formal. Denying the antecedent. This is a result of the writernot really caring if his/her brain is going backwards orforwards. He/she does not recognize that the same result canbe produced by different causes.2. Informal (linguistic). Amphiboly. This is the fallacy of carelessgrammatical construction which causes an ambiguity ofmeaning.3. Informal (relevance presumption). Analogical fallacy.Analogies are often a useful way of describing unfamiliarconcepts by talking about them in terms of which theaudience already has experience. The fallacy comes inassuming that further similarities, other than those alreadyidentified, inevitably occur. None of the false analogieslikening the state to a ship ever seem to say much about itscabins, air-conditioning system or boiler capacity!4. Another example of informal (relevance presumption). Dictosimpliciter. This is the fallacy of sweeping generalization. Itignores the fact that applying a broad general rule to anindividual case overlooks the special features which mightmake the case exceptional.5. Informal (relevance intrusion). Argumentum ad ignorantium.Just because we lack knowledge about something, it is afallacy to infer that the opposite is the case, e.g. in this casethat it does not exist.6. Informal (relevance omission). One-sided assessment. Thisfallacy is committed when only one side of the situation isconsidered. A balanced view is usually required in order tomake an informed decision.7. Formal. Undistributed middle. A test with the Venn diagramwill show you the shortcomings of this argument. The middleterm ‘really sensitive people’ in this example does not coverthe whole of its class at least once, so there is a fallacy in theargument (i.e. in this example the middle term never oncerefers to all really sensitive people).8. Informal (relevance intrusion). Blinding with science. Pseudo-scientific jargon can be used to try to impress people withyour superior knowledge. ‘The cat sat on the mat’ is a simpleway of translating the ‘gobbledegook’.9. Formal. Affirming the consequent. In an ‘if … then’construction, the ‘if’ part is the antecedent, and the ‘then’ part
is the consequent. It is all right to affirm the antecedent inorder to prove the consequent, but not vice versa. Affirmingthe consequent is fallacious because an event can beproduced by different causes.10. Informal (omission). Concealed quantification. Whenstatements are made about a class, sometimes they areabout all of the members of it, sometimes about some ofthem. This fallacy occurs when ambiguity of expressionpermits a misunderstanding of the quantity which is spoken of(i.e. in this case is it all or some careless people who are badat golf?).Exercise 3.81. Correct.2. Correct, unless there are finer divisions, e.g. slightly hard,slightly soft etc.3. More than one basis of division. How many crusts has a mudpie?4. Correct, unless you feel you should add nursery schoolbuildings.5. Correct, though there might be some considerable dispute asto the precision of the categories.6. Overlapping categories; cold days could be also sunny etc.7. Correct, unless you can think of any other basic pen types. Ifso, then the classification is with gaps.8. Gaps left: what about wind, waves, solar etc.?9. Correct, if you are only talking about planets circling our sun.10. More than one basis of division. The two types of divisionalso overlap.Chapter 4Exercise 4.1You will progressively develop your skills in tracking downinformation. The important point is to keep yourself well informed
and not to hesitate to explore different sources and techniques ofinformation retrieval.Exercise 4.2Obviously you can devote only a short time to exploring theresources of the Internet at this stage. You will develop skills themore you use it. Take care that you do not get diverted from thesubjects that you are searching information on. Like looking up aword in an encyclopaedia, it is very easy to get stuck reading allthe other interesting topics on the same page!Exercise 4.31. When you re-read your notes, now that you have read thearticle more thoroughly, does your initial summary contain allthe essential points in the argument?2. These notes will probably be rather fragmented, but willcontain useful definitions and descriptions. The list form isoften a useful way of recording information about concepts.3. Make sure that your notes are not longer than the originaltext! Note that the first sentence of a paragraph usuallyintroduces the subject of the paragraph. The followingsentences elaborate or explain further. The last sentencemight make a conclusion or a bridge to the next paragraph.4. These last paragraphs often contain the real essence of thearticle or paper. This is why it is often good to begin reading along thesis or paper by starting at the end! It is important notjust to make notes summarizing the information you collect,but also to consider, as you are reading, the significance ithas for your present thinking about your subject.5. Try a Barbara Cartland novel (she wrote one every twoweeks!)Exercise 4.4The outcomes to this exercise are for you to judge! Aim to devisethe simplest system of note-recording, sorting and retrievalsystem possible adjusted to your needs and inclinations. The
move from a paper to a computerized system will save you lots oftime in the future as the notes are already in electronic format, soyou can easily select, paste and revise them when you make thefirst draft of your report. You will need, ideally, to have your ownlaptop or equivalent, so that you can make the notes whereveryou want to, just as you can with writing on paper.Chapter 5Exercise 5.1If you want to have a further check on the strength of yourargument, you could test your result out on a colleague to see ifhe/she is equally convinced with the result.Chapter 6Exercise 6.11. Tampering with medical records to support his claim to haveperformed a pioneering operation. He also pretended to haveconducted extensive research trials into ovarian diseases.2. Other obstetricians and gynaecologists might have tried toperform the same reputed operation with possible dangerousresults, and have based treatments on the evidence putforward by the reputed research trials.3. Several safeguards are there, but they are not, on their own,defence against fraud. Claims must be substantiated withaccessible and reliable evidence (though as seen in thearticle, this evidence can sometimes be maliciously altered),and the research methods used must be open to scrutiny andrepeatable (after several deaths, the success of the operationmethods will be questioned!). While many types ofexperiment can be repeated without harmful effects, even ifthey are not successful, in medicine this is definitely not thecase.
4. It is difficult to imagine how a ‘fraud squad’ could operate in adirect and active way without the inside help of members ofthe research community. As it appears to be a small (thoughperhaps important) problem with little obvious effects onsociety, general policing or detective work like that used infinancial fraud is unlikely to be appropriate. However, asystem which provides anonymous and discreet consultationwith the ‘squad’ by researchers who suspect fraud within theirdepartment might alert the authorities to criminal activities.5. The accusation is that Jung falsified documents which heused as evidence to support his argument. This is similar tothe type of falsifying of which obstetrician Malcolm Pearcewas convicted. The greatest difference is that the accusationagainst Jung cannot be substantiated because the evidencelies in an archive in the Library of Congress to which accesshas been refused by the Jung family. If the accusation is true,it might be difficult to assess whether anybody has beenphysically or mentally harmed by the fraud (in Mr Pearce’scase, this would have been much simpler). However, if true,the credibility of Jung’s other writings will be put in question,and the evidence he puts forward to support all his theorieswill need to be closely investigated.6. It must be very tempting to ‘bend’ the evidence if it does notquite fit your radical theory, which you think will make yourespected as a leader in your field of research. Ambition is apowerful emotion. There are several reasons why scientistsdo not lie: they are, like most people, honest; professionaldisaster awaits those who are found out; and scientificadvance is not achieved only by ‘proving’ new theories,because the falsifying of an existing hypothesis or theory alsocounts as a scientific success.Exercise 6.2The issue of what labels to use to categorize people is somethingof a minefield! PC (politically correct) terms are subject to constantchange and even some gernerally accepted ones are contentiousin some circles. Individual groups have particular ways to describethemselves, so this should be noted where possible. So, I take therisk of listing here the words I found inappropriate, with
suggestions for suitable terminology to replace them. For acomprehensive review of officially recommended categorizationand labelling terms see:www.ons.gov.uk/methodology/classificationsandstandards/measuringequality/ethnicgroupnationalidentityandreligionor just type: ONS-recommended categories into Google.old folk, old dears. These are condescending terms. Use ‘olderpeople’ or possibly ‘senior citizens’.the normal ones. This implies that disabled people are notnormal people! Use ‘able-bodied people’ or ‘people withoutdisabilities’.the disabled. Not acceptable any more as it is depersonalizing.Use ‘disabled people’ or ‘people with disabilities’. Rememberthat this does not necessarily mean old people.the blind. A vague term without the personal element.Alternatives are ‘blind’ or ‘partially sighted’ person, ‘someonewith visual impairment’, ‘people with little or no sight’.the deaf. A vague term again. Add the personal element, and ifpossible whether partially deaf or profoundly deaf. Somepeople use ‘the Deaf’ as a term for a linguistic minority ratherthan a disabled group.the wheelchair-bound. Substitute with ‘wheelchair users’ or‘people with impaired mobility’.mental handicap. ‘Handicap’ is seen as a discriminatory term.Use ‘learning difficulties’ or ‘learning disabled’.businessmen. Is it only the men who are mentioned here?Probably not. Use ‘business person’, ‘manager’ or ‘executive’.headmasters. The same gender question applies here. Use‘head teacher’.putting their wallets into their trouser pockets. This assumesthat all the previous professions are male occupations.Rephrase to include female terms or non-gender terms suchas ‘purse’ and ‘inside pocket’.lower class. Seen by many as derogatory and demeaning. Use‘working class’, though this is rather old-fashioned, or ‘peoplewith lower earnings’.
charwomen, deliverymen, salesgirls and waiters, policemen,laymen. Again, these are all gender specific when this isprobably not intended. Use gender-neutral terms such as‘cleaners’, ‘couriers’, ‘sales assistants’, ‘waiting staff’, ‘policeofficers’, ‘lay persons’.British people. As used here, this implies that these are a whiteethnic group – patently not true. Use instead ‘white people’.the blacks. Depersonalized. Use ‘black people’ or ‘persons ofcolour’. According to the ONS, ‘Skin colour is an option forconsidering ethnic group. However, it’s not an adequatecriterion in its own right, and for some, its use is seen asunacceptable’ (ONS site address as above).other immigrants. This implies that ‘the blacks’ are necessarilyimmigrants. This is of course not the case in an establishedmulti-ethnic society.druggies, homosexuals and tramps. If these are the termsused by the young people, this should be made explicit. ‘Drugdealers’ or ‘drug users’ is more precise, ‘gay persons’overcomes the use of medical terminology that is usuallyassociated with males, and perhaps ‘vagrants’ is a lessderogatory term.geriatrics. A medical term for a branch of medicine. Use ‘olderpeople’.typical young person … himself … his. Ignores the femalegender of this category of people. Use ‘him/herself’, ‘his/her’,or the now quite grammatically acceptable ‘their’ even if it is inthe singular.Exercise 6.3Here are my brief thoughts. I am sure that you will have a range ofreactions that are worth discussing with your colleagues. It isunlikely, though, that your research will take such a radicalexperimental direction.1. In defence of Milgram’s methods is the fact that no physicalharm was incurred by any of the subjects. Despite thethorough debriefing after the experiment, I question whetherthe anguish suffered by the participants caused by theirsituation was trivial and whether the psychological effects of
the experience might not have been harmful in the longer term.I would definitely have felt very cheated and angered had Ibeen a participant in that research.2. For:1. This is the only way to discover facts about this importantissue.2. Full precautions are taken to explain to the participantsthat they have been deceived and that no pain at all wasadministered.3. The participants are allowed to stop the experiment at anystage.Against:1. An experiment based on deception is basically unethical.Other methods must be devised to gain the required data.2. The participants are put into a situation that inevitably causesthem stress beyond what they should be asked to bear.3. The longer-term effects on the participants cannot be gauged.Trauma might result in post-traumatic stress. The experimentdoes not include follow-up monitoring.3. It is difficult to see how participants could be subjected to asimilar situation without the need for deception, and still haveto make such onerous choices of action. The whole point of theexercise was to measure the extent to which ordinary peoplego against their normal principles when faced with authority.Perhaps nowadays, a computer simulation might be used tocreate a scenario that involves this type of conflict of interest.4. Assuming that I behaved like most of the other participants,my main shock would have been my willingness to inflict painagainst my normal natural inclinations. I would have been veryrelieved that I had not, after all, caused harm to anyone, but Iwould have felt angry that I had been exploited in this way.Chapter 7Exercise 7.1
1. Researchers use sampling procedures in order to be able tostudy a manageable number of cases. Sampling procedureshelp to ensure that the cases studied are representative ofthe whole population.2. The single case is unlikely to be representative of the wholegroup, unless the group is very homogeneous in the aspectstudied. There are situations when studying a single case isnot only useful, but unavoidable. For example, when only onefossilized skeleton of a type of dinosaur has been discovered,many generalizations have been made about other possibledinosaurs of the same type, even in the absence of otherskeletons.3. 1. Random and non-random sampling. Random samplingrelies on pure chance to select cases, which eliminatesany possible influence of the researcher. Non-randomsampling consists of selecting cases on the basis ofconvenience, even though there might be an attempt toachieve an element of representativity.2. Random sampling must be used whenever possible ifyou wish to make generalizations from your collecteddata. When the numbers in the population are very small,when time is very short or it is impossible to gain accessto the full range in the population, non-random samplingmay be used, with inherent limitations on the validity ofgeneralizations made from the sample.4. You must examine the nature of the population: is ithomogeneous, or does it consist of strata? If the latter, whatis the nature of the strata? Alternatively, the population mayconsist of clusters. Appropriate sampling techniques must bechosen in each case.5. The required size of the sample depends on the nature of thepopulation and the detail of the survey. Great variability withinthe population and detailed information requirement demandlarge samples. If great accuracy is needed, then the samplealso needs to be large. If the population is very homogeneousand the survey limited in scope, then a small sample willsuffice. Minimum sample sizes are normally stipulated fordifferent statistical tests. Take account also of the practicallimits imposed by the project.
Exercise 7.21. 1. Too vague. Do you want to know which street, town,country or continent the person has come from?2. What reply can people give if they have five or morechildren? You could add the category ‘more than 4’ tocover this eventuality.3. The gradation in the scale is too fine. What is thedifference between 54% and 56%? It is meaningless. Afive-point scale is common: like very much, like, neitherlike nor dislike, dislike, dislike very much. Another maybe added: don’t know!4. This assumes that you have drunk a bottle ofchampagne within the last year.5. Too complicated! Unless you are aiming thisquestionnaire at a very specialist and knowledgeablesample, the responses will be pure guesswork.2. The main difference between all the four different kinds ofpersonal data collection is the amount of control that theresearcher has in determining the questions, and hence thetype of answers which are expected. The most directed kindis the structured interview, where the respondent is stronglyguided to limit him/herself to answering the series ofquestions posed by the researcher. At the other extreme,accounts are elicited from respondents with prompts, ratherthan questions, in order to encourage them to speak freelywith as little direction from the researcher as possible. Semi-structured and open interviews fall between these twoextremes.3. No. When talking about a ‘population’, we are not only talkingabout people. A sample can be taken from a population ofobjects, events, situations and phenomena.Exercise 7.31. 1. Assuming that you want to keep all of your students inyour lecture, a true experiment will not be possible. Aone-shot case study (testing the students after thelecture) is possible, but takes no account of what they
might already know. The one-group pre-test/post-testdesign would be better (testing the students before andafter the lecture and comparing their increase in artistrecognition).2. Much depends on whether you have several bricks of thesame type which you can choose from randomly. If so,you could carry out a true experiment using a controlgroup (pre-test/post-test control group). I suppose youwould have to assume that any unbaked bricks thatdissolved when you soaked them would also not standup to frost!3. Difficult, as there are so many uncontrollable variableswhich determine how long people sleep. Try setting up apseudo-experiment and give a sleeping drug to theparticipants sleeping in your bed!2. 1. Mechanical (physical), mathematical, diagrammatic.2. Diagrammatic.3. Physical, diagrammatic (including computer generated).4. Diagrammatic.5. Diagrammatic, possibly mathematical.3. 1. Input data: egg size. Output data: time. No feedback.2. Input data: room geometry, furniture number and typewith sizes, minimum spaces required around furniture inuse. Output data: room arrangements. Feedback:possible changes to the space needs of using furniturethrough overlaps.3. Many inputs: speed of buses, position of bus stops, stateof traffic, etc. Output data: behaviour of buses relative toeach other. Feedback: many possibilities to explorechanges in input data suggested by results, e.g. changebus speed, alter bus stops.4. Input: glass type and strength, type and shape offraming, method of fixing glass, type of impact. Output:resistance to impact data. Feedback: different and moreefficient combinations of the input values might besuggested by the output data.Exercise 7.4
1. 1. The large number of scripts suggests that somequantitative analysis is possible, assuming that there issome standardization in the questions, and hence in thereplies to them. Qualitative analysis would have to beused to assess the wider range of opinions given.2. This appears to be a purely qualitative study.3. Quantitative at first sight. It does depend, though, onwhat data are collected.4. Both quantitative and qualitative analysis are possiblehere.5. Again, this looks like a quantitative study, though it ispossible that some qualitative analysis could be made byinference.2. Parametric statistics are used to study data that conform to acommon parameter, normally the Gaussian curve. Non-parametric statistics do not conform to a common parameter.3. It is difficult (impossible?) to prove anything with statistics!You could, however, support an argument with statisticalanalysis that suggests correlations between variables whichyou claim demonstrate a connection between smoking andcancer.4. I agree that you do have to use your creative and imaginativefaculties when doing qualitative analysis, but your argumentsmust be backed up with logic and a coherent set of evidence.The use of human abilities to simplify, order and theorizecomplex, open-ended social phenomena is essential forgaining an understanding of our everyday existence.However, this is not to say that science is devoid of creativity.Some of the greatest discoveries have been made on anintuitive basis, and only later supported by empirical data.Unlike quantitative analysis, qualitative analysis must relyless on systematic experiment and testing than oncontinuous comparison of data. Concepts and theories aremore likely to be developed during, and as a result of, thiscontinuous comparison, rather than being formulated atthe outset of the research. Quantitative analysis has thepotential to provide precise answers, which qualitativeanalysis lacks. However, qualitative analysis can deal withsubjective data that cannot be precisely measured.
5. 1. A network in the form of an organization chart wouldshow the hierarchical structure. A role-ordered displaywould also add information on the type of relationshipsbetween the roles.2. A simple matrix could be used here to plot types ofschool vertically and social classes horizontally, givingfields to record the data on attitudes with eachcombination.3. A network in the form of an activity record would clearlyshow the sequence of steps required to complete thetask successfully.4. An effects matrix would help to show how this changeaffected the work in the factory, and even, if necessary,the levels of pay.Chapter 8Exercise 8.1Have you got key words referring to the following aspects? If yes,are they really needed? If no, have you missed somethingimportant?specific place, e.g. specific country or citygeneral place or situation, e.g. rural, urban, building/socialcontext (school, home, factory), the natural environment, localgovernmentcontext, e.g. social, political, economictime, e.g. dates, past, contemporarymeasures, e.g. effectiveness, impact, performance, values,relationships between, appraisal, influences, implicationsconcrete concepts, e.g. steel, interior planting, localregulations, houses, professional practices, environmentallegislation, crime figurestheoretical concepts, e.g. wealth, leisure, disability, heritage,conservation, ecology, education, access, housing, theenvironment
Exercise 8.2This list, together with your answers to the questions, is goodmaterial for discussion with your tutor. Why not make anappointment to see him/her? Bring along the list of key words youhave selected in Exercise 8.1 together with the title you havecomposed.Exercise 8.3The result of this exercise will be the first draft of your proposal(without the description of methods, which you can write only afteryou know exactly what the research problem is!). Show this toyour tutor, and discuss and revise it several times before you goon to the methods section. The process of regular consultationwith various members of tutorial staff is essential at this stage, sothat your thinking about your intended research develops rapidlyand along the right lines.Exercise 8.4This is a preparation for the plan of work in your proposal thatdescribes the methods you will use. You will need to discuss thiswith your tutor or supervisor. The main aspects to check are theappropriateness of the methods in relation to what you want tofind out, and the practical issues involved (access to sources ofinformation, necessary equipment, time requirements, costs,travel, etc.). It is best to set out this section broadly following thesequence in which the work will be carried out. You must showthat you have a good understanding of what is involved in carryingout the methods you propose and the reasons why you are goingto use them. In an extended study, e.g. a PhD thesis or fundedresearch project, it is likely that part of your plan will be to find outmore about appropriate methods and to refine or adapt them forapplication to your particular subject.The last part of this exercise will help you to focus on the essentialaspects of your project and the reasons why you are undertakingsuch a major task. Do not hesitate to discuss these with your
colleagues and supervisors. It is also particularly useful to get thereactions of people who are not familiar with your field of study inorder to test the clarity of the contents and the language.Exercise 8.5You do not really need an answer from me here, as you are bynow completely competent to do this kind of analysis. The point ofthe exercise is to make you look afresh at your own proposal, as ifit were another example of the kind shown in Boxes 8.3 and 8.4,so that you can analyse it from a purely technical and structuralstandpoint. Are all the ingredients of a successful proposal there,and are they presented clearly in order to build up a goodargument for doing the research? Can it convince the researchcommittee that the project is practical and can be completedwithin your resources of time, skills, equipment and finance?
Key Words GlossaryAbstractnessA characteristic of research findings that makes themindependent of specific time and place. Such research findingsare useful as they can be applied to other situations.Accidental samplingAlso called convenience sampling. A non-random samplingtechnique that involves selecting what is immediately available,e.g. studying the building you happen to be in, examining thework practices of your firm.AlgorithmA process or set of rules used for calculation or problem-solving, especially using a computer. It can be expressedgraphically or, more often, as a mathematical formula. Anexample is a formula that summarizes the interior conditionsthat lead to the feeling of climatic comfort, with factors such asair temperature, humidity, air movement, amount of clothing,etc.AnalogyA comparison of two different types of thing in order to detectthe similarities between them. It may be possible to infer thatthey possess further undetected similarities.ArgumentA type of discourse that not only makes assertions but alsoasserts that some of these assertions are reasons for others.Argument is often based on the rules of logic in order toprovide a solid structure.Associational statementsMake an assertion that two concepts are associated in someway, positively or negatively, or are not associated at all. The
word ‘correlation’ is often employed to refer to the degree ofassociation. Association does not imply causation.AuthenticationChecking on historical data to verify whether they areauthentic. Typical techniques used are textual analysis, carbondating, paper analysis, cross-referencing, etc.Axiomatic theoryA theory that comprises an initial set of statements (axioms) orself-evident truths, each independent of the others, and fromwhich it is possible to logically derive all the other statements(propositions) of the theory. A good example of one of these isPythagorean geometry.Bibliographic databaseAn electronic list of bibliographic information. Databases maybe on CD-ROM, or online on the Internet. Many of these areavailable through university and other libraries and can besearched using key words, subjects, authors, titles, etc.BibliographyA list of key information about publications. They can becompiled on particular subjects or in relation to a particularpiece of academic work. There are standard systems forcompiling bibliographies, e.g. Harvard. Libraries usuallycompile their own bibliographies to guide students to literaturein their specialist subject.CategorizationInvolves forming a typology of objects, events or concepts.This can be useful in explaining what ‘things’ belong togetherand how.Causal statementsThese make an assertion that one concept or variable causesanother – a ‘cause and effect’ relationship. This can bedeterministic, meaning that under certain conditions an eventwill inevitably follow, or if the outcome is not so certain,
probabilistic, meaning that an event has a certain chance(which may be quantifiable) of following.CitationA reference to a source of information or quotation given in atext. This is usually in abbreviated form to enable the fulldetails to be found in the list of references.ClassA set of persons or things grouped together or graded ordifferentiated from others. Classes can be formed by collectionor division. Classes can be divided into subclasses to form ahierarchy.Cluster samplingSelection of cases in a population that share one or somecharacteristics, but are otherwise as heterogeneous aspossible, e.g. travellers using a railway station. Also known asarea sampling when random segments are chosen from alarge area of population distribution.CodingThe application of labels or tags to allocate units of meaning tocollected data. This is an important aspect of formingtypologies and facilitates the organization of copious data inthe form of notes, observations, transcripts, documents, etc. Ithelps to prevent ‘data overload’ resulting from mountains ofunprocessed data in the form of ambiguous words. Coding ofqualitative data can form a part of theory building. Codes canalso be allocated to responses to fixed choice questionnaires.Coefficient of correlationThe measure of a statistical correlation between two or morevariables. There are many types of these, the Pearson’s rbeing the most common.ConceptGeneral expression of a particular phenomenon, or words thatrepresent an object or an idea. Concepts can be concrete, e.g.
dog, cat, house; or abstract, independent of time or place, e.g.anger, marginality, politics. We use concepts to communicateour experience of the world around us.ConsistencyA quality of argument concerned with the compatibility ofbeliefs, i.e. a set of beliefs that can be shown to be compatiblewith each other is said to be consistent.ControlHaving the ability to determine the influences on variables in aphenomenon, e.g. in an experiment. The crucial issue incontrol is to understand how certain variables affect oneanother, and then be able to change the variables in such away as to produce predictable results. Not all phenomena canbe controlled, as many are too complex or not sufficientlyunderstood.Critical rationalismAn approach usually associated with Popper. It maintains thatrival theories can be judged against specific, unchanging,universal criteria, which are divorced from or set beyond theinfluences of time or society.Critical realismA non-empirical (i.e. realist) epistemology that maintains theimportance of identifying the structures of social systems, evenif they are not amenable to the senses. This will enable thestructures to be changed to ameliorate social ills.DeductionThe inferring of particular instances from a general law, i.e.‘theory then research’ approach.Descriptive statisticsA method of quantifying the characteristics of parametricnumerical data, e.g. where the centre is, how broadly they arespread, the point of central tendency, the mode, median and
means. They are often explained in relation to a Gaussian(bell) curve.DiscourseCommunication in the form of words as speech or writing oreven attitude and gesture. Discourse analysis studies the waypeople communicate with each other through language in asocial setting, where language is not seen as a neutral mediumfor transmission of information, but is loaded with meaningsdisplaying different versions of reality.Empirical generalizationA generalization based on several empirical studies that reveala similar pattern of events. All concepts in an empiricalgeneralization must be directly measurable.Empirical relevanceThe measure of the correspondence between a particulartheory and what is taken to be objective empirical data.Empirical truthA truth derived from experience. In science, experiments are acommon way of establishing the truth by empirical methods.EpistemologyThe theory of knowledge, especially about its validation andthe methods used. Often used in connection with one’sepistemological standpoint – how one sees and makes senseof the world.EthicsThe rules of conduct. In this book, particularly about conductwith other people and organizations, aimed at causing no harmand providing, if possible, benefits.EvaluationMaking judgements about the quality of objects or events.Quality can be measured either in an absolute sense or on acomparative basis.
Existence statementsThese claim that instances of a concept exist in the real world,and provide a typology or a description.ExperienceActual observation or practical acquaintance with facts orevents that results in knowledge and understanding.ExplanationOne of the common objectives of research.External validityThe extent of the legitimate generalizability of the results of anexperiment.FalsificationThe process by which a hypothesis is rejected as a result oftrue observational statements that conflict with it.Formal fallaciesThese occur due to some error in the structure of the logicused, causing the chain of reasoning to be defective.GeneralityThe assumption that there can be valid relationships betweenthe particular cases investigated by the researcher and othersimilar cases in the world at large.HypothesisA theoretical statement that has not yet been tested againstdata collected in a concrete situation, but which it is possible totest by providing clear evidence for support or rejection.Hypothetico-deductive methodSynonymous with scientific method. Progress in scientificthought by the four-step method of: identification of a problem,formulation of a hypothesis, practical or theoretical testing ofthe hypothesis, rejection or adjustment of the hypothesis if it isfalsified.
InductionThe inference of a general law from particular instances. Ourexperiences lead us to make conclusions from which wegeneralize.Inferential statisticsStatistical analysis that goes beyond describing thecharacteristics of the data and the examination of correlationsof variables in order to produce predictions through inferencebased on the data analysed. Inferential statistics are also usedto test statistically based hypotheses.Informal fallaciesThese occur when the ambiguities of language admit error toan argument, something is left out that is needed to sustain anargument, irrelevant factors are permitted to weigh on theconclusions, or unwarranted presumptions alter the conclusion.Informative (language)Language used to communicate information.Informed consentConsent given by participants to take part in a research projectbased on having sufficient information about the purposes andnature of the research and the involvement required.Interim summaryA short report prepared about one-third of the way throughdata collection in qualitative research in order to review thequantity and quality of the data, your confidence in theirreliability, and the presence and nature of any gaps or puzzlesthat have been revealed, and to judge what still needs to becollected in the time available.Internal validityA measure of the level of sophistication of the design andextent of control in an experiment. The values of data gainedshould genuinely reflect the influences of the controlledvariables.
InterpretationAn integral part of the analysis of qualitative data that requiresverification and extrapolation in order to make out or bring outthe meaning.InterpretivismThe standpoint that recognizes the ‘embedded’ nature of theresearcher, and the unique personal theoretical stances uponwhich each person bases his/her actions. It rejects theassertion that human behaviour can be codified in laws byidentifying underlying regularities, and that society can bestudied from a detached, objective and impartial viewpoint bythe researcher. Attempts to find understanding in research aremediated by our own historical and cultural milieu.Interpretivist approachAn approach to research based on interpretivism. This requiresa trained researcher and human subjects, and uses aqualitative approach to data collection and analysis. Thetheoretical bases of this approach are called hermeneutics,phenomenolegy and symbolic interactionism.IntersubjectivityAgreement between people about the meaning of conceptsused in statements, attained by precise definition of theconcepts. Put another way: inter-observer agreement aboutthe veracity of an operational definition. Intersubjectivity is alsopromoted by the use of appropriate logical systems such asmathematics, statistics and symbolic logic.Interval level of measurementMeasurement scale that makes use of equal units, but withouta significant zero value, e.g. the Fahrenheit or Celsiustemperature scales.LawStatement that describes relationships that are so wellsupported by evidence, and engenders such strong confidence
in their reliability, that they are considered to express the ‘truth’in the cases to which they apply.Levels of abstractionThe degree of abstraction of a statement based on three levels– theoretical, operational and concrete, the last being the leastabstract.Levels of measurementThe four different types of quantification, especially whenapplied to operational definitions, namely nominal, ordinal,interval and ratio.Library catalogueBibliographic details of items in a library. The databases arenow usually accessed by computer as online public accesscatalogues (OPACs).Logical truthLogically true statements can be divided into three varieties:trivial (obvious), true by necessity (according to rules, e.g.mathematical) and true by definition (conforming tounambiguous definition).MatricesTwo-dimensional arrangements of rows and columns used tosummarize substantial amounts of information. They can beused to record variables such as time, levels of measurement,roles, clusters, outcomes, effects, etc. Latest developmentsallow the formulation of three-dimensional matrices.MemoShort analytical description based on the developing ideas ofthe researcher reacting to the data and development of codesand pattern codes. Compiling memos is a good way to explorelinks between data and to record and develop intuitions andideas.Mixed methods
Involves a purposeful mixing of qualitative and quanititativemethods in data collection, data analysis and interpretation ofthe evidence. An essential feature is the linking or integrationof the data at appropriate stages in the research process.Model(a) A term used to describe the overall framework that we useto look at reality, based on a philosophical stance (e.g.postmodernism, post-structuralism, positivism, empiricism).(b) A simplified physical or mathematical representation of anobject or a system used as a tool for analysis. It may be able tobe manipulated in order to obtain data about the effects of themanipulations.NetworksMaps or charts used to display data, made up of blocks(nodes) connected by links. They can be produced in a widevariety of formats, each with the capability of displayingdifferent types of data, e.g. flow charts, organization charts,causal networks, mind maps.Nominal level of measurementThe division of data into separate categories by naming orlabelling.Null hypothesisA statistically based hypothesis tested by using inferentialstatistics. A null hypothesis suggests no relationship betweentwo variables.Operational definitionA set of actions that an observer should perform in order todetect or measure a theoretical concept. Operationaldefinitions should be abstract, i.e. independent of time andspace.OrderThe condition that things are constituted in an organizedfashion that can be revealed through observation.
Ordinal level of measurementOrdering data on a scale by rank without reference to specificmeasurement, i.e. more or less than, bigger or smaller than.ParadigmThe overall effect of the acceptance of a particular generaltheoretical approach, and the influence it has on the scientist’sview of the world. According to Kuhn, normal scientific activityis carried out within the terms of the paradigm.ParsimonyEconomy of explanation of phenomena, especially informulating theories.ParticipantSomeone who takes part in a research project as a subject ofstudy. This term implies that the person takes an active role inthe research by performing actions or providing information.Pilot studyA pre-test of a questionnaire or other type of survey on a smallnumber of cases in order to test the procedures and quality ofresponses.PlagiarismThe taking and use of other people’s thoughts or writing asyour own. This is sometimes done by students who copy outchunks of text from publications or the Internet and include it intheir writing without any acknowledgement of its source.PopulationA collective term used to describe the total quantity of cases ofthe type which are the subject of the study. It can consist ofobjects, people and even events.Positivist approachAn epistemological stance that maintains that all phenomena,including social, can be analysed using scientific method.
Everything can be measured and, if only one knew enough, thecauses and effects of all phenomena could be uncovered.PostmodernismA term applied to a wide-ranging set of developments in criticaltheory, philosophy, architecture, art, literature and culture,which are generally characterized as either emerging from, inreaction to, or superseding, modernism. In sociology,postmodernism is described as being the result of economic,cultural and demographic changes – related terms in thiscontext include post-industrial society and late capitalism – andit is attributed to factors in the rise of the service economy, theimportance of the mass media and the rise of an increasinglyinterdependent world economy.Post-structuralismAny of various theories or methods of analysis, includingdeconstruction and some psychoanalytic theories that deny thevalidity of structuralism’s method of binary opposition andmaintain that meanings and intellectual categories are shiftingand unstable.PredictionOne of the common objectives of research.Primary sourcesSources from which the researcher can gain data by direct,detached observation or measurement of phenomena in thereal world, undisturbed by any intermediary interpreter. It is amatter of philosophical debate as to what extent thedetachment and undisturbed state are possible or evendesirable.Problem areaAn issue within a general body of knowledge or subject fromwhich a research project might be selected.Proportional stratified sampling
Used when cases in a population fall into distinctly differentcategories (strata) of a known proportion of that population.PropositionA theoretical statement that indicates a clear direction andscope of a research project.Qualitative analysisUses subjective judgement based on non-quantifiableinformation, such as opinions, beliefs, traditions etc. It tends touse an inductive rather than deductive approach. Instead ofworking from predetermined hypotheses, the researcherdevelops theories on the basis of coded qualitative data after ithas been collected analysed.Quantification (of concepts)Measurement techniques used in association with operationaldefinitions.Quantitative analysisThe process of collecting and evaluating measurable andverifiable data using statistical methods to describe andanalyse the data.Quota samplingAn attempt to balance the sample by selecting responses fromequal numbers of different respondents. This is an unregulatedform of sampling as there is no knowledge of whether therespondents are typical of their class.Ratio level of measurementMeasurement scale with equal units and containing a true zeroequal to nought – the total absence of the quantity beingmeasured.ReasoningA method of coming to conclusions by the use of logicalargument.
Reconciliatory approachThis approach, advocated by Roy Bhaskar, promotes areciprocal movement between the study of the individual,his/her beliefs and values, and the study of structures ofsociety, in order to check the results of the one against theother.Relational statementsThese impart information about a relationship between twoconcepts. They form the bedrock of scientific knowledge andexplain, predict and provide us with a sense of understandingof our surroundings.RelativismThe stance that implies that judgement is principally dependenton the values of the individuals or society and the perspectivesfrom which they make their judgement. No universal criteriacan be ‘rationally’ applied, and an understanding of decisionsmade by individuals or organizations can only be gainedthrough knowledge of the historical, psychological and socialbackgrounds of the individuals.ReliabilityIn relation to human perception and intellect, the power ofmemory and reasoning to organize data and ideas in order topromote understanding.Research problemA general statement of an issue meriting research. It is usuallyused to help formulate a research project and is the basis onwhich specific research questions, hypotheses or statementsare based.Research questionA theoretical question that indicates a clear direction andscope for a research project.Sample
The small part of a whole (population) selected to show whatthe whole is like. There are two main types of samplingprocedure: random and non-random.Sampling errorThe differences between the random sample and thepopulation from which it has been selected.Scientific methodThe foundation of modern scientific enquiry. It is based onobservation and testing of the soundness of conclusions,commonly by using the hypothetico-deductive method. Thefour-step method is: identification of a problem, formulation of ahypothesis, practical or theoretical testing of the hypothesis,rejection or adjustment of the hypothesis if it is falsified.Secondary sourcesSources of information that have been subject to interpretationby others, usually in the form of publications.SemioticsThe study of signs. A type of analysis, particularly of text andvisual material, intent on revealing hidden or implicit meanings.Sense of understandingA complete explanation of a phenomenon provided by a widerstudy of the processes that surround, influence and cause it tohappen.Set of laws theoryA theory that comprises a set of separate, though interrelated,laws.Simple random samplingA method used to select cases at random from a uniformpopulation.Simple stratified sampling
A method that recognizes the different strata in the populationin order to select a representative sample.StatementAn assertion based on a combination of concepts.StructuralistAn approach that all cultural phenomena are primarily linguisticin character, and that subjectivity is formed by deep ‘structures’that lie beneath the surface of social reality.SubjectThe participant in a research project. The term implies apassive role in the project, i.e. that things are done to thesubject in the form of a test or an experiment.Sub-problemA component of a main problem, usually expressed in lessabstract terms to indicate an avenue of investigation.SymbolA sign used to communicate concepts in the form of natural orartificial language.Systematic samplingA sampling method that selects samples using a numericalmethod, e.g. selection of every tenth name on a list.TermA word used to express a definite concept. Terms can beprimitive, which cannot be described by using other terms, orderived, which can.Theoretical samplingSelection of a sample of the population that you think knowsmost about the subject. This approach is common in qualitativeresearch where statistical inference is not required.Theory
A system of ideas based on interrelated concepts, definitionsand propositions, with the purpose of explaining or predictingphenomena.ValidityThe property of an argument to correctly draw conclusionsfrom premises according to the rules of logic.
ReferencesAckrill, M. (1994) ‘A few notes to assist you with the presentationof your report/thesis’. Unpublished.Alldred, P. and Gillies, V. (2002) ‘Eliciting research accounts:re/producing modern subjects?’, in M. Mauthner, M. Birch, J.Jessop and T. Miller (eds), Ethics in Qualitative Research.London: Sage. Chapter 8, pp. 148–65.Barry, A., Born, G. and Weszkalnys, G. (2008) ‘Logics ofInterdisciplinarity’. Economy and Society, 37(1) 20–49.Beardsley, M.C. (1975) Thinking Straight: Principles of Reasoningfor Readers and Writers. Englewood Cliffs, NJ: Prentice-Hall.Beck, R.N. (1979) Handbook to Social Philosophy. New York:Macmillan.Bell, L. and Nutt, L. (2002) ‘Divided loyalties, divided expectations:research ethics, professional and occupational responsibilities’,in M. Mauthner, M. Birch, J. Jessop and T. Miller (eds), Ethics inQualitative Research. London: Sage. Chapter 4, pp. 70–90.Benfield, J. A. and Szlemko, W. J. (2006) ‘Internet-Based DataCollection: Promises and Realities’. Journal of ResearchPractice, 2 (2): article D1 2006Bhaskar, R. (1989) Reclaiming Reality: A Critical Introduction toContemporary Philosophy. London: Verso.Blaxter, L., Hughes, C. and Tight, M. (2010) How to Research, 4thedn. Buckingham: Open University Press.
Booth, W.C., Colomb, G.G. and William, J.M. (2016) The Craft ofResearch, 4th edn. Chicago: University of Chicago Press.Broadbent, G. (1988) Design in Architecture: Architecture and theHuman Sciences. London: Fulton.Brown, J. and Sime, D.J. (1981) ‘A methodology of accounts’, inM. Brenner (ed.), Social Method and Social Life. London:Academic.Bryman, A. and Burgess, R.G. (eds) (1994) Analysing QualitativeData. London: Routledge.Burnett, A., Sheehy, J. and Digby, A. (1994) Theory, Methodologyand Techniques of Research in the Humanities 1: ResearchDesign. Oxford: EMU Oxford Brookes University.Camphell, D.T. and Stanley, J.C. (1966) Experimental and Quasi-Experimental Designs for Research. Boston: Houghton Mifflin.Chalmers, A.F. (1999) What Is This Thing Called Science? 3rdedn. Milton Keynes: Open University Press.Clark, V.L.P. and Creswell, J.W. (2008) The Mixed MethodsReader. Thousand Oaks, CA: Sage.Cohen, L., Manion, L. and Morrison, K. (2017) Research Methodsin Education, 8th edn. London: Routledge.Collier, A. (1994) Critical Realism: An Introduction to RoyBhaskar’s Philosophy. London: Verso.Cottrell, S. (2019) The Study Skills Handbook, 5th edn. London:Red Globe Press.Creswell, J.W. (2014) Educational Research: Planning,Conducting, and Evaluating Quantitative and Qualitative
Research, 4th edn. Harlow: Pearson Education.Davis-Berman, J. and Pestello, F.G. (2018) Exploring Anti-Depressant Medication Use: Using Postings to an InternetMedical Discussion Board. Sage Research Methods Cases.http://dx.doi.org/10.4135/9781526441461Denzin, N. (2010) ‘Moments, mixed methods, and paradigmdialogs’. Qualitative Inquiry, 16(6): 419–27.Dominowski, R.L. (1980) Research Methods. Englewood Cliffs,NJ: Prentice-Hall.Doucet, A. and Mauthner, N. (2002) ‘Knowing responsibly: linkingethics, research practice and epistemology’, in M. Mauthner, M.Birch, J. Jessop and T. Miller (eds), Ethics in QualitativeResearch. London: Sage. Chapter 7, pp. 123–45.Duncombe, J. and Jessop, J. (2002) ‘“Doing rapport” and theethics of “faking friendship’”, in M. Mauthner, M. Birch, J. Jessopand T. Miller (eds), Ethics in Qualitative Research. London:Sage. Chapter 6, pp. 107–22.Fairbairn, G.J. and Winch, C. (2011) Reading, Writing andReasoning: A Guide for Students, 3rd edn. Buckingham: SRHEand Open University Press.Fearnside, W.W. and Holther, W.B. (1959) Fallacy. EnglewoodCliffs, NJ: Prentice-Hall.Feyerabend, P. (1981) ‘How to defend society against science’, inI. Hacking (ed.), Scientific Revolutions. Oxford: OxfordUniversity Press. pp. 157–67.Foucault, M. (1972) The Archaeology of Knowledge. London:Tavistock.
Fox, D. J. (1969) The Research Process in Education. New York:Holt, Rinehart and Winston.Freeman, R. and Meed, J. (1993) How to Study Effectively, 3rdedn. London: Collins Educational.Gall, M.D., Gall, P. and Borg, W.R. (2006) Educational Research:An Introduction, 8th edn. London: Pearson.Goetz, J.P. and LeCompte, M.D. (1993) Ethnography andQualitative Design in Educational Research, 2nd edn. London:Academic.Gottschalk, L. (1951) Understanding History. New York: Knopf.Guba, E. and Lincoln, Y. (1989) Fourth Generation Evaluation.Newbury Park, CA: Sage.Hacking, I. (ed.) (1981) Scientific Revolutions. Oxford: OxfordUniversity Press.Hague, P.N. (1993) Questionnaire Design. London: Kogan Page.Harré, R. (1972) The Philosophies of Science. Oxford: OxfordUniversity Press.Harré, R. (1977) ‘The ethnogenic approach: theory and practice’,in L. Berkowitz (ed.), Advances in Experimental SocialPsychology. Vol. 10. New York: Academic Press.Hewson, C., Yuyle, P., Laurent, D. and Vogel, C. (2016) InternetResearch Methods: A Practical Guide for the Social andBehavioural Sciences, 2nd edn. London: Sage.Hill, J.E. and Kerber, A. (1967) Models, Methods and AnalyticalProcedures in Educational Research. Detroit: Wayne StateUniversity Press.
Hodges, W. (2001) Logic, 2nd edn. Harmondsworth: Penguin.Hughes, J.A. (1997) The Philosophy of Social Research, 3rd edn.Harlow: Longman.Hughes, J.A. and Sharrock, W.W. (1997) The Philosophy of SocialResearch, 3rd edn. Harlow: Longman.Hyland, K. (1999) ‘Academic attribution: citation and theconstruction of disciplinary knowledge’. Applied Linguistics, 20:341–67.Kerlinger, F.N. (1999) Foundations of Behavioral Research, 4thedn. New York: Wadsworth.Krantz, H. and Kimmelman, J. (1992) Keys to Reading and StudySkills, 4th edn. Fort Worth, TX: Harcourt Brace Jovanovich,College.Kuhn, T.S. (2012) The Structure of Scientific Revolutions, 3rd edn.Chicago: University of Chicago Press.Lakatos, I. (1980) Methodology of Scientific ResearchProgrammes: Volume 1: Philosophical Papers, edited by JohnWorrall and Gregory Currie. Cambridge: Cambridge UniversityPress. pp. 91–196.Leedy, P.D. and Ormrod, J. (2015) Practical Research: Planningand Design, 11th edn. Harlow: Pearson Education.Locke, L.C., Spirduso, W.W. and Silverman, S.J. (2007) ProposalsThat Work: A Guide for Planning Dissertations and GrantProposals. Newbury, CA: Sage.Lofland, J., Snow. D., Anderson, L. and Lofland, L.H. (2005)Analysing Social Settings: A Guide to Qualitative Observationand Analysis, 5th edn. Belmont, CA: Wadsworth.
Lyubomirsky, S. (2007) The How of Happiness: A ScientificApproach to Getting the Life you Want. New York: PenguinPress.Margolin, J.B. (1983) The Individual’s Guide to Grants. London:Plenum.Medawar, P.B. (2009) Induction and Intuition in Scientific Thought.Abingdon: Routledge.Miles, M.B. and Huberman, A.M. (2013) Qualitative Data Analysis:An Expanded Sourcebook, 3rd edn. London: Sage.Milgram, S. (1963) ‘Behavioral study of obedience’. Journal ofAbnormal and Social Psychology, 67: 371–8.Mill, J.S. (1973) A System of Logic, Ratiocinative and Inductive:Books I–III and Appendices. London: Routledge and KeganPaul.Miller, T. and Bell, L. (2002) ‘Consenting to what? Issues ofaccess, gatekeeping and “informed” consent’, in M. Mauthner,M. Birch, J. Jessop and T. Miller (eds), Ethics in QualitativeResearch. London: Sage. Chapter 3, pp. 53–69.Moreno, M.A., Goniu, N., Moreno, P.S. and Diekema, D. (2013)‘Ethics of social media research: common concerns andpractical considerations’, Cyberpsychology, Behavior, andSocial Networking, 16(9): 708–13.Morton-Williams, J. (1993) Interviewer Approaches. Aldershot:Dartmouth.Moser, C.A. and Kalton, G. (2016) Survey Methods in SocialInvestigation. Basingstoke: Routledge.
Mouly, G.J. (1978) Educational Research: The Art and Science ofInvestigation. Boston, MA: Allyn and Bacon.Newman, R.J. (1989) Study and Research: A SystematicApproach for All Students. Oxford: Bookmarque.OEED (1991) The Oxford Encyclopedic English Dictionary.Oxford: Oxford University Press.Oliver, P. (2010) The Student’s Guide to Research Ethics, 2ndedn. Maidenhead: Open University Press.Open University (2008) Equality & Diversity in Language andImage. Milton Keynes: Open University.Oppenheim, A.N. (1998) Questionnaire Design, Interviewing andAttitude Measurement, New edn. London: Pinter.Oxford Brookes University (2016) Ethical Standards for ResearchInvolving Human Participants: Code of Practice.www.brookes.ac.uk/research/research-ethicsPhillips, E.M. and Pugh, D.S. (2015) How to Get a PhD, 6th edn.Buckingham: Open University Press.Pirie, M. (1985) The Book of the Fallacy: A Training Manual ofIntellectual Subversives. London: Routledge and Kegan Paul.Preece, R. (1994) Starting Research: An Introduction to AcademicResearch and Dissertation Writing. London: Pinter.Quine, W.V.O. (1969) Ontological Relativity and Other Essays.New York: Columbia University Press.Reynolds, P.D. (2016) A Primer in Theory Construction.Basingstoke: Routledge.
Richards, L. and Richards, T. (1994) ‘From filing cabinet tocomputer’, in A. Bryman and R.G. Burgess (eds), AnalysingQualitative Data. London: Routledge. pp. 146–72.Robson, C. and McCartan, K. (2016) Real World Research, 4thedn. Oxford: Blackwell.Roeser, K., Obergfell, F., Meule, A., Vögele, C., Schlarb, A. andKübler, A (2012) ‘Of larks and hearts –morningness/eveningness, heart rate variability andcardiovascular stress response at different times of day’.Physiology & Behavior, 106(2): 151–7.Russell, B. (1912) The Problems of Philosophy. London: Williams& Norgate.Salmon, W.C. (2012) Introduction to Logic and Critical Thinking,6th edn. Boston: Centage Learning.Schimel, J. (2011) Writing Science: How to Write Papers That GetCited and Proposals That Get Funded. Oxford: OxfordUniversity Press.Seale, C. (2004) Researching Society and Culture, 2nd edn.London: Sage.Shadish, W.R., Cook, T.D. and Campbell, D. (2001) Experimentaland Quasi-Experimental Designs for Generalized CausalInference, 2nd edn. Andover: Cengage Learning.Silverman, D. (2012) ‘Research and theory’, in C. Seale (ed.),Researching Society and Culture, 3rd edn. London: Sage.Slater, D. (1995) ‘Analysing cultural objects: content analysis andsemiotics’, in C. Seale (ed.), Researching Society and Culture,1st edn. London: Sage.
Smith, M. and Smith, G. (1990) A Study Skills Handbook, 2ndedn. Oxford: Oxford University Press.Strauss, A.L. (1987) Qualitative Analysis for Social Scientists.Cambridge: Cambridge University Press.Strauss, A.L. and Corbin, J. (2015) Basics of QualitativeResearch: Techniques and Procedures for DevelopingGrounded Theory, 4th edn. Newbury Park, CA: Sage.Swales, J. and Feak, C. (2000) English in Today’s ResearchWorld: A Writing Guide. Michigan: University of Michigan Press.Symonds, J. and Gorard, S. (2008) The Death of Mixed Methods:Research Labels and Their Casualties. The British EducationalResearch Association Annual Conference, Heriot WattUniversity, Edinburgh, September 3–6.Tait, J. and Lyall, C. (2007) Short Guide to DevelopingInterdisciplinary Research Proposals. ISSTI Briefing Note, 1.Townend, D. (2000) ‘Can the law prescribe an ethical frameworkfor social science research?’, in D. Burton (ed.), ResearchTraining for Social Scientists. London: Sage.Turner, B.A. (1994) ‘Patterns of crisis behaviour: a qualitativeenquiry’, in A. Bryman and R. Burgess (eds), AnalysingQualitative Data. London: Routledge. pp. 192–215.van Dalen, D.B. (1979) Understanding Educational Research: anIntroduction, 4th edn. New York: McGraw–Hill.van Dijk, T.A. (1994) ‘Discourse and cognition in society’, in D.J.Crowley and D. Mitchell (eds), Communication Theory Today.Palo Alto, CA: Stanford University Press.
Wallace, L.N. (2018) Researching Gun Control Attitudes with WebSurveys. SAGE Research Methods Cases Part 2.https://dx.doi.org/10.4135/9781526437631Weber, M. (1949) The Methodology of the Social Sciences (trans.and ed. E.A. Shils and H.A. Finch). Glencoe, IL: Free Press.Wiersma, W. and Jurs, S. (2008) Research Methods in Education:An Introduction, 9th edn. New York: Pearson.Williams, M. and May, T. (2002) Introduction to the Philosophy ofSocial Research. London: Routledge.Yanchar, S. and Williams, D. (2006). ‘Reconsidering thecompatibility thesis and eclecticism: five proposed guidelines formethod use’. Educational Researcher, 35(9): 3–12.Yin, R.K. (2018) Case Study Research and Applications: Designand Methods, 6th edn. London: Sage.Zohar, D. and Marshall, I. (1993) The Quantum Society. London:Bloomsbury.
Indexabstract, 326, 346, 351abstract concept, 82–83, 88, 233, 262abstractness, 26accounts, 18, 33, 41, 151, 154, 174, 189, 216, 223, 233, 237,264, 266, 269action research, 44–45, 315–316, 320aims of research, 38, 81algorithm, 246analogy, 110–113analysing problems, 31analysisarchival, 213content, 19, 218–221, 262discourse, 19, 82qualitative, 214, 249, 255, 264, 265quantitative, 255, 264statistical, 14, 221–222, 232, 252systems, 16analyticalreading, 95review, 142survey, 12archival analysis, 213area sampling, 225argument, 5–6, 20, 36, 54, 57, 75–77, 80, 95–111, 117, 131,148–150, 154, 158, 164–165, 188, 266, 276, 288, 297, 306,309, 322, 329–334Aristotle, 5, 56, 96, 109assessment, 10, 19, 47, 50, 70, 77, 81, 85, 116–117, 144, 148,152, 153, 156, 165, 181, 203, 205, 207, 218, 234, 276, 298,317, 319, 324, 336association, 12, 13, 89, 90, 97associational statement, 97authentication, 146, 151, 218axiomatic theory, 94
backgroundinformation, 36, 50, 80, 121investigation, 51, 303reading, 31, 115, 275, 293, 303, 328to research, 54social, 63, 312studies, 31theoretical, 7, 8, 92, 157, 158, 271bias, 10, 12, 85, 151, 173, 177, 180, 191, 223, 229, 235, 236,239, 240, 243bibliographicaids, 124, 125database, 124, 143bibliography, 122, 137, 141, 144, 342–343catalogue, 8, 124, 145, 217category, 7, 86, 109, 115, 256, 274causalprocess, 93statement, 97cause and effect, 13–15, 97, 241, 262citation, 136, 138, 143, 155, 156, 169, 170, 171, 202, 204, 321,327, 343class, 11, 178, 222, 224, 277, 312classification, 81, 91, 97, 98, 110–113, 219, 256cluster sampling, 225coding, 19, 44, 190, 219, 256–259, 280common mistakes, 1, 29comparativeentity, 87measurement, 86comparisonof differences, 11of sets of data, 29computercourse, 309program, 46, 257, 260, 302technique, 238terminal, 124concept
abstract, 82, 83, 88, 233, 262concrete, 82, 83measurement, 84quantification of, 98, 250of research, 2theoretical, 84–88, 98, 99, 291conceptual model, 90conclusions, 3, 5, 9, 12, 15, 17, 21, 24, 28, 29, 40, 48, 54, 56–58, 76, 80, 95, 101, 104, 107, 108, 120, 130–135, 140, 148–150, 163, 190, 191, 203, 204, 227, 238, 255, 266, 271, 292,304, 309, 321, 331, 332, 336, 346, 351conclusion indicators, 100, 101, 149consistency, 105, 235, 319, 339, 346, 351content analysis, 19, 218, 219, 220, 221, 262controldata, 196, 257quality, 46, 253samples, 15of variables, 15correlation, 12, 13, 43, 87, 88, 89, 97, 101, 243, 252, 255coefficient of, 30, 252degree of, 29research, 12, 13, 252costs, 17, 46, 47, 103, 110, 121, 155, 218, 235, 300, 301, 310criticalpath diagram, 245rationalist approach, 62, 67realism, 71, 72culturalbackground, 18context, 16exchange, 19meaning, 19, 219reality, 67, 71texts, 19dataanalysis, 34, 35, 174, 187, 189, 205, 208, 211, 273, 274,275, 280, 299, 304, 324
collection, 211–212, 256–260, 268, 270, 273, 274, 275, 276,277, 278, 299, 313, 318, 355display, 255, 259, 260interpretation of, 18, 29measurement, 88mining, 218, 221primary, 190, 214–218, 222, 235, 275, 278, 313protection, 197secondary, 214–218, 235, 275, 277, 313sets of, 15, 29, 250database, 45–47, 124, 126, 136, 139, 140, 142, 143, 144, 146,171, 221, 260, 323deduction, 5–6, 56, 58, 74, 76, 104deductiveargument, 6, 103–108reasoning, 5, 57, 76definitionoperational, 84–86, 90, 114–116of research, 23of research problem, 23dependent variable, 14, 97, 98, 236, 261, 290derived term, 81diagram, 21, 214, 244, 245, 262, 264, 274, 285diagrammatic models, 244diaries, 217, 230, 233, 264discourse, 19, 20, 82, 99, 100, 106, 114, 160, 265, 268analysis, 19, 82display, 237, 245, 251, 256, 260–262, 264, 311drafting, 333, 334, 337, 339empirical, 3, 6, 26, 27, 68, 90, 91, 92, 94, 106, 213, 293, 295,312, 316research, 94epistemology, 174, 208, 268estimate, 102, 121, 253ethics, 167–209, 268, 278, 304, 308ethnogenic, 18, 233ethnography, 19ethnomethodology, 54, 69
evaluation, 8, 9, 15–17, 46, 47, 62, 149, 156, 266, 300, 315,316evaluation research, 15, 17evidence, 3, 9, 10, 43, 50, 57, 61, 63, 91, 92, 93, 94, 122, 131,133, 149, 152, 162, 171, 172, 173, 216, 260, 272, 294, 295,308, 314, 316, 336empirical, 94, 213existence statement, 96experience, 4–6, 10, 11, 27, 32, 39, 40, 43, 56, 58, 68, 80, 92,106, 160, 161, 183, 233, 268, 271, 286, 301, 303, 307, 309,312, 315, 323, 341, 345experiment, 11, 14, 55, 58, 68, 188, 238–243, 247, 248, 297experimental research, 14, 17explanation, 7, 16, 90, 107, 122, 150, 213, 233, 234, 258, 260,281, 287, 288, 302, 321, 345external, 68, 92, 243validity, 243fallacy, 73, 109, 114, 117falsification, 58, 60, 172feministresearch, 18, 208theory, 72figures, 154, 162, 313, 340, 341population, 54footnotes, 337, 341formal fallacies, 109Foucault, 20, 67, 73, 268Gaussian curve, 252, 254generality, 251generalizability, 17, 19, 92, 219, 236, 240, 270historicalrecords, 126research, 9, 10, 126honesty, 167–173, 185, 202, 203hypotheses, 6, 9, 55, 59, 60, 90, 91, 92, 94, 114, 115, 134,249, 253, 256, 292–296, 303, 308, 336
hypothesis, 5, 14, 58, 59, 80, 91, 94, 95, 240, 244, 254, 273,293, 295, 296, 299, 303hypothetical construct, 84hypothetico-deductive method, 56, 60, 61, 74, 171illustrations, 133, 219, 283, 337, 340, 349, 351independent variable, 14, 17, 97, 236, 238, 243, 290index, 126, 130, 137, 192, 257, 278indicator, 85, 100, 122, 295induction, 56, 57, 58, 74, 104, 107inductive, 5, 6, 56, 57, 59, 61, 76, 103, 104, 105, 106, 107argument, 5, 57, 103, 104, 105, 106, 107reasoning, 5, 6, 56, 61inferential statistics, 253, 264informed consent, 181, 186, 187, 188, 191, 195–197, 205, 208,313instrumentation, 35, 94, 235intellectual property, 190, 202, 302interim summary, 257, 259internal validity, 243Internet, 46, 121, 123–128, 145, 186, 187, 198, 209, 216–218,235, 236, 279, 281, 282, 344interpretation, 3, 4, 10, 14–19, 29, 60, 68, 80, 85, 95, 184, 190,207, 219, 230, 233, 252, 259, 267, 268, 280, 298of data, 18, 29interpretivism, 69, 71, 73intersubjectivity, 26interviews, 12, 41, 42, 44, 45, 54, 74, 132, 170, 182, 184, 202,216, 219, 220, 223, 230–232, 236, 237, 256, 257, 270, 278,299, 313, 318journal, 120, 121, 124, 125, 138, 141, 190, 277, 326, 328, 342,343, 351knowledgecontribution to, 27, 288, 300, 304, 308, 346nature of, 53, 61, 74, 76, 77scientific, 6, 24, 61, 77, 93, 95, 97, 175Kuhn, 62, 63, 64, 66, 73, 78, 117
language, 2, 3, 18–20, 72, 79, 81, 82, 95, 99, 107, 109, 114,149, 151, 171, 177, 182, 199, 230, 236, 250, 266–268, 283,285, 294, 313law, 80, 92, 94, 95, 195, 208levelof abstraction, 83, 96, 286of association, 90of measurement, 87, 88library, 32, 45, 51, 119–126, 128, 134–138, 141, 142, 144, 145,216, 217, 276, 279–281, 344catalogue, 124, 134, 142, 145, 276, 280, 281literature review, 33, 35, 77, 95, 146, 147–166, 221, 288, 289,291, 303, 313, 321, 323, 326logic, 15, 57, 58, 64, 70, 96, 99, 102, 105, 107, 109, 117, 127,149, 186, 213, 221, 235, 265, 266, 297of discovery, 64logicalindicator, 99, 100, 101, 102rigour, 24truth, 106map, 50, 150, 158, 244, 245, 262, 277mathematicalmodel, 246proposition, 107sign, 81mathematics, 26, 81matrix, 214, 262, 273measurementconcept, 84levels of, 86, 88, 89, 249, 260memo, 258methodology, 3, 18, 38, 40, 43, 44, 45, 63, 115, 135, 157, 178,212, 235, 266, 268, 282, 283, 299, 307, 308mistakes, 9, 29, 98, 108, 204, 334, 340mixed methods, 211–213, 268–274, 282–283model, 17, 18, 43, 63, 71, 72, 81, 141, 162, 224, 238, 244–248, 263, 314–317decision tree, 263, 266diagrammatic, 244
evaluation, 16mathematical, 246physical, 245, 247network, 50, 136, 146, 207, 260, 262, 266causal, 255computer, 298nominal level of measurement, 86non-parametric statistics, 251, 264non-random sampling, 225, 236notesfield, 44, 256, 258, 259taking, 132, 134, 135, 146null hypothesis, 254numbering, 124, 299, 341, 342objectives of research, 1, 6, 26objectivity, 10, 19, 73, 173, 219, 266observation, 5, 10–12, 17, 18, 26, 37, 44, 54–62, 68, 76, 92,160, 190, 195, 217, 228, 230, 234, 236, 266, 313operational definition, 84, 85, 86, 90, 114–116operationalization, 295orderrank, 86, 87ordered displays, 262ordinal level of measurement, 86, 87, 286outline of methods, 289, 298, 323outputs, 200, 289, 298, 300, 302, 313paradigm, 63, 64, 67, 95, 152, 162, 268, 315, 320paragraphs, 332, 335parameters, 16, 37, 67, 121, 127, 253, 254, 288parametric statistics, 251, 254parsimony, 92personnel, 125, 216, 289, 300, 301, 309, 313perspective, 63, 67, 70, 75, 77, 163, 164, 174, 214philosophy, 20, 53, 54, 72, 73, 77, 117, 153, 154, 163, 208of research, 53, 117physical model, 245, 247physical survey, 234
pilot study, 229, 317, 345plagierism, 121, 168, 169, 204planning, 24, 31, 34, 133, 153, 155, 245, 282research, 38–40, 194, 211, 282, 304, 325, 327, 331–332,354–355population, 12, 54, 84, 103, 194, 216, 223–227, 230, 231, 236,237, 250, 251, 253, 254, 275, 299, 310positivism, 67, 69, 71postmodernism, 19, 73prediction, 7, 13, 14, 90, 182, 185, 221, 254, 290prediction studies, 13preliminary analysis, 42, 50premise, 5, 99, 100–104, 107indicators, 100presentation, 20, 101, 103, 204, 230, 236, 250, 261, 262, 268,294, 328, 340, 342, 346–351primarydata, 190, 214–218, 222, 235, 275, 278, 313research – using the internet for, 235–237sources, 214primitive term, 81problemarea, 23, 29–33, 48–51, 75, 308definition, 33, 307of induction, 57, 58research, 1, 2, 4, 27–38, 40, 48–51, 55, 67, 74–76, 80, 93,101, 113, 115, 120, 128, 133, 134, 148, 157, 158, 164, 165,211, 212, 219, 223, 233, 248, 269, 272, 273, 276, 285, 288,291–298, 302–304, 306, 322, 323, 330of scaling, 245solving, 269sub-, 34–36, 49, 292, 299, 331of writing, 285programme of work, 321, 353proposal, 36–38, 46–47, 77, 116, 148, 154, 156, 159, 181, 200,201, 211, 273, 275–292, 297–310, 314, 320–328proposition, 30, 94, 104, 296hypothetical, 3mathematical, 107punctuation, 136, 184, 204, 338, 339, 341, 342
purposive sampling, 225qualitativedata, 210, 231, 233, 234, 250, 255, 257, 261, 270, 274,278, 280research, 12, 174, 189, 212, 213, 225, 257, 258, 268, 274,280, 299, 304quantification, 86, 87, 88, 89, 90, 97, 98, 115, 250quantitativeanalysis, 255, 264data, 213, 214, 231, 245, 250, 255, 270, 271model, 244, 245, 246research, 13, 258, 268, 269, 273, 274quasi-experiment, 15, 214questionresearch, 93–94, 148, 152, 158–161, 164, 165, 213, 215,252, 259, 269, 275, 277, 286, 294, 295, 303, 310–312, 315,321questionnaire, 42, 43, 46, 47, 85, 115, 180, 182, 190, 223,228–232, 235, 237, 264, 270, 318, 345Quine, 92quotation, 156, 170, 337–339, 343, 345mark, 337–339, 345randomnumber generation, 246sampling, 224, 225, 228rank, 86, 87, 254ratio level of measurement, 88readinganalytical, 95background, 31, 115, 275, 293, 303, 328realism, 55, 78realitycultural, 67, 71social, 54, 69, 73reasoningdeductive, 5, 57, 76inductive, 5, 6, 56, 61reconciliatory approach, 67, 70, 71
redrafting, 50, 139, 334references, 11, 40, 41, 43, 117, 124, 125, 128, 133, 136, 139,141, 143, 144, 145, 154, 156, 157, 159, 162, 170, 171, 233,289, 291, 297, 307, 308, 314, 321, 332, 335, 341, 342, 343,344, 345relational statement, 97relativism, 62reliability, 15, 19, 57, 85, 94, 95, 151, 189, 218, 219, 223, 233,235, 236, 238, 240, 258, 317, 318, 319researchapproach, 13, 34, 36, 39, 47, 55, 75, 76, 80, 113–116, 164,165, 174, 296, 298, 307, 309definition, 297design, 34, 35, 77, 278, 283, 311, 313methods, 2, 21, 23, 24, 27, 55, 76, 80, 117, 120, 154, 163,164, 165, 167, 183, 200, 202, 210, 211, 214, 218, 220, 268,269, 270, 271, 273, 276, 277, 283, 288, 295, 296, 302, 304,307, 309, 321, 323, 324, 348, 351objectives, 40, 50, 217, 289, 298problem, 1, 2, 4, 27–40, 48–51, 55, 67, 74–76, 80, 93, 101,113, 115, 120, 128, 133, 134, 148, 157, 158, 164, 165, 211,212, 219, 223, 233, 248, 269, 272, 273, 276, 285, 288,291–298, 302, 303, 304, 306, 322, 323, 330process, 1, 2, 3, 14, 21–25, 44, 45, 50, 120, 176, 179, 181,203, 262, 277, 278, 303proposal, 23, 36, 38, 115, 148, 154, 157, 162, 164, 174,200, 201, 210, 211, 272, 282, 284, 285, 287, 289, 292, 300,302, 307, 309, 321, 322, 325, 335qualitative, 12, 174, 189, 212, 213, 225, 257, 258, 268, 274,280, 299, 304quantitative, 13, 258, 268, 269, 273, 274question, 93, 94, 148, 152, 158, 159, 161, 164, 165, 213,215, 252, 259, 269, 275, 277, 286, 294, 295, 303, 310, 311,312, 315, 321skills, 38strategy, 1, 50, 75, 213, 274respect, 83, 87, 130, 139, 168, 178, 179, 198, 202, 250, 292responsive evaluation, 16, 17retrieving notes, 136
sample, 12, 15, 41–44, 144, 187, 223, 224–231, 236, 240, 242,243, 250, 253, 254, 270, 275, 299, 317samplingcluster, 225error, 227random, 224, 225, 228stratified, 224scale, 5, 12, 17, 43, 48, 75, 80, 86, 87, 88, 89, 113, 172, 189,233, 238, 245, 262, 286, 288, 299, 311, 317–319scanning, 130, 132, 143scientificenquiry, 6, 16, 54–62, 68, 73, 75, 76, 113knowledge, 6, 24, 61, 77, 93, 95, 97, 175method, 2, 5, 6, 10, 53, 55, 60, 66, 67, 69, 76, 78, 174scope, 4, 10, 12, 20, 31, 35, 36, 39, 45, 50, 76, 93, 115, 154,223, 229, 238, 246, 272, 276, 286, 287, 292, 294, 297, 299,323search engine, 46, 121, 123, 127secondarydata, 214–218, 235, 275, 277, 313sources, 124, 170, 215, 218, 219, 277semiotics, 19, 267sense of understanding, 7, 26, 81, 97sequential analysis, 259, 260simple random sampling, 224socialcontext, 20, 268facts, 54, 67, 68phenomena, 11, 68, 71, 256, 257problems, 29, 148reality, 54, 69, 73science, 13, 18, 54, 66–71, 76, 78, 84, 87–93, 141, 149,152, 166, 171, 208, 211, 271, 278, 281system, 70, 83theory, 20, 268sociology, 68, 72, 73, 280spelling, 127, 190, 334, 339, 340stability, 7, 176standard deviation, 88, 227, 253, 254statement
of intent, 293, 297observation, 56, 57, 60, 61theoretical, 61, 90, 93, 94, 95, 171statisticsinferential, 253, 264non-parametric, 251, 264official, 54, 122, 215parametric, 251, 254stratified sampling, 224structureof argument, 105structured interview, 37, 41, 232, 237, 264, 313, 317–319Strucuralsim, 73sub-hypotheses, 295problems, 34–36, 49, 292, 331questions, 49, 164, 165, 219, 271, 286, 294, 295, 303subjectivity, 73surveyanalytical, 12physical, 234research, 214, 276, 278technique, 12telephone, 232visual, 234syllogism, 5, 109symbol, 81, 82, 127symbolic, 26, 69, 312interactionism, 69logic, 26systematic sampling, 187systems, 16, 19, 26, 44, 48, 123, 141, 143, 170, 181, 183, 217,219, 225, 244, 245, 247, 269, 298, 313, 342analysis, 16tables, 337, 340, 341taxonomy, 277technical term, 80, 99, 104, 252, 267theoreticalconcept, 84, 85, 86, 87, 88, 98, 99, 291
sampling, 225statement, 61, 90, 93, 94, 95, 171theorygrounded, 43, 44, 92, 223, 311social, 20, 268timetable, 289, 302, 304, 324, 328, 345title, 114, 120, 124, 125, 130, 131, 134, 138, 140, 141, 285,289, 290, 296, 303, 307, 321, 323, 336, 340–343, 349, 351truth, 57–59, 62, 65, 93–95, 103–107, 171, 206typology, 7, 81, 96, 97, 160, 244unstructured interview, 235, 266validity, 39, 54, 73, 91, 92, 94, 105–109, 114, 131, 153, 174,190, 229, 235, 236, 240, 242, 243, 271, 317–319valueof concepts, 81of research, 202statistical, 254variable, 8, 14, 15, 97, 98, 121, 222, 242, 246, 295dependent, 14, 97, 98, 236, 261, 290independent, 14, 17, 97, 236, 238, 243, 290vignette, 259Weber, 67, 70, 71Wittgenstein, 19World Wide Web, 121, 122writingnotes, 133, 232process, 304, 329, 330a proposal, 325
Collepals.com Plagiarism Free Papers
Are you looking for custom essay writing service or even dissertation writing services? Just request for our write my paper service, and we'll match you with the best essay writer in your subject! With an exceptional team of professional academic experts in a wide range of subjects, we can guarantee you an unrivaled quality of custom-written papers.
Get ZERO PLAGIARISM, HUMAN WRITTEN ESSAYS
Why Hire Collepals.com writers to do your paper?
Quality- We are experienced and have access to ample research materials.
We write plagiarism Free Content
Confidential- We never share or sell your personal information to third parties.
Support-Chat with us today! We are always waiting to answer all your questions.
