The research lifecycle is the process a researcher takes to conduct a ?study from inception to completion. For this assignment, you must write a ?paper that describes the research

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/200629644

Research as a Process: A Comparison between Different Research Approaches

Article · January 2009

CITATIONS

7 READS

22,941

2 authors, including:

Some of the authors of this publication are also working on these related projects:

MomPower View project

Virtual Collaborative Learning (VCL) View project

Helena Lovasz-Bukvova

IMC University of Applied Science Krems

38 PUBLICATIONS   766 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Helena Lovasz-Bukvova on 22 May 2014.

The user has requested enhancement of the downloaded file.

Working Papers on Information Systems ISSN 1535-6078

Research as a Process: A Comparison between Different Research Approaches

Helena Bukvova Technische Universität Dresden, Germany

Abstract The research process plays a central role in science. To support research and researchers, it is necessary to understand the research process and its phases. Although the literature offers different research processes, these are often focused on specific research paradigms and methods. The aim of this working paper is to define a research process common to different research approaches: behavioral science, design science and action research. The paper reviews research process models from the literature and finally introduces a model of a general research process. The model can be used to further explore the tasks of researchers, with the aim of identifying possibilities of supporting researchers.

Keywords: research process, research, behavioral science, behavioral science, design science, action research

Permanent URL: http://sprouts.aisnet.org/9-29

Copyright: Creative Commons Attribution-Noncommercial-No Derivative Works License

Reference: Bukvova, H. (2009). "Research as a Process: A Comparison between Different Research Approaches," . Sprouts: Working Papers on Information Systems, 9(29). http://sprouts.aisnet.org/9-29

Sprouts – http://sprouts.aisnet.org/9-29

1

Introduction

Scientific research provides the means of refining and widening scientific knowledge. The understanding of the principles and methods of scientific research is therefore cru- cial for every scientist. The education of research methods plays an important role in academic education, particularly at universities. Different scientific disciplines have de- veloped different research methods and practices. However, as this paper documents, these differences are on the operative level. With regard to the research process as a whole, researchers throughout different disciplines follow the same procedure.

Research, as a systematic enquiry leading to the construction of new knowledge (compare Bordens & Abbott, 2007, p. 2; Graziano & Raulin, 2009, p. 26), does not take place solely in science. It is also carried out daily by each individual (compare Bannister & Fransella, 2003). However, scientific research (also scholarly research or academic research) follows particular guidelines and procedures to ensure the quality of research results. Scientific research intents to create scientific knowledge in particular field (Hockey, 2000, p. 3) through the process of systematic scientific enquiry, the research process (Clark & Hockey, 1989). The research process as well as the research results have to fulfil certain standards (Heinrich, 1993, pp. 62-66; Shugan, 2004, pp. 174-175). Among others, scientific research must be public, replicable, unprejudiced and independent and it must advance the state of the art (Heinrich, 1993, pp. 62-66; Shugan, 2004, pp. 174-175). Due to this crucial role of the research process in science, the understanding as well as the theoretical analysis of the research process are relevant for any research directed towards improving and supporting science (compare Söldner, Haller, Bullinger, & Möslein, 2009).

In this working paper I introduce several models and descriptions of the research process. My aim is to discuss the models, contrast the approaches and point out similar- ities. I show, that the models and procedures share a number of phases. Differences are based on different aims of the research approaches, which the processes use (behavioural and design science), and other understanding of the role of the researcher. The result of the working paper is a general research process model that can be used to describe the research across the approaches. This working paper is based solely on literature study. It is not the purpose of this paper to introduce new phenomena, but to review und sum up the existing publications on this topic. The paper should serve researchers as a foundation for further enquiries into research practice. The working paper is further based on the radical constructivist understanding of reality (compare e.g. Kenny, 2009; Rusch, 2007).

Research process in the behavioural science

Most research processes represented in the literature follow the principles of be- havioural science. Behavioural science has its origins mainly in natural sciences. Meth- ods of behavioural science seek do describe, explain and predict phenomena (often human behaviour). The result of behavioural research process is the creation or the evaluation of theories. (Hevner, March, Park, & Ram, 2004, p. 2; March & Smith, 1995, 253) This

Sprouts – http://sprouts.aisnet.org/9-29

2

section discusses five models of behavioural research processes: Björk ’s (2007) model of scientific communication, research process by Bordens and Abbott (2007), research process by Graziano and Raulin (2009), process of nursing research by Lacey (2006) and business research process by Blumberg, Cooper, and Schindler (2008).

Idea

Develop idea into testable hypothesis

Choose an appropriate research

design

Choose subject population

Decide on what to observe and the approp. measures

Conduct study

Analyse data

Report results

Library research Casual or systematic

observation

Deductive reasoning

Figure 1. Research process by Bordens & Abbott (2007, p. 24)

The research process by Bordens and Abbott (2007) is a good example for behavioural research. The process is strictly oriented on behavioural science. According to Bordens and Abbott (2007, p. 8), research methods outside of behavioural science are consid- ered non-science. Further, Bordens and Abbott acknowledge only empirical (mainly quantitative) research methods. Figure 1 shows the steps of the process.

The process begins with the generation of new ideas for studying behaviour. If an idea is to be used in scientific research, it has to be clearly defined and the involved variables

Sprouts – http://sprouts.aisnet.org/9-29

3

have to be isolated. The expected relationship of the variables is then described in a hypothesis. The hypothesis serves as a foundation for the study. Once the hypothesis is clear, it is possible to choose the research design (e.g. experimental or correlational study). Then the study subjects are selected. When it was decided, what behaviour exactly will be observed and how it will be measured, the study can take place. Finally, the results of the study are analysed and presented to the scientific community. The study results (as well as the study or the analysis of the data) can trigger a new research process. (Bordens & Abbott, 2007, pp. 22-25)

The research process by Graziano and Raulin (2009) also focuses on behavioural science, but it is less strict than the process by Bordens and Abbott (2007). Graziano and Raulin (Graziano & Raulin, 2009, 2) also define science as a process of enquiry. Science acquires its knowledge through observation (empiricism), but also through reasoning (rationalism) (Graziano & Raulin, 2009, 9-10).

Initial idea

Personal experience

Other's research

Problem definition

Procedures design

Observation

Data analysisInterpretationCommunication

Stimulating other's research

Figure 2. Research process by Graziano & Raulin (2009, p. 40)

The process (see Figure 2) begins with the generation of an initial idea. Personal experience or existing research can serve as an inspiration for a new research process. To explore the idea with the help of scientific research, it has to be clearly defined. In the next step, therefore, the problem to be addressed is described in the form of a research question. The research procedure that should lead to the solution of the research question is defined in the procedure-design phase. The resulting research design determines the study participants and conditions as well as the data-collection and data- analysis methods. After the observation has been carried out, the data is analysed and interpreted. The final communication of the results to the scientific community can trigger a new research process or stimulate the activity of other researchers. (Graziano

Sprouts – http://sprouts.aisnet.org/9-29

4

& Raulin, 2009, pp. 36-41)

Björk (2007) describes a very complex model of a research process (see Figure 3). Al- though Björk does not clearly define his understanding of science, the terms he uses place the model in behavioural science. The focus of the model is scientific communication. The model therefore distinguishes activities that serve to acquire existing knowledge from activities that generate new knowledge. The inputs of the process are ”scientific problems” and ”existing knowledge”. By studying the existing research knowledge, the researchers devise a conceptual framework and hypotheses for further research. Then, data from existing repositories are collected and analysed. The researchers then do ex- periments and make observations with selected scientific methods. The collected data as well as the new empirical data is analysed in order to draw conclusions and create new scientific knowledge. (Björk, 2007, p. 16) The research process is further embedded in a broader process called ”Do research, communicate and apply the results” consisting of the stages Fund R&D, Perform the research, Communicate the results and Apply the knowledge (Björk, 2007).

Figure 3. Research process by Björk (2007, p. 16)

Lacey (2006) presents a research process from the area of nursing. The process (Figure 4) and its steps are described in a very general way, making them suitable also for other disciplines.

The research process begins with the development of a research question. The re- search question is often based on an idea or a ‘hunch’. To serve as a foundation for a research, the idea has to be refined to a research question. The researchers then study

Sprouts – http://sprouts.aisnet.org/9-29

5

Developing the research question

Searching and evaluating the literature

Choice of methodology and research design

Preparing a research proposal

Gaining access to the data

Sampling

Data collectionData analysis Dissemination of

the results Implementation of the research

Figure 4. Research process by Lacey (2006, p. 17)

existing literature to determine the state of research related to the problem. The lit- erature search can lead to a further refinement of the research question. Afterwards, the methodology of the study is chosen and the research design is determined. Different research designs (e.g. quantitative or qualitative) are possible. Lacey (2006) points out that all approaches are valid, given that they fit the research question and the needs of the research. In the next stage, a research proposal is prepared, giving details about the planned research. Lacey mentions gaining the access to the necessary data as a separate step, due to ethical and legal issues connected to the use of patients’ data. Once the data is available, a suitable sample can be selected and the data collection takes place (preceded if necessary by a pilot study). The data is analysed according to the selected methods. The results of the research are disseminated, mostly through publications in scientific journals or presentation at conferences. Lacey sees the implementation of the research results as an important part of the research process in nursing.

Finally, Blumberg et al. (2008) describe a business research process (see Figure 5). Their process begins with the development and the exact definition of the research question. The research question of the business research process has to be connected to an existing management problem. Preceding the research design, researchers might have to provide a written research proposal. The research proposal describes the exploration of the management research question. The proposal can be used to obtain funding for the research project. The next phase, the research design, describes the activities leading to the fulfilment of the research objectives. Blumberg et al. point out the benefits of using different methods to prevent bias. The research design begins with the definition of an overall design strategy. Based on this, the relevant population and sampling methods

Sprouts – http://sprouts.aisnet.org/9-29

6

as well as data collection methods are determined. The actual collection of data is generally preceded by pilot testing, to detect problems with the selected instruments. If necessary, the instruments are revised before collecting the data. The data is then analysed and interpreted. Blumberg et al. focus on quantitative research. The data analysis therefore covers mainly aggregation of data and the use of statistical methods. Finally, the research results have to be reported to the target audience. These are, in the case of business research, managers facing management decisions. (Blumberg et al., 2008, pp. 55-76)

Although the number of process steps in the presented research processes from the behavioural science vary, there are clear similarities. All processes begin with an idea. The source of these ideas is not clear. They can be connected to practical problems (e.g. in business management or nursing) or triggered by existing research. The ideas have to be further developed before they can serve as a foundation for research. This ensures, that the research project will have a clear focus. Bordens and Abbott (2007) call for a hypothesis as grounds for the new research. Others also suggest refining the initial idea into a research question. Using research questions as a starting point enables research design on different level of constraint (see Graziano & Raulin, 2009, pp. 41-44), allowing e.g. also qualitative research methods. The planned research methods with regard to sampling, data collection and data analysis are determined in the research design. Deciding upon research methods before actively gathering the data helps maintain the rigour of the research process. It also ensures, that the combination of methods is suitable. After the data-collecting and data-analysing activities have taken place, the results are reported to the targeted audience. Besides the scientific community, for research projects derived from practical problems these are also relevant practitioners. Similarly, applied research sees the implementation of the research results in practice as a part of the research process (compare Björk, 2007; Blumberg et al., 2008; Lacey, 2006). The process models mostly also demonstrate the cyclic character of research. New research ideas can be triggered by research results or practical application of the findings.

Design science

Although most researched process descriptions are founded on behavioural science, there is another possible scientific approach: the design science. The design science (also design research) seeks to develop artefacts (constructs, models, methods and instantia- tions) that solve a particular problem (compare Cole, Purao, Rossi, & Sein, 2005; Hevner et al., 2004; March & Smith, 1995; Simon, 1996). To distinguish between design science and ”simple” design, the artefact has to present an innovation of the existing scientific knowledge base by improving technical, social or informational resources (Hevner, 2007; Järvinen, 2007). The design science and the behavioural science are not incompatible. On the contrary, there is a consensus, that methods of behavioural science can support steps of the design science research(Cole et al., 2005; Hevner et al., 2004; March & Smith, 1995). However, there are clear distinctions. Most prominently, the design science is prescriptive, using the prescriptions to create artefacts, whereas behavioural science fo-

Sprouts – http://sprouts.aisnet.org/9-29

7

Research proposal

Data collection and preparation

Data analysis and interpretation

Research reporting

Policy management decision

Discover research dilemma

Define management dilemma

Define research question(s)

Refine the research question(s)

Design strategy

Data collection design Sampling design

Pilot testing

Instrument revision

Research design

Figure 5. Research process by Blumberg et al. (2008, p. 57)

Sprouts – http://sprouts.aisnet.org/9-29

8

cuses on description and explanation (March & Smith, 1995, 254). This section contains three research processes from the design science: the research process by Vaishnavi and Kuechler (2007), the three cycles by Hevner (2007) and the research process by March and Storey (2008).

The research process by Vaishnavi and Kuechler (2007) (see Figure 6) begins with the awareness of the problem. The awareness can have different sources. It can come from the practice or from research development in science. The output of this process step is a proposal for new research. In the next step, the researchers suggest a tentative design, which becomes a part of the research proposal. The tentative design is an output of a creative, ill-defined activity. The suggested tentative design is implemented in the development phase. The focus of the development phase lies on the methods used, not on the implementation of the artefact itself. The resulting artefact can be only a prototype. The artefact is evaluated in the next process step, using defined performance measures. Finally, in the conclusion of the process the results are consolidated and the knowledge gained in the project is classified.

Awareness of Problem

Suggestion

Development

Evaluation

Conclusion

Proces steps Outputs

Proposal

Tentative design

Artefact

Performance measures

Results

Figure 6. Research process of design science by Vaishnavi & Kuechler (2007)

Hevner (2007) offers a different representation of the research process. Instead of describing linear process steps, Hevner suggests three connected process cycles (see Fig- ure 7). The core of the design process is the design cycle. The cycle iterates between development of design alternative and the evaluation until a satisfactory artefact is reached (Simon, 1996). Design science attempts to improve the environment by design-

Sprouts – http://sprouts.aisnet.org/9-29

9

ing artefacts. The second cycle, the relevance cycle, therefore connects the design to the environment by providing requirements on the design as well as testing the acceptance of the developed artefact. Finally, the rigour cycle ensures that the design research is grounded on theories existing in the knowledge base and contributes new knowledge to the discipline. The three cycles are interconnected. They have to be present and clearly defined in every design science research project.

Application domain

– People – Organisational systems – Technical systems – Problems and opportunities

Foundations

– Scientific theories and methods – Experience and expertise – Meta artefacts (design producs and design processes)

Build design artefacts and processes

Evaluate

Design cycle

Relevance cycle

Rigour cycle

Environment Design science

research Knowledge base

Figure 7. Three cycles of deign science by Hevner (2007) (compare also Hevner et al., 2004)

March and Storey (2008) describe six steps of a design science research process (see Figure 8). The process begins with the identification of a new problem. The researchers use the existing knowledge base of the scientific community to demonstrate, that no adequate solution exists. If the knowledge base offers a suitable solution, this can be used to solve the problem. In this case, the process could not contribute a new artefact to the knowledge base, but it would still be possible to gather new experience from the evaluation of the existing artefact. If no suitable solution is present in the knowledge base, the researchers develop a new artefact addressing the problem. The artefact is evaluated. At the end of the process, the researchers articulate their contribution to the knowledge base, but also to the practice. Similarly, the researchers describe the implications of their research for the scientific community as well as for the practice.

Just as in case of the behavioural science, the research process descriptions from the design science show a number of similarities (although some of them are not explic- itly visible in the process models). All processes begin with an existing problem. This problem generally comes from the practice, although research advances can also be a source of new problems. Because the research problems are grounded in the practice, all three processes stress the necessity for continuous connection with the environment (particularly Hevner, 2007). Also, all three processes point out the need for the inter- action with the existing knowledge base, to identify and compare existing artefacts but

Sprouts – http://sprouts.aisnet.org/9-29

10

Knowledge base

Identification and description of the

problem

Demonstration that no adequate solution exists

Development and presentation of a problem adequate artefact as solution

Evaluation of the artefact

Articulation of the contribution to the knowledge-base and to practice

Explanation of the implications for management and

practice

Figure 8. Research process of design science by March & Storey (2008, 726)

also as a source of appropriate methods. The core of design science is the design of the artefact solving the selected problem. To ensure scientific rigour and practical relevance, the artefact is evaluated. Hevner explicitly depicts the iteration between design and evaluation. Finally, at end of the research process, the researchers have to ensure that new artefacts, methods, theories and experiences become a part of the knowledge base and thus available for further research (Hevner, 2007; March & Storey, 2008; Vaishnavi & Kuechler, 2007).

Action research

Besides the two introduced approaches, I want to introduce another research con- cept: the action research. Action research is not seen here as a different approach, but as a different perspective on the research process. Whereas some representatives par- ticularly of the behavioural science emphasise the need for objective and thus detached perspective of the researcher (compare e.g. Bordens & Abbott, 2007), the action re- search advocates participative research methods (Baskerville, 2001; Cole et al., 2005). In action research, researchers deliberately influence the environment they are studying. Instead of distancing themselves from the subjects of their study, the researchers actively interact and collaborate with them. The creation of new scientific knowledge is thus not separable from the real problem. On the contrary, new knowledge is generated through action. The research itself as well as the research results are directly connected to the specific problem and environmental settings. Finally, the research process has a cyclic form, iterating repeatedly between action and evaluation until a satisfactory result is reached. (Cole et al., 2005; Holter & Schwartz-Barcott, 1993; Hult & Lennung, 1980;

Sprouts – http://sprouts.aisnet.org/9-29

11

Järvinen, 2007; Peters & Robinson, 1984)

Action research has been linked and compared to design science, because both aim to change the environment to solve a given problem (Cole et al., 2005; Iivari, 2007; Järvinen, 2007). Iivari, however, points out that despite their similarities they are ”historically, practically, ontologically, epistemologically and methodologically quite different” (p. 53). Although design science can profit by adopting practices from the action research, this should be reflected with regard to the particular research problem.

Susman and Evered (1978) describe an action research process consisting of five phases (see Figure 9). The process begins with the identification and the definition of the problem (diagnosing). Alternative actions that could solve the problem are then considered (action planning). One of the alternatives is selected and performed (action taking). The consequences of this a

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.