to identify and assess critical and 0.6% other (Mundi, 2018). Singapore is also an immigrant-seeking factors affecting workforce diversity

Received: 8 September 2020 Revised: 5 November 2020 Accepted: 19 November 2020 DOI: 10.1002/sd.2155 RESEARCH ARTICLE Assessing the effects of workforce diversity on project productivity performance for sustainable workplace in the construction industry Daeyoun Won1 | Bon-Gang Hwang2 | Soo Jing Chng2 1 Department of Civil and Environmental Engineering, Seoul National University, Seoul, South Korea Abstract Recent studies on workforce diversity have preached that workforce diversity could 2 Department of Building, National University of Singapore, Singapore, Singapore have a positive impact on productivity if it is managed and utilized well. Due to the global trend of increasing workforce diversity in construction projects, it is crucial to Correspondence Bon-Gang Hwang, Department of Building, National University of Singapore, Singapore 117566, Singapore. Email: [email protected] understand the factors affecting workforce diversity and their impact on productivity for developing a sustainable environment for workforce diversity. However, little attention has been given to how workforce diversity may affect productivity performance in the construction industry. Thus, this study aims to assess the impact of workforce diversity on project productivity performance (PPP). Twenty-one diversity factors were identified via literature review and validated by industry experts, followed by a survey conducted with 58 firms working in Singapore. The responses were analyzed and used to develop a partial least squares structural equation model. The outcomes from the model signified that the diversity aspects such as efficient decision-making and countering the issue of skilled labor shortages had the highest impact in their respective categories of “Skill and education” and “Age and experience.” Therefore, this study contributes to the core body of knowledge and practice both in defining workforce diversity factors and in assessing the relationship between diversity factors and productivity. Furthermore, feasible strategies that match the factors prioritized by the level of the impact on PPP were proposed. These can help construction organizations to understand the untapped potential of workforce diversity and its impact on PPP, ultimately enhancing the productivity of the industry and sustainable diversity in the workplace. KEYWORDS built environment, construction industry, project productivity performance, structural equation modeling (SEM), sustainable workplace, workforce diversity 1 | I N T RO DU CT I O N Singapore [MTI], 2018). However, this productivity growth was primarily driven by sectors such as manufacturing, accommodation, food, Recently, Singapore has made an effort to promote the productivity and finance instead of construction (MTI, 2019). Therefore, these sec- of the entire industry; however, productivity in the construction tors increased their productivity while the construction sector was industry is still in recession. In 2017, Singapore’s overall productivity less productive—even underperforming expectations. Although not grew by 4.5%, which was the highest recorded since 2010, following the worst-performing industry, the construction sector is still deemed the Global Financial Crisis (Ministry of Trade and Industry Republic of as weak in productivity, negatively impacting the country’s overall 398 © 2020 ERP Environment and John Wiley & Sons Ltd. wileyonlinelibrary.com/journal/sd Sustainable Development. 2021;29:398–418. 399 WON ET AL. productivity. In addition, the construction sector is also seen to be of this study can contribute to the body of knowledge and practice underperforming due to its predicted failure to meet the target of a both in defining workforce diversity factors and in assessing the rela- 2–3% productivity growth by 2020 set by the government. tionship between diversity factors and project productivity perfor- On the other hand, Singapore has been encouraging more talent mance (PPP) in the context of the construction industry. Furthermore, from abroad, with a highly diverse workforce in terms of culture and this study can serve as a cornerstone for the industry to handle its race. It is also a multi-ethnic and multi-racial society with a significant workforce diversity properly and achieve the benefits of increased foreign population. In Singapore, there are four major ethnic groups: productivity, by proposing viable strategies to harness diversity to Chinese, Malay, Indian, and others (MTI, 2018)—showing just how increase productivity in the construction industry. diverse Singapore’s society is in terms of ethnicity and religion, partic- This study investigated the relationships between workforce ularly in comparison to other countries such as Japan, which has an diversity and productivity in the construction industry. Therefore, the ethnic breakdown of 98.5% Japanese, 0.5% Koreans, 0.4% Chinese, following objectives were identified: (i) to identify and assess critical and 0.6% other (Mundi, 2018). Singapore is also an immigrant-seeking factors affecting workforce diversity, (ii) to assess the impact of work- country, aiming to attract foreign talent to boost its human resource force diversity on project productivity, and (iii) to propose feasible rec- capability in line with Singapore’s white paper prediction of an influx ommendations to enhance diversity in the construction industry. The of foreigners. Thus, the growth potential and significance of diversity main goal of the study was to identify the potential benefits that a in the population encapsulate the labor force of Singapore (National more diverse workforce can bring to productivity in the context of the Population and Talent Division, 2013). construction industry, as well as to discover solutions to help better Many studies on workforce diversity have been conducted in integrate diversity into the labor force in construction firms. The ben- recent years, highlighting that workforce diversity could have a posi- efits and reasons for the solutions were analyzed and reviewed. As tive impact on productivity and workplace sustainability if it is man- workforce diversity concerning project productivity is underexplored aged and utilized well (Gladwin, Krause, & Kennelly, 1995; in Singapore’s construction industry, this paper aimed to draw conclu- Plummer, 2006; Saxena, 2014; Scholtens & Zhou, 2008). As such, the sions on and propose feasible solutions to this matter. construction industry also has a high level of workforce diversity, and To accomplish each objective, the study proposed the following taking full advantage of this workforce diversity could have a positive methods as shown in Figure 1: (i) a comprehensive literature review impact on project productivity (Loosemore, 2014; Shifnas and Sutha, was conducted to identify the crucial factors affecting workforce 2016). Therefore, by tapping into Singapore’s diverse labor force, diversity in Section 2, (ii) pilot interview and survey questionnaire there could be a potential impact on project productivity growth and development were performed for data collection, and structural equa- sustainable production in the construction industry. Through more tion modeling (SEM) method was proposed for the data analysis to skilled and diverse workers from various sectors, their experiences, assess the impact of workforce diversity on project productivity as culture, skills, and professional ambition across firms and industries described in Section 3, and (iii) post-interview with the analysis results could be transplanted into Singapore. To this end, it is imperative to was performed to discuss the results and propose the feasible strate- understand the relationship between workforce diversity and project gies to improve the workforce diversity and project productivity productivity correctly. However, studies on the relationships existing accordingly as described in Section 5 and 6. in Singapore’s construction industry are still insufficient. Also, not much attention has been given to how worker diversity may affect project productivity, although Singapore’s focus on 2 | LI T E RA T U R E RE V I E W increasing productivity mostly falls on skills training, implementing new construction technology, and workforce health and environment. In a country as diverse as Singapore, it should fully capitalize on its 2.1 | Productivity improvement initiatives in Singapore labor force diversity to boost productivity (Selvaraj, 2015). Although there are guidelines and specific laws in place to maintain age and The workforce in the construction industry is still developing, as the gender quotas in the workforce, these are mostly to prevent unfair issue lies with labor availability and training (Arditi & Mochtar, 2000). dismissal instead of looking at how they can increase productivity The study conducted by Singapore Contractors Association Ltd (Ministry of Manpower [MOM], 2011). There are also not enough (SCAL) (2016) stated that firms had identified the need for human studies on both Singapore’s and the construction industry’s context. resources with relevant skills and a lack of effort to manage the work- Given the weak performance regarding project productivity and force as reasons for issues in human resource utilization in the pro- Singapore’s position as a uniquely diverse society, there is a greater ductivity of the construction industry. In this context, initiatives to need better to understand the relationship between workforce diver- boost productivity come in the form of a construction productivity sity and productivity to develop proposals. This study thus aimed to roadmap. The main objectives pushed included points such as regulat- bridge the research gap and solidify a model of construction labor ing the demand and supply of construction, introduce more construc- diversity productivity that will enable industry stakeholders to under- tion technology, uphold standards when hiring labor, and further stand how labor diversity can cause productivity to change and to develop the skills of workers (Building and Construction Authority propose solutions by capitalizing on this model. Therefore, the results [BCA], 2015a). The other aim of the roadmap was to enable the 400 WON ET AL. Research Objectives Methods Research Outcomes To identify and assess critical factors affecting workforce diversity • Literature review Section 2 • Workforce diversity factors affecting workforce diversity To assess the impact of workforcediversity on project productivity • Data collection: Pilot interview and survey questionnaire development • Data analysis: Structural equation modelling Section 3 & 4 • Survey questionnaire • Analysis results To propose feasible recommendations to enhance diversity in the construction industry • Post interview with the analysis results Section 5 & 6 • Discussions of the results • Recommendations F I G U R E 1 Connections among research objectives, methods, and research outcomes industry to meet the national target of a 2–3% average annual growth educated, with the literacy rate also rising every year; for instance, by 2020. The areas of focus were a higher-quality workforce, higher 2017 recorded a 97.2% literacy rate compared to 97% in 2016 capital investment, and a better-integrated construction value chain. (Department of Statistics Singapore [SingStat], 2020). By encouraging In 2015, an additional S$450 million was provided to help firms invest more diversified education backgrounds among locals, the diversity of in “impactful productive technologies” and improve the quality of skills will only stand to increase in the future workforce. their workforce from 2015 to 2018 (BCA, 2015b). Building and Con- Diversity in age would similarly reflect the “Age and experience” struction Authority (BCA) is continuously pushing for the implementa- distribution in the workforce. Variances in workers’ age would gener- tion and advancement of construction technology, believing it to be ally indicate their level of working experience (Chung et al., 2015). the solution to increasing productivity (BCA, 2015a). Moreover, BCA Thus, the experience is a component of diversity under the age cate- developed a research and development plan for construction produc- gory. An increase in the 60 and above age group in the working popu- tivity, where the construction industry aimed to work on and improve lation reflects the data in 2017, where the median age of Singapore’s technologies adopted (BCA, 2016). These technologies will occur in population stood at 40.5 years (SingStat, 2017). An increasing median several research and development clusters with a future goal and tar- age potentially means an aging population and, in turn, a shrinking get in sight. Therefore, technology is seen as the solution to boost labor force. However, in response to its aging society, Singapore’s Singapore’s construction productivity, and there is less emphasis on retirement age is also increasing. Ministry of Manpower (MOM) reasons for the lagging productivity of the construction sector in com- announced that the proportion of workers aged 60 and above active parison to other sectors in Singapore. in the labor force had increased by 6.5% over 9 years from 2006. Otherwise, gender diversity in Singapore’s workforce has been steadily equal over the past few years, with the labor force gender ratio 2.2 | Workforce diversity in Singapore closely reflecting the gender ratio of the population. Although close to having an equal proportion of males and females in Singapore’s popu- A diverse workforce reflects the country’s community. With regard to lation, the percentage of females in the labor force was reported at labor diversity in the workforce, categories are often placed on groups 44.97% in 2017 (SingStat, 2017). of people for ease of categorization when clustering and segmenting Diversity initiatives typically mean the programs, policies, and them for differentiation’s sake. The following four types are com- strategies enrolled to promote diversity within companies. In Sin- monly used when addressing Singapore’s workforce diversity gapore, because the workforce is inherently diverse due to its (Selvaraj, 2015). Firstly, “ethnicity” diversity typically refers to race, multi-racial community, MOM has crafted a toolkit for organiza- but it goes beyond that to include other aspects such as religion and tions to help them better manage their workforce diversity spoken language. The ethnic breakdown of Singapore is estimated to (National Integration Working Group [NIWG], 2014). MOM pro- be 74% Chinese, 13% Malay, 9% Indian, and 3% others, while religious motes the idea that a diverse workforce, along with proper guid- groups equate to 43.2% Buddhists and Taoists, 18.8% Christians, ance, can help drive better business performance and higher 18.5% no religion, 14% Islam, 5% Hindus, and 0.6% others employee engagement. The MOM toolkit provides detailed strate- (MTI, 2018). “Skills and education” diversity refers to the qualifica- gies and reasons why firms should look toward adopting a more tions of the workforce required to enter the job market, as different inclusive workforce. As workforces in Singapore consist of various skill sets are certified to enroll in different jobs. The education level workers from different age groups, gender, nationalities, and eth- and skills also vary across different industries. An increase in skilled nicities who work together, the toolkit aims to help organizations labor with a rise in the percentage of higher education graduates in and managers understand the importance of workforce diversity Singapore from 2007 to 2017 symbolizes the country becoming more and to maximize its potential. 401 WON ET AL. 2.3 | Impact of workforce diversity on project productivity performance men are typically more logical, independent, and competitive, whereas women are known to be empathetic, dependent, and compromising (Jost & Kay, 2005). Therefore, men are expected to disagree more often There are several reasons why MOM is promoting a more diverse and show stronger task behaviors (Myaskovsky, Unikel, & Dew, 2005), workforce. Firstly, organizations that create a more inclusive environ- which can push the group to be more task-oriented. Women, on the ment for their diverse workers can enhance their company’s reputa- other hand, are perceived as more agreeable and supportive, and this tion in the job-seeking market to attract more workers. Secondly, can facilitate more teamwork and interaction. Therefore, when both employees from different backgrounds, who feel more integrated into genders are present in the workforce, their behaviors can be comple- the workforce, will be more engaged and motivated (Pollitt, 2005; mentary to support project tasks (Jost & Kay, 2005; Pucheta-Martínez, MOM, 2011). Lastly, higher employee engagement across all types of Olcina-Sempere, & López-Zamora, 2020). workers may lead to a reduced turnover rate (Dernovsek, 2008; In addition, a 2010 Singapore study on inclusive and harmonious MOM, 2011); a 2008 study by Gallup Management Group revealed workforces showed that 87% of companies surveyed agreed that that engaged employees had a 51% lower turnover on average workforce harmony was essential to business outcomes, and also (Dernovsek, 2008). Therefore, a more diverse workforce allows for demonstrated that a better-managed team with more diverse workers more cross-cultural interactions and a more socially inclusive space could outperform homogeneous teams with more creative solutions reflective of Singapore’s multiracial society, where employees stand to alongside more effective problem-solving methods, thus driving benefit from having good relationships among themselves and with workers to be more efficient and productive (NIWG, 2014). However, the firm. when diverse groups are not well managed, communication will be Concerning the impact that more workforce diversity has on PPP, scattered, and trust becomes weak, compromising the team each of the following diversities, “Ethnicity,” “Skills and education,” dynamics—showing how people management is essential to the “Age,” and “Gender,” are elaborated, based on past research and stud- impact of a diverse workforce on project productivity (Srikanth, ies conducted. Firstly, the exact impact and results of ethnic diversity Harvey, & Peterson, 2016). on firm performance is somewhat unclear due to conflicting theories and reasoning. On the one hand, ethnic-cultural diversity has been theorized to negatively affect firm performance as it may hinder the exchange of information and knowledge among workers due to lin- 2.4 | Identification of workforce diversity factors affecting project productivity performance (PPP) guistic and cultural barriers as well as weaken social ties and trust; people tend to doubt and be warier of people of different ethnicity, A comprehensive literature review was carried out to identify a set of and they prefer to interact with others belonging to the same enclave workforce diversity factors that affect PPP. For the first diversity type, as themselves (Glaeser & Vigdor, 2001). However, ethnic diversity can “Ethnicity,” diversity factors with negative impact were identified as benefit how well a firm performs as it can help facilitate communica- language barriers and lack of trust. Communication among different tions and decisions, and stimulate the brainstorming of new ideas ethnic groups, especially from different countries, often results in a (Hong & Page, 2004). It can also provide useful insights into the global language barrier that has the potential to negatively affect productiv- market and demands to increase a company’s competitiveness ity due to progress being slower from more rounds of clarification (Parrotta, Pozzoli, & Pytlikova, 2014), therefore reflecting inconclusive among workers. Also, because of the poor communication among dif- results on the opposite end of the spectrum. ferent ethnic groups, technical details and management instructions Diversity in “Skills and education” can trigger a knowledge stand to be misinterpreted; this can lead to different ethnic groups exchange among workers within a project group or the firm, which can not trusting each other, which will affect the coordination required for positively affect firm performance (Tsang, Rumberger, & Levin, 1991). In productivity (Makulsawatudom, Emsley, & Sinthawanarong, 2004). On contrast, diversity in “Age” can help stimulate firm productivity because the other hand, as for the diversity factors with a positive impact, PPP the interactions between workers of different age groups (young or old would depend on how much motivation and job commitment the workers), as well as the less and more experienced, can complement workers possess as increased morale can influence commitment to each other for fresh perspectives (Prieto, Phipps, & Osiri, 2009). On the the project and boost workforce motivation and, in turn, improve pro- one hand, younger workers would possess knowledge of new technolo- ductivity (Thomas & Sudhakumar, 2013). In addition, a group of peo- gies currently in use, while older workers have a clearer understanding ple with diverse ethnicity could boost productivity by contributing of the moving process and technical skills (Lazear, 2001). Besides, firms different effective work styles to the project, therefore allowing for a with mentorship programs also stand to gain more from having more more significant inflow of ideas, which would increase productivity diverse ages within the firm (Mor Barak, 2000). (Parrotta et al., 2014). Lastly, “Gender” diversity has been theorized to have potential For “Skill and education” type of diversity, the negative diversity benefits, such as where mixed-gender groups complete tasks quicker factors were identified as a lack of training provided, shortage of and make improved decisions (Ali, Kulik, & Metz, 2011; Sabatier, 2015), skilled labor, inspection delays, and incomplete technical specifica- thus potentially enhancing group performance. The general theory tions. A plethora of unskilled workers have come to the construction states that men and women partake in different societal roles, where industry in developed countries from developing countries 402 WON ET AL. (Abdul-Rahman, Wang, Wood, & Low, 2012; Kaming, Olomolaiye, workforce of different ages can result in higher productivity due to Holt, & Harris, 1997; Khadria, 2006). Due to unskilled and unqualified the incentive gained. With the proper positioning of experienced workers, inspection delays and incomplete technical specifications supervisors, productivity would increase as supervisors have to be occur and eventually decrease productivity. However, contractors assigned to positions and places where they can properly utilize and usually invest less in labor training as they are dissuaded by the short- transfer knowledge gained from experience and translate that into term costs incurred, which result in potential long-term benefits not productivity on the job (Lim & Alum, 1995; Thomas, 2015). As the being reaped, rendering construction firms with little option but to construction industry is becoming mechanized with the introduction hire less-skilled, less-qualified, and less-trained workers to save costs, of new equipment and technology, it is paramount to recruit and train which can severely affect construction productivity (Lim & the younger generation of skilled local workers who are adept at using Alum, 1995). Also, an unskilled supervisor in the workforce can lead technology in comparison to their older but more experienced to an incompetent leader, resulting in unproductive activities, such as counterparts (Lim & Alum, 1995). delayed inspections, more mediocre quality work produced, and Regarding gender diversity, the diversity factors negatively affect- increased idle time of resources (Dai, Goodrum, & Maloney, 2007). ing productivity mainly include the physicality and health of workers. Therefore, without a skilled and trained supervisor and workforce, The physical aspect of construction consists of the time and workload incomplete, unclear, or outdated technical specifications will also of the construction site (Soham & Rajiv, 2013). As men are more likely occur due to inadequate site management, resulting in multiple accustomed to hard manual labor, a more physical job is usually requests for clarification meetings, leading to more interruptions to undertaken by men for increased productivity. Taking the health of the work progress (Jarkas & Bitar, 2012). workers into consideration as a factor affecting productivity, for Otherwise, in the “Skill and education” diversity type, transfer of maternity leave, women in Singapore are entitled to 16 weeks of paid skills and knowledge, effective project planning and execution, quali- leave. In the context of Singapore’s construction industry, workers’ fied supervisors, and efficient decision-making were identified as the health may affect productivity as cross-training and handing over as positive diversity factors in “Skill and education.” The transfer of well as hiring temporary replacements all consume time and knowledge and skills would help generate knowledge spillover among resources, which can impact project productivity. However, with a a firm’s employees, as long as workers’ knowledge sets do not overlap more diverse distribution of gender in the workforce, productivity and are relevant to one another, which positively affects firm perfor- may stand to gain from the contrasting behavior and work pattern of mance (Parrotta et al., 2014). With more relevant skills and training men and women as they have different sets of social behaviors that for construction projects, qualified supervisors can make quicker and could be complementary (Jost & Kay, 2005; Pucheta-Martínez more effective decisions to prevent wastage of time and, thus, et al., 2020). increase productivity on site (Jarkas & Radosavljevic, 2013; Kazaz, Manisali, & Ulubeyli, 2008). With more highly qualified and trained workers, there can be more effective project planning and execution 3 M E T H O D S A N D D A T A C O LL E C T I O N | occurring on-site for construction projects, hence resulting in a positive boost to project productivity through a more skilled and educated 3.1 | Survey design workforce. “Age and experience” usually encompasses work experience as a Before conducting a wide-ranging survey, a pilot interview was con- factor concerning productivity. With a negative impact of “Age and ducted with three industry experts to validate the identified factors experience,” shortage of experienced labor, high rate of labor turn- from the literature review as well as to certify the applicability and over, and unrealistic deadlines for project completion were identified understandability of the questionnaire. This interview helped to elimi- as diversity factors. Since the labor-intensive construction industry nate insignificant and repetitive diversity factors identified concerning strongly relies on and demands excellent skills and experience of the their impact on the productivity of construction projects. During the workforce, a lack of experienced labor can severely affect project interview, the questions were posed to the respondents to list addi- deadlines, cost, and quality of works done (Alinaitwe, Mwakali, & tional factors that may not have been identified through the literature Hansson, 2007). Unlike older workers who stay in the construction review. Targeted respondents of this pilot interview were industrial industry in which they accumulated their experience, a younger labor experts who were chosen due to their relevance and involvement in force results in a higher turnover rate in search of new jobs in differ- the construction industry, who have worked closely with various asso- ent sectors (Thomas, 2015). Such a volatile workforce with more ciates of different diversities. Therefore, their ability to provide valu- youthful workers may result in lower productivity (Bandhanpreet, able insights and views were taken into consideration to develop the Mohindru, & Pankaj, 2013). final survey questionnaire. Finally, as shown in Table 1, a total of Conversely, for the “Age and experience” diversity factors with a positive impact, wages have to be attractive to recruit incoming 21 diversity factors categorized into four principal diversity types were included in a survey questionnaire. workers as the construction sector has to compete with other indus- The survey questionnaire was designed based on workforce tries for its workforce, such as the engineering market, banking, and diversity factors. The questionnaire consists of three main sections, as finance (Thomas, 2015). Therefore, using wages to attract and retain a shown in Appendix S1. The first two sections aimed to solicit Shortage of skilled labor Inspection delays Incomplete technical specifications Transfer of skills and knowledge Effective project planning and execution Qualified supervisors 7 8 9 10 11 Contribution of different work ethic 4 6 Motivation and job commitment 3 Lack of training provided Lack of trust between groups 2 5 Language barrier 1 Ethnicity Skill and education No. Diversity factors With more relevant skill and training for construction projects, the qualified supervisors can make quicker Through a more skilled and educated workforce, there can be more effective project planning and execution occurring on-site for the construction projects, resulting in a positive boost to labor productivity The transfer of knowledge and skill would help generate knowledge spillover among the employees within a firm, as long as the sets of workers knowledge do not overlap and are relevant to one another, which positively affects firm performance Incomplete technical specifications are usually caused by the lack of skill and education. Thus, it can be included in the workforce diversity factor as a mediator affecting project productivity One of the reasons for inspection delay is the lack of skill and education, and therefore it was included in the workforce diversity factor as a mediator affecting project productivity accordingly Unskilled labor in the workplace would lead to unproductive activities such as poorer quality work done and increased idle time of resources. Due to unskilled and poorly trained workers, productivity decreases as faulty works will eventually result in more cost spent on maintenance and corrections A group of people with diverse ethnicity can boost productivity by contributing different effective work styles in the project, therefore allowing for more inflow of ideas The project productivity would also depend on how much motivation and job commitment the workers possess. As increased morale can influence the commitment to the project and boost the motivation of the workforce and in turn increase productivity Because of lousy communication, technical details and management instructions stand to be misinterpreted which can lead to the different ethnic groups not trusting each other which will affect the coordination required for productivity Communication between different ethnic groups, especially from different countries, often result in a language barrier, which has the potential to negatively affect productivity due to progress being slower from more rounds of clarification between workers Definition ● ● ● ● ● Kazaz et al., Makulsawatudom 2008; Jarkas, Thomas & et al., 2004; Jarkas Radosavljevic, Sudhakumar, & Radosavljevic, 2013 & Wuyi, 2014 2013 Diversity factors affecting labor productivity Diversity type TABLE 1 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● Enshassi, Jarkas & Lim & Mohamed, Makulsawatudom Parrotta Radosavljevic, El-Gohary & Jarkas & Jarkas, Alum, Thomas, Bandhanpreet Dai et al., Mustafa, & et al., et al., 2014 2013 Aziz, 2013 Bitar, 2012 2015 1995 2015 et al., 2013 2007 Mayer, 2007 2004 (Continues) Mojahed & Soham & Jost & Aghazadeh, Rajiv, Kay, 2008 2013 2005 WON ET AL. 403 Gender Age and experience Positioning of experienced consultants More dependence on equipment and technology 17 18 Physical fatigue Amount of pay and wages 16 19 If there is a lack of diverse labor with little experienced project managers, there may also be negative impacts on productivity, such as milestones and deadlines that are unrealistic. If the deadline is not possible to meet, it would place production pressure on the workers and cause a negative impact on the project productivity Unrealistic deadlines for project completion 15 Where the physical aspect of construction includes the time and workload of the construction site which men are more accustomed to hard As the construction industry is becoming more mechanized with the introduction of more new equipment and technology, it is paramount to recruit and train a younger generation of skilled local workers who are adept at the usage of technology in comparison to the older but more experience counterparts The proper positioning of experienced supervisors can positively affect productivity as they have to be assigned to positions and places where they can properly utilize and transfer knowledge gained from experience and translate that into productivity on the job The amount of wages has to also be attractive in order the recruit the incoming workers as the construction sector has to compete for workforce with other more attractive industries such as the engineering market or banking and finance. Therefore, by using wages to attract and retain the workforce of different ages, it can result in higher productivity due to the incentive gained High rate of labor Unlike older workers who stay in turnover the construction industry in which they accumulate experience in, a younger labor force projects a higher turnover rate in search of new jobs in different sectors. Such a volatile workforce with more youthful workers may result in a lower productivity 14 Where since the labor-intensive construction industry strongly relies on and demands good skills and experience of the workforce. Therefore, a lack of experienced labor can severely affect the project deadlines, cost, and quality of works done Shortage of experienced labor With more highly qualified and trained workers more efficient decision-making can be occurred on site for the construction projects, resulting in a positive boost to labor productivity and more effective decisions to prevent wastage of time and thus, increase productivity on site Definition 13 Efficient decisionmaking No. Diversity factors Diversity type 12 (Continued) TABLE 1 ● Kazaz et al., Makulsawatudom 2008; Jarkas, Thomas & et al., 2004; Jarkas Radosavljevic, Sudhakumar, & Radosavljevic, 2013 & Wuyi, 2014 2013 ● ● ● ● ● ● ● Enshassi, Jarkas & Lim & Mohamed, Makulsawatudom Parrotta Radosavljevic, El-Gohary & Jarkas & Jarkas, Alum, Thomas, Bandhanpreet Dai et al., Mustafa, & et al., et al., 2014 2013 Aziz, 2013 Bitar, 2012 2015 1995 2015 et al., 2013 2007 Mayer, 2007 2004 ● ● ● ● Mojahed & Soham & Jost & Aghazadeh, Rajiv, Kay, 2008 2013 2005 404 WON ET AL. 405 WON ET AL. information regarding the respondent’s organization and their per● Mojahed & Soham & Jost & Aghazadeh, Rajiv, Kay, 2008 2013 2005 sonal work experience in different construction project categories to ● The last part of the questionnaire required the respondents to indicate their perception of the significance and PPP of all the diversity factors on projects based on their experience. The significance Enshassi, Jarkas & Lim & Mohamed, Makulsawatudom Parrotta Radosavljevic, El-Gohary & Jarkas & Jarkas, Alum, Thomas, Bandhanpreet Dai et al., Mustafa, & et al., et al., 2014 2013 Aziz, 2013 Bitar, 2012 2015 1995 2015 et al., 2013 2007 Mayer, 2007 2004 referred to how they perceived the level of impact that the respective diversity factor had on the project, which was measured using a fivepoint Likert scale: 1 = not significant, 2 = a little significant, 3 = uncertain and neutral, 4 = moderately significant, and 5 = very significant. The PPP referred to how respondents perceived the diversity factor to affect the project’s level of productivity, which was also measured using a five-point Likert scale: 1 = very bad, 2 = bad, 3 = neutral, 4 = good, and 5 = very good. 3.2 construction industry, such as developers, contractors, architects, or quantity surveyors who had experience in construction projects. They were selected in the belief that they could provide accurate and constructive responses on the workforce diversity factors and their significance and impact on PPP. Firms of the targeted respondents were sourced through BCA’s Directory of Registered Contractors and Licensed Builders, which includes global companies working in Singapore, such as the list of architectural consultancy firms and registered contractors. Simple random sampling was conducted to select 140 companies from the BCA’s directory, and the questionnaires were distributed via e-mail. Among the selected companies, 28 survey responses were collected. In addition, the other 36 questionnaires were printed and distributed to relevant personnel in the industry, and 30 responses were received from them. All in all, the questionnaire achieved a 33% response rate, with a total of 58 complete sets of surveys. Four types of organizations participated in the survey. The respondents that completed the survey included 26 contractors, 3 developers, 23 consultants, and 6 architects. Table 2 summarizes the respondents’ respective job positions in the firms as well as their relevant years of experience in the construction industry. As seen in Table 2, the majority of the respondents had at least 5 years of experiWith a more diverse distribution of genders in the workplace, productivity may stand to gain from the contrasting behavior and work pattern of men and women as they have different sets of social behaviors, which can be complementary at least 3 years of experience. Only one respondent had less than a Complementary social behaviors (56.5%), 42 infrastructure (27.3%), 11 heavy industrial (7.1%), and 21 20 Health of workers Especially in the context of maternity leave where women in Singapore are entitled to 16 weeks of paid leave may affect the productivity in the workplace as cross-training and handing over as well as hiring temporary replacement all consume time and resources which can impact the productivity of projects ence in the construction industry, with the next highest group having manual labor, therefore, a more physical job is usually undertaken by men for increased productivity Definition Data collection The target population of the study belonged to firms involved in the Kazaz et al., Makulsawatudom 2008; Jarkas, Thomas & et al., 2004; Jarkas Radosavljevic, Sudhakumar, & Radosavljevic, 2013 & Wuyi, 2014 2013 No. Diversity factors | year of experience, while five respondents had more than 10 years of experience in the industry. The data collected also reflected the various project characteristics of the projects managed and participated in by the respondents, as shown in Table 3. There were a total of 154 projects: 87 building 14 light industrial (9.1%) projects. A large percentage of the projects involved new construction projects (71.4%), while the remaining 28.6% of projects were addition and alteration projects. Similarly, the majority (38.3%) of the projects cost between S$10 million and S$100 Diversity type (Continued) TABLE 1 validate the survey, targeting respondents using their qualifications. million. 406 WON ET AL. TABLE 2 Survey respondents’ job positions and years of experience Respondent characteristics Job position Number Percentage (%) Project manager 15 25.9 Architects 8 13.8 Quantity surveyors 12 20.7 Consultant 9 15.5 Contractor Total Years of experience 24.1 100 Less than 1 year 1 1.7 1–2 years 8 13.8 3–4 years 20 34.5 5–10 years 24 41.4 More than 10 years 5 8.6 Total 58 Project characteristic Project type Number Percentage (%) Building 87 56.5 Infrastructure 42 27.3 Heavy industrial 11 7.1 Light industrial 14 9.1 154 100 New construction 110 71.4 Total Project nature Addition and alteration 44 28.6 154 100 Less than 1 million SGD 32 20.8 1–10 million SGD 42 27.3 10–100 million SGD 59 38.3 100 million–1 billion SGD 21 13.6 154 100 Total Project cost Total 3.3 14 58 | Data analysis methods 100 T A B L E 3 Project characteristics undertaken by survey respondents through survey questionnaires, a hypothesized model of diversity factor significance and its perceived productivity performance was 3.3.1 | Partial least squares structural equation modeling established, assuming there was a positive relationship between The Shapiro–Wilk test (Shapiro & Wilk, 1965) was firstly used to test ables using the PLS method, the model was divided into a structural for the normality of the dataset collected from the questionnaires. model, a measurement model, and a weight relation that represented According to the result of the Shapiro–Wilk test, the method to be latent variables as linear combinations of measurement variables used for analysis was determined between parametric and nonpara- (Fornell & Cha, 1994). The structural model represented the causal metric methods. In this study, diversity factors identified from the lit- relationship between latent variables, including error variables that erature review were assessed by industry experts through survey were not explained through the model by following the basic matrix questionnaires. The survey asked respondents to indicate their gen- equation: the two. In order to analyze a structural equation model with latent vari- eral perception of the significance of each diversity factor and its impact on productivity performance based on their project experi- η = Β η + Γ ξ + ζ, ð1Þ ences using a five-point Likert scale. A structural equation model, partial least squares structural equation modeling (PLS-SEM), was where each η, ξ, and ζ represent the matrixes for endogenous latent subsequently applied. The PLS-SEM aimed to test the proposed variables, exogenous latent variables, and error variable; Β and Γ are hypothesis, which, in this study, represented the relationship between the matrixes of path coefficients between latent variables. On the diversity factors and perceived PPP. Based on the data collected other hand, the measurement model represented the causal 407 WON ET AL. relationship between the latent variable and the measured variable, applied in this study. Based on the obtained weights, the weighted including the measurement error occurring during the measurement sum of the latent variables is re-estimated in the model, and the process by following the basic matrix equation: values of the latent variables are renormalized for the next step. In the external approximation, the regression coefficients are esti- y = Λy η + ε, ð2Þ x = Λx ξ + δ, ð3Þ mated by simple or multiple regression analysis using normalized latent and measurement variables, and the estimated latent variables are reestimated using the weighted sum of the regression coefficients of measured variables using Equations (4) and (5). The internal and exter- where y and x are the matrixes of measurement variables η and ξ, Λy nal approximations are repeated using the estimated latent variables. and Λx are coefficient matrixes of the measurement model, and ε and This iterative process is repeated until the difference between the value δ are the matrixes of error. The weighting relationship in the PLS esti- of each weight in the external approximation becomes less than 10−3 mation process required calculating the predictive value of the latent variables, η^ and ξ^ using a linear combination of the measured variables, (Chin, 1998). After estimating the latent variables using the final weight defined in the following equations: the measurement and structural models are estimated by simple or mul- obtained through the iterative process, all the path coefficients of both tiple regression analysis using the estimated values. η^ = wη y, ð4Þ After the development of the structural equation model, evaluation of the model needs to be performed to ensure confidence that the ξ^ = wξ x, ð5Þ measures represent the construct of interest, hence adequately using them to examine the structural relationship. The model evaluation where wη and wξ are the matrixes of weights needed to compute the included the assessment of model measures with regard to four predictive value of a latent variable. aspects: (i) internal consistency reliability, (ii) indicator reliability, The PLS method was chosen for analyzing the structural equation (iii) convergent validity, and (iv) discriminant validity. The evaluation was model due to its propriety for non-distributional form presumed in the processed by applying four kinds of validity following certain criteria measured variables (Aibinu & Al-Lawati, 2010), which is applicable to with the statistical indicators that were generated in the SmartPLS soft- the dataset of this study as determined by the Shapiro–Wilk test. Fur- ware: composite reliability, indicator loadings, and average variance thermore, PLS-SEM works well and has high statistical power with a extracted (AVE). For internal consistency reliability to be regarded as smaller sample size (Fornell & Bookstein, 1982; Hair, Hollingsworth, satisfactory, composite reliability values should be higher than 0.70 Randolph, & Chong, 2017; Lim, Ling, Ibbs, Raphael, & Ofori, 2012; (Hair, Ringle, & Sarstedt, 2011). Indicator reliability requires values of Raymond & Bergeron, 2008). PLS is a method of repeating a regres- indicator loadings to be at least higher than 0.40, and, for the conver- sion analysis to find the optimized parameter combination that mini- gent validity assessment, AVE values should be higher than 0.50 (Hair mizes the prediction error by using the parameters whose values are et al., 2011). Lastly, discriminant validity was assessed by two measures: known or fixed by specific conditions. The iterative process can be cross-loadings, where an indicator’s loading should be higher than all of primarily divided into the external approximation process that calcu- its cross-loadings; and the Fornell–Larcker criterion, where the AVE of lates the value of the latent variable using the measured variables and each latent construct should be higher than the construct’s highest internal approximation process that calculates the value of the latent squared correlation with any other construct (Hair et al., 2011). variable using the relationship between the latent variables (Fornell & Cha, 1994). After ensuring the evaluation of the model, the significance of path coefficients was generated by bootstrapping, which is a statisti- In the internal approximation, the value of the latent variable is cal method for the inference of a population using sample data (Hair estimated based on its initial weight, which is commonly initialized as et al., 2011). The critical t-value for a two-tailed test is 2.58 at a signif- one, and its normalized measurement variables, which represent the icance level of 1% (Hair et al., 2011). A t-value of below 2.58 accepts attribute of the latent variable. After standardizing the estimated the null hypothesis, concluding in this case that workforce diversity latent variable values, the weight is calculated using one of the weight does not have a significant impact on PPP. The various indicator load- selection methods (Fornell & Cha, 1994) defined according to the ings of each management diversity factor define how impactful they causal relationship of latent variables. There are three main methods are, with higher loadings deemed as having greater effectiveness for weight selection: (i) the path-weighting scheme, in which weights within the category. are differentiated by distinguishing the roles of causes and effects among potential variables, (ii) the factor-weighting scheme that uses the correlation coefficients of two variables as weights without dis- 3.3.2 | Post-interviews tinguishing roles between potential variables, and (iii) the centroidweighting scheme that uses the sign of the correlation coefficient In order to fine-tune the recommendations, a follow-up interview was between the latent variable to be found and the latent variable affect- conducted with the selected specialists listed in Table 4. The post- ing the variable as a weight. The path-weighting method is most com- interviews were performed with the three professionals in both face- mon and generally utilized among the above methods, which was to-face and untact manners. The professionals’ information is 408 WON ET AL. TABLE 4 Profile of pilot and post-interviewees Respondents Company type Years of experience in the industry Job position A Developer firm 12 Project director B Construction company 9.5 Project manager C Construction company 4 Project executive Significance Productivity performance Diversity factors Statistic DoF Significance Statistic DoF Significance F1 0.790 10 0.011 0.844 10 0.049 F2 0.830 10 0.033 0.814 10 0.022 F12 0.704 10 0.001 0.744 10 0.003 F13 0.783 10 0.009 0.654 10 0.000 F3 0.811 10 0.020 0.746 10 0.003 F4 0.828 10 0.031 0.690 10 0.001 F5 0.661 10 0.000 0.776 10 0.007 F6 0.835 10 0.039 0.696 10 0.001 F14 0.797 10 0.013 0.746 10 0.003 F15 0.729 10 0.002 0.766 10 0.006 F16 0.830 10 0.033 0.842 10 0.046 F17 0.820 10 0.025 0.781 10 0.009 F7 0.785 10 0.010 0.671 10 0.000 F8 0.812 10 0.020 0.792 10 0.012 F9 0.778 10 0.008 0.825 10 0.029 F18 0.725 10 0.002 0.805 10 0.017 F19 0.826 10 0.030 0.842 10 0.046 F20 0.776 10 0.007 0.530 10 0.000 F10 0.830 10 0.034 0.744 10 0.003 F11 0.844 10 0.049 0.823 10 0.028 F21 0.812 10 0.020 0.694 10 0.001 TABLE 5 Normality test results described in Table 4, which include one project director from a devel- H0. The data are normally distributed. opment firm as well as two project managers and a project executive H1. The data are not normally distributed. from construction companies. Their feedbacks were carefully consid- Table 5 shows the results of the Shapiro–Wilk test for the data ered when analyzing the results and devising the strategies. More for the perceived significance and productivity performance of the details on the outcomes from the post-interviews are provided in Sec- 21 diversity factors. The p values of all the diversity indexes reflected tions 5 and 6. a value lower than 0.05, thus rejecting the null hypothesis and concluding that the dataset does not fit a normal distribution. As the dataset is not of a normal distribution, a nonparametric statistical test, 4 4.1 DATA ANALYSIS RESULTS | | the PLS-SEM, was then adopted for analysis in this research. Normality test In order to determine the suitable test methods for the data, the 4.2 | Deriving the relationship between workforce diversity and project productivity via PLS-SEM Shapiro–Wilk test was conducted. A significance level (α) of 0.05 was used to assess the hypothesis. If the p value falls below 0.05, the null The study mainly aimed to investigate which workforce diversity fac- hypothesis will thus be rejected, concluding that the dataset does not tor was significantly correlated with construction PPP. In order to fit a normal distribution. The null hypothesis and alternative hypothe- derive the relationship, the SEM technique PLS-SEM was applied. This sis are established below: section presents the results of the modeling and an in-depth 409 WON ET AL. TABLE 6 Measurement models Diversity type Ethnicity Skill and education Age and experience Gender Code Loading t-value AVE CR 0.561 0.864 0.514 0.892 0.583 0.847 0.665 0.854 E1 0.713 24.219 E2 0.759 32.929 E3 0.776 44.556 E4 0.822 47.182 SE5 0.746 28.207 SE6 0.685 22.441 SE7 0.647 23.257 SE8 0.443 10.671 SE9 0.695 28.713 SE10 0.721 33.884 SE11 0.816 44.025 SE12 0.899 92.952 AE13 0.807 42.139 AE14 0.697 24.014 AE15 0.635 20.369 AE16 0.550 13.493 AE17 0.695 22.554 AE18 0.755 37.118 G19 0.849 39.121 G20 0.921 145.241 G21 0.654 18.636 Note: E = ethnic, SE = skills and education, AE = age and experience, G = gender. discussion on which diversity factors are most effective and which satisfactory diversity aspects need to be enhanced. The hypothesis for the model (CR) (Equation [6]) values shown in Table 6 are all higher than 0.7, development was proposed as follows: implying satisfactory internal consistency reliability (Nunnally, Bern- indicator reliability. The composite reliability stein, & Berge, 1967). On the other hand, for the validity of the model, H0. Workforce diversity does not have a significant impact on PPP. several measures needed to be evaluated, such as AVE, cross-loading, and the Fornell–Larcker criterion (Hair et al., 2011). Firstly, all the H1. Workforce diversity has a significant impact on PPP. values of AVE (Equation [7]) presented in Table 6 are all more than 0.5, thus indicating a sufficient degree of convergent validity The significance of the path coefficient was generated using bootstrapping. If the t-value is below 2.58, the null hypothesis is accepted, concluding that workforce diversity does not have a significant impact on PPP. Conversely, the null hypothesis is rejected if the t-value falls (Bagozzi & Yi, 1988; Fornell & Larcker, 1981): P ð λi Þ2 Composite Reliability ðCRÞ = P 2 P , ð λi Þ + 1− λi 2 higher than 2.58. With the proposed hypothesis, a structural equation model was developed via SmartPLS software. Then, the developed model was evaluated through measurement model analysis, and the Average Variance Extracted ðAVEÞ = P P 2 λi , P 1− λi 2 λi 2 + ð6Þ ð7Þ model was analyzed and discussed to suggest recommendations to improve workforce diversity and construction PPP. where λ is the factor loading of the i-th measurement variable. Also, two measures, cross-loading, and the Fornell–Larcker criterion (Fornell & Larcker, 1981), needed to be assessed for discriminant 4.2.1 | Measurement model analysis validity to determine which factors are distinct and uncorrelated. Table 7 shows that each factor’s loading (bolded in the table) is higher The evaluation of the model reliability involved separate assessments than all of the other cross-loadings for each category: ethnicity (E), of the measures from the measurement model (Hair et al., 2011), as skills and education (SE), age and experience (AE), and gender (G); for presented in Table 6. If the indicator loadings for each diversity factor example, the loading of E1, 0.713, is the highest value compared to are higher than 0.4, and if t-values are higher than 2.58, this indicates other categories such as SE, AE, and G. Table 8 presents each 410 WON ET AL. TABLE 7 Cross-loadings for discriminant validity substantiate the null hypothesis, as presented in Figure 2. The effects of each diversity factor in relation to productivity performance are Factor category Factor E SE AE G deduced from the factor loadings generated by the bootstrapping, E1 0.713 0.557 0.508 0.267 with higher loadings having a more substantial impact. Diversity fac- E2 0.759 0.605 0.588 0.652 tors were grouped into four categories, with a total of 21 factors mea- E3 0.776 0.617 0.614 0.416 E4 0.822 0.616 0.600 0.381 SE5 0.600 0.746 0.565 0.323 SE6 0.453 0.685 0.553 0.260 SE7 0.606 0.647 0.604 0.681 SE8 0.506 0.742 0.382 0.566 model showed the path coefficient and t-values. The t-values, 12.697 SE9 0.504 0.695 0.604 0.294 for “Skill and education” and 10.947 for “Age and experience” with SE10 0.613 0.721 0.604 0.532 “Productivity performance” (as shown in Figure 2), are both higher SE11 0.613 0.816 0.565 0.562 than 2.58, indicating its statistical significance at the 0.01 level (Hair SE12 0.520 0.899 0.561 0.596 et al., 2011). Therefore, the null hypothesis proposed is rejected, con- AE13 0.530 0.558 0.807 0.497 cluding that the two categories of workforce diversity have a signifi- AE14 0.604 0.534 0.697 0.328 cant impact on PPP in the construction industry. AE15 0.439 0.524 0.735 0.242 “Skills and education” was the top set of diversity factors with a path coefficient of 0.552. “Age and experience” had the second- AE16 0.509 0.571 0.650 0.637 AE17 0.547 0.557 0.695 0.445 AE18 0.541 0.509 0.755 0.593 G19 0.338 0.455 0.512 0.849 G20 0.552 0.577 0.470 0.921 G21 0.493 0.443 0.385 0.754 Note: E = Ethnic, SE = Skills and education, AE = Age and experience, G = Gender. sured and tested for their correlation to perceived productivity performance. This was achieved by computing the significance and productivity performance indicated by all respondents. A path analysis between the four diversity categories and productivity performance of the factors was implemented. The outer model showed the loadings and t-values for the respective diversity factors, while the inner highest path coefficient of 0.359, and “Ethnicity” was the third with a path coefficient of 0.046. Lastly, “Gender” ranked fourth with a path coefficient of 0.031. “Skills and education” had the highest path coefficient among the four diversity categories, which implies a stronger influence on perceived productivity performance than the rest, followed by “Age and experience” also having a strong influence. As for the other two diversity categories, “Ethnicity” and “Gender,” they have a small path coefficient of less than 0.100. A discussion on the various diversity factors in the two significant categories (“Skill and TABLE 8 education” and “Age and experience”) and their corresponding effect Fornell–Larcker criterion E are discussed in the following section. Category AE G AE 0.723 E 0.585 0.747 G 0.661 0.567 0.841 SE 0.695 0.649 0.662 SE 5 0.744 Note: E = Ethnic, SE = Skills and education, AE = Age and experience, G = Gender. DISCUSSION OF THE RESULTS | The structural model presents the impacts of the diversity factors on PPP as perceived by the respondents. Two out of the four diversity categories, namely “Skill and education” and “Age and experience,” were shown to have significant effects on the productivity of construction projects, while the other two categories, “Ethnicity” and construct’s AVE, bolded figures in the table higher than the squared “Gender,” had statistically insignificant and small correlations to PPP correlation with any other construct, which satisfies the Fornell– due to the t-values being less than 2.58. Table 9 summarizes the rank- Larcker criterion; for example, for the AVE value for AE to AE, 0.723 ings of the diversity factors belonging to the two significant catego- is the highest value compared to other categories such as E, G, and ries. The categories’ assorted diversity factors are discussed based on SE. Both the above measurements indicate that the factors rightfully the descending order of their loadings as follows. relate more strongly to their diversity category than to another category, which satisfied discriminant validity. 5.1 4.2.2 | Structural equation model analysis 5.1.1 Skill and education | | SE12: Efficient decision-making The structural model showed a t-value higher than 2.58 between two Efficient decision-making received the highest loading in this category. diversity categories and their perceived productivity performance to Decisions are usually made by an executive who should know how to 411 WON ET AL. Outer model Inner model Productivity performance Loading t-value 0.713 23.159 E1 0.759 32.491 E2 0.776 44.063 E3 0.822 50.274 E4 t-value Loading t-value Loading t-value SE5 0.746 28.212 0.046 1.347 0.552 12.697* SE6 0.685 24.016 SE7 0.647 23.309 SE8 0.442 10.438 SE9 0.695 29.385 Skill and education Ethnic Loading t-value Loading t-value SE10 0.721 34.405 0.031 1.565 0.359 10.497* SE11 0.816 44.308 SE12 0.899 91.064 Age and experience Gender FIGURE 2 Loading G19 G20 G21 Loading 0.849 0.921 0.654 t-value 36.553 150.224 19.414 AE13 AE14 AE15 AE16 AE17 AE18 Loading 0.807 0.697 0.635 0.550 0.695 0.755 t-value 41.887 24.127 20.113 13.153 22.304 37.798 Best-fitted structural equation model T A B L E 9 Summary of the effectiveness of diversity factors by category Code Diversity factors Loading Rank Skill and education (SE) SE12 Efficient decision-making 0.899 1 SE11 Qualified supervisors 0.816 2 SE5 Lack of training provided 0.746 3 SE10 Effective project planning and execution 0.721 4 SE9 Transfer of skills and knowledge 0.695 5 SE6 Shortage of skilled labor 0.685 6 SE7 Inspection delays 0.647 7 SE8 Incomplete technical specifications 0.442 8 Age and experience (AE) AE13 Shortage of experienced labor 0.807 1 AE18 More dependence on equipment and technology 0.755 2 AE14 High rate of labor turnover 0.697 3 AE17 Positioning of experienced consultants 0.695 4 AE15 Unrealistic deadlines for project completion 0.635 5 AE16 Amount of pay and wages 0.550 6 make wise and logical decisions based on principles and how to weigh decision-making process. Therefore, a determination is valid only when the pros and cons. This can be attributed to the vital role that efficient the initial steps are taken to ensure that the process is streamlined, and decision-making plays throughout the entire project life cycle, where the respective superiors are on the same page for them to quickly give the post-interviewees have emphasized how meetings are often del- approval. The decision-maker (which in the context of the construction ayed due to a conclusion not being made. The most time-consuming industry is a project manager) must also consider the worker capabilities step in the whole process is not merely making the decision but putting that the decision will affect (Thorngate, 1980). Therefore, a single deci- it into effect (Thorngate, 1980). Thus, effectively making a decision and sion can have a long-term impact on the construction project and has swiftly translating it into action are highly essential steps in the whole the potential to affect productivity levels. 412 5.1.2 WON ET AL. | SE11: Qualified supervisors precise directions for the usage of resources, such as people, money, time, or raw materials, can drive higher effectiveness and productivity Having a qualified leader strongly relates to the quality of the decisions (Mason, 2000). Therefore, in order to ensure the project flows being made. A more qualified supervisor has the capability to make more smoothly, without facing a shortage of resources and causing delays, effective decisions. A qualified supervisor, in turn, would also step into companies have to ensure that planning occurs before project execu- more leadership roles with the opportunity to lead various groups of pro- tion so that the product can be delivered on time and schedule. ject teams. The constant push for higher levels of productivity requires more effective and ethical leadership (Mason, 2000). Competent supervisors with the necessary leadership skills are crucial in the organizational 5.1.5 | SE9: Transfer of skills and knowledge hierarchy as processes and decisions have to be run by them before being executed (Lee, 2007; Russell & Petrie, 1994). In general, supervi- The transfer of learned skills and knowledge is crucial to develop com- sion is about establishing relationships in the workforce to work with petencies in the respective fields. This is especially essential in a and guide people in new directions (Mason, 2000). In addition, the post- project-based environment where workers need to have positive inter- interviewees affirmed the positive impact of qualified supervisors as they actions and efficiency in order to meet milestones, as skill transfer helps are needed for quality control of the project and a stringent reporting create a more dynamic learning and working environment where inter- structure of project task teams. More qualified supervisors can influence actions enable workers to learn more from each other (Parrotta the people that work under them as they heed the instructions given by et al., 2014). By engaging in the transfer of skills and knowledge, the supervisor so that they can finish their goal on target (Hui, Chiu, Yu, workers can improve their capabilities. Therefore, a communicative pro- Cheng, & Tse, 2007)—ultimately improving project productivity. How- ject environment promotes active and sustainable collaboration in the ever, the post-interviewees also mentioned that smaller projects require workforce, which is optimum for transferring skills and knowledge for less manpower; therefore, with more qualified supervisors managing the self- and group improvement and improves the project productivity staff, the effect of the supervisor’s good management skills is more accordingly. However, a post-interviewee mentioned with a different noticeable among the small workforce in contrast to among a larger perspective that construction projects are temporary-based where workforce where the impact is not so perceptive (Disselkamp, 2013). teams assembled only interact as long as the duration of the project, and, therefore, the transfer of skills and knowledge have lower significance in the construction industry in comparison to the manufacturing 5.1.3 | SE5: Lack of training provided industry, which is more continuous based on factory and products. A lack of training can negatively impact workforce safety and the health of the construction workforce. Thus, workers need to receive adequate 5.1.6 | SE6: Shortage of skilled labor health and safety training. Also, poorly trained employees will lead to more mediocre performance in the workforce, where they hinder the Many industries, including the construction industry, are currently deal- productivity of the project progress and disrupt the workflow of their ing with a skills mismatch amid higher competition for job-seekers team, which can lead to more delays and pressure to meet tighter dead- (Ejohwomu, Proverbs, & Olomolaiye, 2005). The issue is finding quali- lines (Elnaga & Imran, 2013; Latham & Kinne, 1974). Also, with the con- fied workers at the current fast pace of a developing economy. Compa- stant introduction of new technology, the levels of competition in the nies that fail to attract skilled workers lag in terms of productivity and industry increase as workers are expected to be more productive and progress and lose out among the competition (Richardson, 2009). This have a higher output. Hence, the workforce must be well trained to failure may result in unfortunate consequences, such as overworked equip them with the necessary tools through education and skills for staff where they have to work extra hours to make up for the lack of them to work safely and productively. Therefore, companies need to workers or workers with lower skills, thus doubling their workload. A retain well-trained employees by investing in their well-being and shortage of skilled labor also leads to more untrained staff who may not enrolling them in training and education with a long-term view. execute jobs properly, leading to more defects or idling during the construction period—ultimately resulting in an imminent decrease in productivity levels for the whole firm. Therefore, organizations need to 5.1.4 | SE10: Effective project planning and execution ensure that the labor they hire has enough skills to make up for the shortage of workers or make the hiring incentives more attractive in order to attract newcomers as well as retain their current labor force. The planning aspect is especially crucial because of a need to evolve and change (Dvir, Raz, & Shenhar, 2003); in the construction industry, as newer construction technology emerges, it becomes even more 5.1.7 | SE7: Inspection delays important for companies to plan in order to keep their workforce current. Proper planning and execution are paramount for the construction Typically in the construction industry, supervisors or architects are industry due to the handling and management of resources. More responsible for inspections of the construction site, and inspection 413 WON ET AL. delays can potentially have a severe impact on contract completion. unpopularity and the principal motives for leaving are better prospects The inspection delays are usually caused by incompetent project man- in other industries, poor job conditions, and dissatisfaction with the agers, who fail to complete prioritizations for job inspections and profession and industry (Ling & Leow, 2008). Also, other reasons achieve inspectors include the lack of rewards and personal development, exhaustion (Makulsawatudom et al., 2004). Accordingly, it is crucial for project resulting from the lifestyle required by profession, and conflict managers or supervisors to stringently go through the works required, between workers’ requirements and those of other roles (Teixeira & such as quality assurance and safety inspection (Jarkas & Bitar, 2012). Gomes, 2000). The responsibility thus falls on construction firms to Coordination between inspectors and their project team is crucial to ensure that their recruitment terms are attractive enough to hire not ensure that inspections do not interfere with the contractor’s work just young workers but also older and experienced workers who are and delay progress. Delays in inspections that induce an obstruction looking for something new in an existing industry. Besides, safety con- in works can eventually lead to unwanted costs where the respective cerns go along with the hiring of less qualified workers to keep up parties have to be adequately compensated. with growing project demands. The shortage of experienced labor is a insufficient cooperation, and irresponsible severe problem, particularly in the construction industry (which is highly accident-prone and volatile, with many risks and hazards). 5.1.8 | SE8: Incomplete technical specifications Incomplete technical specifications cause repeated demands for clarification, and therefore successive delays and disturbances to work 5.2.2 | AE18: More dependence on equipment and technology progress. These unclear specifications have the potential to severely delay a project by causing rework and more variation orders, which Technological advancements have always played a critical role in driv- involve time-consuming meetings and negotiations with involved ing advancement in the construction industry. Companies that are parties, such as engineers, project managers, and architects, to rush to keen to be competitive are quick to adopt new technologies. New solve the issues. Accordingly, rectifications or substantial revisions technology helps to improve construction productivity by reducing may be required, which can lead to rework on sites caused by a lack the manual steps taken, such as inspections and technical specifica- of precise specifications at the beginning of the project (Jarkas & tions (Goodrum, Zhai, & Yasin, 2009; Loosemore, 2014). New tools Bitar, 2012); therefore, low productivity occurs. Defining a project in also help to build more creatively designed buildings and to monitor its entirety at the beginning is impossible; however, it is vital to have safety on the construction site by giving feedback and data analysis of critical specifications and details rather than going with an open- past incidents. Through such technology, workers can become more ended project. As such, the project should be approached by first interconnected, boosting their communication and access to informa- defining the partial scope and completing it before going forward with tion, which, in turn, can help them to be more efficient in whatever the rest of the project. Any initial underestimation or lack of detail is processes they are involved in (Rivard, 2000). Therefore, applying new likely to contribute in a significant manner to project cost overruns or technologies and tools could support and make the younger and more perceived project failure. Thus, having workers with the appropriate inexperienced workers more productive and safer, particularly in com- technical expertise and project capability is essential to counter the parison to traditional building methods where teams are fragmented, negative impact of this diversity factor. Specifying important technical and the construction sites prove to be more dangerous and riskier. details during project development and contractual discussions and Thus, by fully embracing technology, PPP in the construction industry then renegotiating down the road would be more beneficial to pro- is bound to increase, as seen through visible results from existing ductivity (Yeo, 2002). tools. 5.2 5.2.3 5.2.1 Age and experience | | AE13: Shortage of experienced labor | AE14: High rate of labor turnover A high rate of labor turnover usually results in the cost of hiring new labor to replace the lost workers. The cost of hiring new labor also The shortage of experienced labor had the highest loading in the “Age includes the cost to train them to be as efficient as the previous and experience” diversity category. As modern times and companies worker; this is indispensable in order to ensure that no productivity is create more promises of new infrastructure and construction projects, lost. The theory is, if the outflow of employees is high, the company there is a challenge for construction firms struggling to maintain their stands to incur higher costs from losing existing workers, and produc- project teams to ensure there is enough qualified staff involved in the tivity levels are bound to decrease (Alinaitwe et al., 2007). Therefore, projects (Hafez, Aziz, Morgan, Abdullah, & Ahmed, 2014; Törner & companies should look to retention as a key to prevent this issue, Pousette, 2009). The construction industry in Singapore is often not a which results in a loss of experience, know-how, and capabilities. The popular choice among fresh graduates and is also not a typical indus- prevention of labor turnover can be achieved by ensuring that the try promoted in university courses. The main reasons for its needs and well-being of current employees are taken care of so that 414 WON ET AL. they have little desire to leave the company, as reiterated by Lim and apply for the job or not (Němečková, 2017; Barber, 2018; Garver, Alum (1995). In addition, cross-training is also recommended to Goffnett, Divine, Williams, & Davis, 2019). As companies cannot func- ensure that workers can cover up any losses when a team member tion without a strong and capable workforce, they have to make leaves or is absent. themselves attractive for workers to want to join them. Thus, the higher the salary, the stronger the draw to entice workers to consider the job offer. A higher salary is also often associated with a reward, 5.2.4 | AE17: Positioning of experienced consultants where current workers are expected to work harder and produce a Identifying the right place in the project team and company for a suit- as a form of motivation to drive the workforce to have a higher pro- able consultant is the first and most important decision to put a pro- ductivity output. The motivation, in turn, also acts as a retention tool ject team together. With the appropriate consultant positioned for existing workers who are rewarded for their loyalty and efforts adequately in a role that allows them to exercise their knowledge and dedicated to the company. Therefore, wages can be a useful tool to experience, the project team will have the ability to pursue the level drive up the diversity of “Age and experience” in construction firms, of efficiency and productivity they are targeting (Törner & and eventually improve productivity in the firm as well as the industry Pousette, 2009). A well-positioned consultant can make a difference as a whole. more prodigious output if they want a promotion or a bonus with a higher salary (Kipnis & Schmidt, 1988)—this shows how wages can act in the decisions made, the relationships, and the organizational hierarchy in the team. The post-interviewees also stated that both the proper positioning of experienced consultants and their supply of knowledge facilitate the progress of a project. Thus, by correctly posi- 6 | C O N C LU S I O N S A N D RECOMMENDATIONS tioning experienced consultants, project labors can be impacted in a positive manner, which creates an opportunity for productivity The Singapore Government is continuously striving for and pushing improvements. productivity growth in the construction sector to achieve sustained economic growth. However, the efforts rolled out in the government’s productivity roadmap mainly focus on the use of automation and 5.2.5 | AE15: Unrealistic deadlines for project completion technology to boost productivity levels in the construction industry, leaving less room to consider alternative resources and their potential impact on productivity. Due to the global trend of increasing work- When a project has a deadline that is almost impossible to meet, it force diversity in construction projects, to improve the industry per- can result in several repercussions, such as a threat to the worker’s formance through untapped potential, it is crucial to understand safety and health, productivity pressure, and decreased morale among factors affecting workforce diversity and their impact on productivity. workers (Raymond & Bergeron, 2008). Overworked workers also However, there is limited research investigating the relationship become less productive in time due to symptoms such as exhaustion between workforce diversity and productivity in the context of con- and fatigue. Longer work hours do not necessarily correlate to more struction projects. Therefore, the results of this study contribute to output; the output of workers pushed beyond their capacity may also the core body of knowledge and practice both in defining workforce be compromised in terms of quality, leading to issues such as defec- diversity factors and in assessing the relationship between diversity tive work. Worst-case scenarios could include fatalities due to con- factors and PPP. In addition, the study proposes viable strategies to stant overexposure leading to a weak and dangerous work ethic harness diversity to increase productivity in the construction industry, where workers cut corners to meet deadlines and, as a result, the which serves as a starting point for the industry to properly handle its work environment and productivity suffer. When the project team workforce diversity and enjoy the benefits of increased productivity. becomes too deadline-oriented—where everyone is under pressure to In turn, the productivity gains will re-promote workforce diversity and meet the demands placed on them by their superiors—spontaneity, create a virtuous cycle in the construction industry. Moreover, the innovation, and creation also decrease (Bunce & West, 1994); that is, cycle can further contribute to the establishment of a sustainable workers may not consider alternatives as they do not want to be side- working environment in the current trend that workforce diversity is lined, thus creating a tension-filled work environment that is not suit- increasing across different countries. able for productivity. This study aimed to analyze the relationship between labor diversity and productivity of construction projects. Diversity in the workplace brings about various benefits, and this study examined diversity 5.2.6 | AE16: Amount of pay and wages for its potential impact on PPP to understand precisely which type of diversity has the most significant impact on the construction sector in The power of how salary influences workers should not be under- the Singaporean context. A literature review was conducted to gain a estimated as it is the deciding factor for many in the workforce nowa- thorough understanding of the types of diversity present in the labor days; people use salary and wages as a deciding factor in whether to force as well as methods to measure the levels of diversity. Through 415 WON ET AL. the comprehensive literature review, diversity factors that could communication channels for better and faster decisions to be made, potentially affect construction productivity were identified. Based on ultimately raising productivity. this review, questionnaires were developed to obtain responses from As the construction industry is project-based and includes regular local industry workers in order to understand the level of diversity in interactions between different project teams, it is necessary to facili- their workplaces and the perceived impacts of the various identified tate meetings and communications between workers to ensure the diversity factors on project performance. A total of 58 responses were exchange and transfer of skills and knowledge between workers of collated, covering contractors, consultants, developers, and architects. different job positions and trades. The transfer of knowledge and skills Also, post-interviews with industry experts were conducted to acquire between workers of varying diversity groups can also help the more extensive information on Singapore’s construction industry. workers themselves to benefit from more experience gained, and Based on the survey and interview results, the following research companies stand in cross-training the workers. Also, the government’s objectives were achieved. plan to introduce more technology into construction projects should In order to assess the impact of workplace diversity on PPP, a be integrated with the utilization of an increase in diversity. Workers structural equation model of the relationship between workforce of different age groups and different levels of skill and education diversity factors and their respective categories and perceived PPP should be made aware of the new technologies present as well as was created using the SmartPLS software. The significance of the have a basic knowledge of construction technology in order for a diversity factors and their perceived productivity performance was more integrated workforce. analyzed using the partial least squares method and bootstrapping in In addition, construction firms need to address and fully implement the software to generate the relevant t-values and loadings to infer the findings from the suggested structural equation model and strate- the correlation levels. Two diversity types out of the four, namely gies. This can be done by creating more tailored workshops focused on “Skill and education” and “Age and experience,” were identified as training workers of different backgrounds, providing more platforms being the most statistically significant, rejecting the proposed null and coworking spaces to boost opportunities for workers to self- hypothesis, which determined that these two diversity types in the instigate cross-training and exchange knowledge and experience. The workforce would have a significant impact on construction hiring of more experienced supervisors can significantly increase the productivity. impact of the benefits of a stable leader influencing the project group, To propose feasible recommendations to enhance workforce and making better, more effective decisions will help to align workers to diversity and productivity in the construction industry, analysis results the same goal and create a more task-oriented team. In general, by from the structural equation model and post-interviews helped to diversifying management-level staff, the workplace productivity rate is fine-tune which areas of diversity efforts should be focused expected to increase. Therefore, it is recommended that organizations on. Factors with a higher impact on PPP within their diversity catego- have a constant gauge on the diversity index of their workforce as well ries were prioritized to create strategies to help productivity growth. as have hiring and management policies in place to ensure a diversified The results highlighted that diversity aspects such as efficient labor force to drive construction productivity as a first step to improving decision-making and countering the shortage of skilled labor had the the organization’s productivity level. The various factors under the “Skill highest impact in their respective categories of “Skill and education” and education” and “Age and experience” are recommended to be and “Age and experience.” Therefore, the bulk of efforts should be monitored or enforced to allow construction firms to mitigate any concentrated on these factors to implement the suggested strategies potential negative impacts of a lack of diversity and benefit from the in the workplace. Regarding the analysis results, the following recom- positive impacts to improve productivity. mendations are made. This study would help local organizations within the construction The “Skill and education” category should be prioritized through industry understand the untapped potential of workforce diversity more training programs to target the educational well-being of and its impact on PPP. With the construction sector in Singapore lag- workers. Establishing a well-trained workforce to execute tasks effi- ging behind other areas amid the government’s push for economic ciently can help to facilitate better workflow to raise productivity. The growth, it is paramount for the construction industry to source alter- shortage of skilled labor was also identified to be a prime concern in native ways to boost productivity levels. The BCA relies heavily on the post-interviewees. Therefore, it is recommended that companies physical and tangible methods to push productivity growth, such as either retain existing skilled and experienced staff or make incentives investing in construction technology. Despite this reliable and proven to work in the construction industry more attractive to avoid losing method, the government and construction organizations should also out to other sectors. These activities should be strategized in order to expand their scope and explore other means of productivity growth. avoid unnecessary downtime and significant loss of productivity. Fol- Thus, their efforts prompt the relevance of this study, which has lowing the identification of a lack of training, the effectiveness of effi- touched on and analyzed the diversity aspect of the labor force—a cient decision-making was highlighted by the SEM as well as the post- vital resource of the construction industry. Through this research, a interviewees. Therefore, companies need to recognize competent significant correlation between diversity factors and PPP has been supervisors as well as establish protocols and clear organizational established, which can help steer the focus and awareness into charts to set out appropriate and frequent meetings. This will support directing more efforts toward implementing and enhancing the diver- better discussion and decision-making sessions through clear sity factors to push for productivity growth within the industry. Thus, 416 WON ET AL. instead of limiting the scope of productivity improvement methods to RE FE RE NCE S standard and more typical strategies, organizations should evaluate Abdul-Rahman, H., Wang, C., Wood, L. C., & Low, S. F. (2012). Negative impact induced by foreign workers: Evidence in Malaysian construction sector. Habitat International, 36(4), 433–443. Aibinu, A. A., & Al-Lawati, A. M. (2010). Using PLS-SEM technique to model construction organizations’ willingness to participate in e-bidding. Automation in Construction, 19(6), 714–724. Ali, M., Kulik, C. T., & Metz, I. (2011). The gender diversity–performance relationship in services and manufacturing organizations. The International Journal of Human Resource Management, 22(7), 1464–1485. Alinaitwe, H. M., Mwakali, J. A., & Hansson, B. (2007). Factors affecting the productivity of building craftsmen-studies of Uganda. Journal of Civil Engineering and Management, 13(3), 169–176. Arditi, D., & Mochtar, K. (2000). Trends in productivity improvement in the US construction industry. Construction Management and Economics, 18 (1), 15–27. Bagozzi, R. P., & Yi, Y. (1988). On the evaluation of structural equation models. Journal of the Academy of Marketing Science, 16(1), 74–94. Bandhanpreet, K., Mohindru, P., & Pankaj, M. (2013). Antecedents of turnover intentions: A literature review. Global Journal of Management and Business Studies, 3(10), 1219–1230. Barber, E. (2018). Job Satisfaction-To determine whether Irish graduates’ favour intrinsic factors over extrinsic factors when choosing a place of employment? (Doctoral dissertation, Dublin, National College of Ireland). Building and Construction Authority (BCA) (2015a). The Second Construction Productivity Roadmap. Retrieved March 10, 2020 from https:// tinyurl.com/y7pwtu9s Building and Construction Authority (BCA) (2015b). Media Release – About 7,000 firms to benefit from S$450 million boosts for construction productivity. Retrieved from December 13, 2019, https://tinyurl. com/ybhtcqyw Building and Construction Authority (BCA) (2016). Construction Productivity R&D Roadmap. Retrieved November from 10, 2019, https:// tinyurl.com/yahurlxn Bunce, D., & West, M. (1994). Changing work environments: Innovative coping responses to occupational stress. Work and Stress, 8(4), 319–331. Chin, W. W. (1998). The partial least squares approach to structural equation modeling. Modern Methods for Business Research, 295(2), 295–336. Chung, J., Park, J., Cho, M., Park, Y., Kim, D., Yang, D., & Yang, Y. (2015). A study on the relationships between age, work experience, cognition, and work ability in older employees working in heavy industry. Journal of Physical Therapy Science, 27(1), 155–157. Dai, J., Goodrum, P. M., & Maloney, W. F. (2007). Analysis of craft workers’ and foremen’s perceptions of the factors affecting construction labour productivity. Construction Management and Economics, 25(11), 1139–1152. Department of Statistics Singapore (SingStat). (2017). Population Trends 2017. Retrieved December 13, 2019 from https://www.singstat.gov. sg/-/media/files/publications/population/population2017.pdf Department of Statistics Singapore (SingStat). (2020). Education, Language Spoken and Literacy. Retrieved March 10, 2020 from https://tinyurl. com/ybrjf9gv Dernovsek, D. (2008). Engaged employees. Credit Union Magazine, 74 (5), 42. Disselkamp, L. (Ed.). (2013). Workforce asset management book of knowledge. Hoboken, NJ: John Wiley & Sons. Dvir, D., Raz, T., & Shenhar, A. (2003). An empirical analysis of the relationship between project planning and project success. International Journal of Project Management, 21, 89–95. Ejohwomu, O. A., Proverbs, D. G., & Olomolaiye, P. (2005). A conceptual demand led model founded on theories of labour market mismatch for the people that make up the sustainable workplace and make an effort to harness the impact of diversity and translate it into productivity growth. Despite achieving its objectives and the findings from this study are still valuable and contribute to the body of knowledge, this study acknowledges some limitations during the process of the study. Although the number of respondents was still valid to perform the statistical analyses, a larger sample size would have helped to assure a more reliable and accurate investigation. As the responses across the three different project types were not equally proportional, and the respondents were from the firms working in Singapore, caution needs to be given when the results are interpreted and generalized. In addition, the proposed method, SEM itself, cannot assess the direct causality between workforce diversity and PPP. However, it can quantify the effect between the two variables by demonstrating discriminant validity among the measures, and enable us to achieve the valuable findings. Nevertheless, the findings are still valuable as they help uncover the impact of labor diversity on productivity, and the results contribute to addressing the meaningful relationship. Furthermore, the proposed strategies help global companies working in diverse countries to deal with not only their workforce diversity issues but also improve their productivity. All in all, this research has provided a stepping stone for the introduction of the correlation between labor diversity and PPP in the construction industry. Further studies exploring more diversity strategies to enhance productivity growth should be considered in order to establish a workforce diversity system and policies not just in the construction industry but across other sectors for productivity improvement. As the research encompasses the evaluation of each organization’s level of workplace diversity, this index can be further standardized and promoted to organizations. In particular, organizations need to keep track of their diversity levels in order to identify what their company may be lacking and develop strategies to counter issues or tap into existing diversification not previously realized. Therefore, future studies can further expand on and explore more comprehensively diversity’s potential role in the workplace and other benefits that this can bring. ACKNOWLEDGMENTS This research was supported by the MOTIE (Ministry of Trade, Industry, and Energy) in Korea, under the Fostering Global Talents for Innovative Growth Program (P0008747) supervised by the Korea Institute for Advancement of Technology (KIAT). Also, this work is supported by the Korea Agency for Infrastructure Technology Advancement (KAIA) grant funded by the Ministry of Land, Infrastructure and Transport (National Research for Smart Construction Technology: Grant 20SMIP-A158708-01). ORCID Bon-Gang Hwang https://orcid.org/0000-0002-9034-2033 WON ET AL. the construction and building services industry. Paper presented at: Proceedings of the 21st ARCOM Annual Conference. El-Gohary, K. M., & Aziz, R. F. (2013). Factors influencing construction labor productivity in Egypt. Journal of Management in Engineering, 30(1), 1–9. Elnaga, A., & Imran, A. (2013). The effect of training on employee performance. European Journal of Business and Management, 5(4), 137–147. Enshassi, A., Mohamed, S., Mustafa, Z. A., & Mayer, P. E. (2007). Factors affecting labour productivity in building projects in the Gaza strip. Journal of Civil Engineering and Management, 13(4), 245–254. Fornell, C., & Bookstein, F. L. (1982). Two structural equation models: LISREL and PLS applied to consumer exit-voice theory. Journal of Marketing Research, 19(4), 440–452. Fornell, C., & Cha, J. (1994). Partial least squares. In R. P. Bagozzi (Ed.), Advanced methods of marketing (pp. 52–78). England: Blackwell Business. Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39–50. Garver, M. S., Goffnett, S. P., Divine, R. L., Williams, Z., & Davis, C. (2019). Analysis of the job preferences of undergraduate supply chain students. Operations and Supply Chain Management: An International, 12(1), 31–41. Gladwin, T. N., Krause, T. S., & Kennelly, J. J. (1995). Beyond eco-efficiency: Towards socially sustainable business. Sustainable Development, 3(1), 35–43. Glaeser, E. L., & Vigdor, J. L. (2001). Racial segregation in the 2000 census: Promising news. Washington: Brookings Institution, Center on Urban and Metropolitan Policy. Goodrum, P. M., Zhai, D., & Yasin, M. F. (2009). Relationship between changes in material technology and construction productivity. Journal of Construction Engineering and Management, 135(4), 278–287. Hafez, S. M., Aziz, R. F., Morgan, E. S., Abdullah, M. M., & Ahmed, E. K. (2014). Critical factors affecting construction labor productivity in Egypt. American Journal of Civil Engineering, 2(2), 35–40. Hair, J. F., Hollingsworth, C. L., Randolph, A. B., & Chong, A. Y. (2017). An updated and expanded assessment of PLS-SEM in information systems research. Industrial Management & Data Systems, 117(3), 442–458. Hair, J. F., Ringle, C. M., & Sarstedt, M. (2011). PLS-SEM: Indeed a silver bullet. Journal of Marketing Theory and Practice, 19(2), 139–152. Hong, L., & Page, S. E. (2004). Groups of diverse problem solvers can outperform groups of high-ability problem solvers. Proceedings of the National Academy of Sciences, 101(46), 16385–16389. Hui, C. H., Chiu, W. C., Yu, P. L., Cheng, K., & Tse, H. H. (2007). The effects of service climate and the effective leadership behaviour of supervisors on frontline employee service quality: A multi-level analysis. Journal of Occupational and Organizational Psychology, 80(1), 151–172. Jarkas, A. M. (2015). Factors influencing labour productivity in Bahrain’s construction industry. International Journal of Construction Management, 15(1), 94–108. Jarkas, A. M., & Bitar, C. G. (2012). Factors affecting construction labor productivity in Kuwait. Journal of Construction Engineering and Management, 138(7), 811–820. Jarkas, A. M., & Radosavljevic, M. (2013). Motivational factors impacting the productivity of construction master craftsmen in Kuwait. Journal of Management in Engineering, 29(4), 446–454. Jarkas, A. M., Radosavljevic, M., & Wuyi, L. (2014). Prominent demotivational factors influencing the productivity of construction project managers in Qatar. International Journal of Productivity and Performance Management, 63, 1070–1090. Jost, J. T., & Kay, A. C. (2005). Exposure to benevolent sexism and complementary gender stereotypes: C…

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.