- Define the concept of social capital as a job resource in the workplace.
- Summarize the new findings on the association of changes in social capital with work engagement, psychological well-being, and job performance.
- Discuss the study implications for workplace interventions to enhance social capital.
The development of well-functioning cooperative relations is important to meet the dual challenge of many contemporary industrial workplaces of sustaining efficient production processes while simultaneously keeping a focus on worker health and well-being.
In this perspective, the concept of job resources acquires particular salience, as the concept of job resources applies its focus on factors in the psychosocial work environment that enhances the possibilities of employees to undertake their work tasks while simultaneously sustaining the motivation and psychological well-being of employees.1,2
Previous studies have shown that job resources such as social support, influence at work, and managerial support have contributed to boosting motivational outcomes, such as work engagement,3–5 psychological well-being,6,7 and job performance.8–10 Other studies indicate that job resources, for example, influence at work and social capital, are negatively associated with adverse outcomes, such as risk of depression11–13 and long-term sickness absence.14,15
In the present study, we, therefore, aim to investigate whether a central job resource, social capital, is associated with outcomes related to work engagement, psychological well-being, and job performance.
In the literature, the concept of social capital refers to actual and potential resources in relationships between people,16–20 and social capital can, therefore, be considered a resource that manifests itself in social networks by affecting opportunities for collaboration and social support—for instance, in a workplace.4,21 Social capital manifests itself in different types of social relations, which has led to a distinction between different types of social capital—bridging, bonding, and linking social capital.22 In a work setting, the concept of bonding social capital has its focus on relations between people who belong to the same group or team. Bridging social capital refers to relations between people who belong to different groups or teams. Finally, linking social capital refers to relations between a work team and the management.
In developing a questionnaire on workplace social capital, Borg et al4,23 found a superior fit to a four-dimensional conceptualization that operationalized bonding social capital within work teams, bridging social capital between work teams, linking social capital in the relation between work teams and their immediate managers, and linking social capital between work teams in relation to the workplace as a whole.
By focusing on collaboration, helping behaviors, communication, mutual trust, and shared understandings of work tasks,19,23 the concept of social capital clearly qualifies as a job resource, as high levels of social capital on the one side may be expected to enhance job performance while simultaneously inducing motivation, well-being, and development of skills in employees.1,2
Only few studies have investigated prospective associations between social capital and relevant outcomes. These studies indicate that social capital is positively associated with work engagement,24 psychological well-being,11,12 and economic performance at the company level.25 In the present study, we add to the literature by investigating whether the four types of social capital predict increased job performance, work engagement, and psychological well-being in a prospective analysis.
Hypothesis 1: The four types of social capital (T1) are positively associated with job performance (T2), work engagement (T2), and psychological well-being (T2).
According to the conservation of resources (COR) theory,26 resources are important for the ability of the individual to cope with situational demands, as, for instance, in the work situation. Resource gain and resource loss are of particular salience to the well-being of the individual. In the present prospective study, we therefore investigate whether changes in social capital are associated with self-reported job performance, work engagement, and psychological well-being.
Hypothesis 2: Changes in the four types of social capital (from T1 to T2) are positively associated with job performance (T2), work engagement (T2), and psychological well-being (T2).
Due to its focus on cooperative relations in work teams, the concept of social capital may, at the theoretical level, be considered a group-level phenomenon. In many studies, however, social capital has been analyzed as an individual-level phenomenon in statistical analyses.11,24 In other studies, it has been argued that social capital may be most appropriately measured at the level of work teams,4,27,28 the reason being that social capital must be considered a shared property of work teams rather than a characteristic of specific individuals. In this study, we, therefore, analyze the associations described in the two hypotheses at the individual level and at the team level to investigate whether there are any differences in the observed associations depending on the level of measurement of social capital.
This study is based on prospective survey data from an intervention study on Danish dairy workers.29 A total of six dairies participated in the study. All employees including management were invited to participate in the survey. The data collection took place from June to August in 2015, and to ensure the highest possible response rate, reminders were sent out. Participation was voluntary. The same procedure was applied for the follow-up study that was conducted in the period from February to April in 2017.
In the baseline study (T1), 945 persons were sent a questionnaire, and we received a total of 791 responses (83.7%). Five hundred thirty-eight of the participants in the baseline study also participated in the follow-up study (T2), and, accordingly, the present study is based on response from these 538 employees. In the analyses at the team level, we excluded participants working in teams with less than three employees and, accordingly, these analyses were conducted on the basis of response from 533 respondents working in 66 teams with three or more employees.
An analysis of nonresponse showed that respondents who did not participate at T2 had significantly lower scores on social capital between teams (P = 0.007), social capital toward the workplace as a whole (P = 0.025), work engagement (P = 0.018), and psychological well-being (P = 0.014) than participants who took part at T1 and T2. The analysis showed no statistically significant differences between the two groups in terms of social capital within teams (P = 0.077), social capital in relation to immediate manager (P = 0.135), and self-reported job performance (P = 0.172).
Social capital was measured using the Danish social capital questionnaire.4,23 The questionnaire aims to capture social capital as a group-level construct from individual responses by applying the method of reference shift consensus.30 The questionnaire consists of four subscales: social capital within teams (bonding) consists of six items (Cronbach's α [T1]: 0.88; ICC2 [T1]: 0.54; Cronbach's α [T2]: 0.89; ICC2 [T2]: 0.49). Sample item: “In my team, we help colleagues who have too much to do.” Social capital between teams (bridging) consists of six items (Cronbach's α [T1]: 0.95; ICC2 [T1]: 0.74; Cronbach's α [T2]: 0.96; ICC2 [T2]: 0.74). Sample item: “My team and Team X acknowledge each other's contribution to solve the work tasks.” Social capital in relation to the immediate manager (linking) consists of six items (Cronbach's α [T1]: 0.95; ICC2 [T1]: 0.71; Cronbach's α [T2]: 0.95; ICC2 [T2]: 0.71). Sample item: “Our immediate manager takes our needs and views into consideration when he/she makes decisions.” Finally, social capital toward the workplace as a whole (linking) consists of three items (Cronbach's α [T1]: 0.78; ICC2 [T1]: 0.56; Cronbach's α [T2]: 0.78; ICC2 [T2]: 0.55). Sample item: “There is a common understanding between the management and employees about how we complete the tasks.” Participants responded on a five point Likert scale ranging from “To a very low extent” to “To a very high extent.” For each subscale, items were added into scales and scales were rescaled from 0 to 100 with high scores indicating high levels of social capital.
Changes in the four types of social capital from T1 to T2 were computed by subtracting social capital scores at T1 from social capital scores at T2. Accordingly, the four variables measuring changes in social capital scores potentially range from −100 to +100.
Scores on the four subscales and changes from T1 to T2 were aggregated to the team level by calculating the team-level average for each of the 66 work teams.
Work engagement was measured with the UWES9,31 consisting of nine items where the participant responded on a seven-point Likert scale ranging from “Never” to “Always” (Cronbach's α [T1] = 0.93; Cronbach's α [T2] = 0.93). The nine items were added into scales and scales were rescaled from 0 to 100 with high scores indicating high levels of work engagement.
Self-reported job performance was measured from two items from the Danish Psychosocial Work Environment Questionnaire.32 The two items were: “Do your working conditions allow you to carry out your work satisfactorily?” and “Can you carry out your work tasks without disturbing interruptions?” The two items exhibited satisfactory interitem correlation to be added to a scale (Pearson's r [T1] = 0.52, Pearson's r (T2) = 0.48). Participants responded using a five-point Likert scale with response options ranging from “To a very low extent” to “To a very high extent.” The two items were added into a scale that was rescaled from 0 to 100 with high scores indicating high levels of job performance.
Psychological well-being was measured using the five-item WHO-5 well-being scale.33 Sample item: “Over the past two weeks, how large a part of the time have you felt active and energetic?” (Cronbach's α [T1] = 0.86; Cronbach's α [T2] = 0.82). Participants responded on a six-point Likert scale ranging from “All of the time” to “None of the time.” All items were added into scales and scales were rescaled from 0 to 100 with high scores indicating high levels of psychological well-being.
Analyses were adjusted for age and sex. Information on these variables was collected in the study questionnaire at T2.
Data were analyzed using linear multilevel linear regression analysis. We chose this mode of analysis because the participants in the study were nested in work teams and by taking random effects at the group level into account in the analyses—that is, the effects of differences between workgroups that are not measured by the specific variables34,35—we were able to take the lack of statistical independence between the observations into account in the analyses.
In the multilevel linear regression analyses, we adjusted the analysis of the association between the four types of social capital and three outcome variables measured at T2 in two steps. First, the analyses were adjusted for random effects at workgroup-level and outcome variable measured at T1; and second, the analyses were additionally adjusted for age and sex of the respondents. In the analyses reported in Table 2, all predictors were measured at T1. In the analyses reported in Table 3, all predictors were measured as the difference between T2 and T1 scores on social capital. Respondents with missing values were excluded from the analyses.
Analyses in model 1 in Tables 2 and 3 were conducted at the individual level, with social capital scores of individual respondents posing as predictors of the three outcomes also measured at the level of individuals. Analyses in model 2 in Tables 2 and 3 were conducted at the team level, with social capital scores aggregated to the level of work teams posing as predictors of the three outcomes measured at the level of individuals. The analyses can be described from the following formula:
To investigate whether the findings of the study were affected by respondents changing from one work team to another, we also conducted a sensitivity analysis on participants who kept working in the same team during follow-up. The sensitivity analysis was performed at the individual level and replicated the analyses reported in Tables 2 and 3.
All analyses were performed using SAS version 9.4 using the PROC MIXED procedure (SAS Institute Inc., Cary, NC).
Table 1 shows descriptive statistics for main study variables.
Results From Analyses With Social Capital Measured at the Individual Level
Model 1 in Table 2 shows that the individual-level measures of social capital between teams and social capital toward the workplace as a whole (T1) significantly predict self-reported job performance (T2). The prospective association between social capital in relation to immediate manager and self-reported job performance is close to statistical significance in model 1, whereas the association between social capital within teams is not significantly associated with self-reported job performance.
Table 2 furthermore shows no statistically significant associations between the individual-level measures of the four types of social capital (T1) and work engagement (T2). Finally, Table 2 shows that all four types of social capital measured at the level of individuals (T1) significantly predict psychological well-being (T2).
Model 1 in Table 3 shows that changes in all four types of social capital are positively associated with all three outcomes. The results from model 1 show that positive changes in all four types of social capital significantly predict increased levels of self-reported job performance, work engagement, and psychological well-being.
Results From Analyses With Social Capital Aggregated to the Team Level
In model 2 in Table 2, we found no statistically significant associations between the four types of social capital aggregated to the team level and the three outcomes measured at the individual level.
In model 2 in Table 3, most of the observed associations between changes in team-level social capital and the three outcomes were statistically significant.
Results in Tables 2 and 3 were adjusted for random effects at team level and outcome measured at T1. Results from analyses that were additionally adjusted for age and sex of participants yielded similar findings (results not shown).
The sensitivity analysis that was performed on the 487 participants who kept working in the same team during the follow-up period showed a similar pattern of results as reported in the analyses reported in model 1 in Tables 2 and 3 (results not shown).
In the present study we investigated the prospective association between four types of social capital in the workplace and three outcomes: self-reported job performance, work engagement, and psychological well-being. When investigated at the individual level, the results provided some support to hypothesis 1, stating that the T1 level of social capital would predict the three outcomes at T2. Moreover, the individual-level results provided full support for hypothesis 2, stating that changes in the four types of social capital from T1 to T2 predicted the three outcomes at T2. In the analyses where the four types of social capital were aggregated to the team level, the results provided no support for hypothesis 1 and some support for hypothesis 2. The study adds to the field of social capital by providing new, prospective evidence on the association between four types of social capital and three important outcomes: job performance, work engagement, and psychological well-being.
When analyzing the association between social capital measured at the individual level and self-reported job performance and psychological well-being, the results lend support to hypothesis 1. The results showed that the T1 levels of all four types of social capital in the workplace were associated with the psychological well-being of the dairy workers at T2. These results are in line with previous findings in the literature11,12 and with the theoretical implications of the concept of job resources that is also expected to be associated with worker well-being.1,2 As stated above, however, none of the statistically significant findings could be replicated in the analyses, where the measures of social capital had been aggregated to the team level.
Moreover, the results showed that social capital between teams and social capital toward the workplace as a whole at T1 were significantly associated with self-reported job performance at T2. These results are in line with previous findings from the literature on job resources8–10 and with the theoretical implications of the concept of job resources that underline the facilitating role of job resources in getting the job done.1,2 In this analysis, we found that the association between social capital in relation to the immediate manager and self-reported job performance was borderline statistically significant (P = 0.051), which also lends some supports to hypothesis 1, but, surprisingly, we found no statistically significant association between social capital within teams and self-reported job performance. This finding is indeed surprising as it could be expected that the proximate collaborative relations within work teams should be important for the workers own perceptions of his or her job performance. More research may be needed to investigate this issue further.
Finally, the results showed that none of the four types of social capital predicted work engagement at T2. These results counter hypothesis 1 and the findings from previous studies that showed positive associations between workplace social capital and work engagement.24,27,36,37 More research may therefore be needed to investigate the prospective association between the four types of social capital and work engagement.
In hypothesis 2, we outlined an expectation that changes in social capital from T1 to T2 were also associated with the three outcomes: self-reported job performance, work engagement, and psychological well-being. The hypothesis was fully supported by the findings as the development in all four types of social capital in the workplace from T1 to T2 was significantly associated with all three outcomes at T2. This hypothesis was inspired by the COR theory26 that states that not just the level of resources but also changes in resources are important for the well-being of individuals. The findings support this theoretical assertion as we find stronger associations between changes in social capital and the three outcomes than in the analyses where the level of social capital at T1 is used to predict the three outcomes. These findings indicate that the development in social capital from T1 to T2 may be a more important determinant of self-reported job performance, work engagement, and psychological well-being than the absolute level of social capital at T1. In these analyses we also found a higher level of uniformity between the results of the individual- and team-level analyses. To the best knowledge of the authors, these associations have not been investigated in previous studies and the study thereby adds to the literature by investigating the association between changes in social capital and important work-related outcomes.
The findings reported in Table 2 indicate that the observed associations between individual-level measures of social capital and the three outcomes were stronger than the associations observed between social capital aggregated to the team level and the three outcomes. This finding is similar to findings from a previous study.11 One explanation may be that the aggregation of social capital to the team level may increase the risk of misclassification. One source of misclassification in the team-level analyses may stem from the level of agreement on the quality of social capital among employees in work teams,38 with higher levels of disagreement leading to higher levels of misclassification. Indeed, a previous study on the study population used in the present study indicated considerable variation in the level of intrateam agreement on the four types of social capital across the participating teams.4 Moreover, the varying levels of intrateam agreement across the participating teams may also contribute to explaining the low intraclass correlations observed for the measures of social capital within teams and social capital toward the workplace as a whole. Another explanation on the weaker associations observed in the team-level analyses may be that the power of the statistical models is reduced when moving from the level of individuals to more aggregated levels, as the number of observations in the aggregated variable is reduced from the number of individuals in the analyses to the number of teams in the analyses. The relatively identical levels of the beta-values observed in the individual and aggregated models reported in Tables 2 and 3 do indeed suggest that the lower P values in the aggregated model vis-á-vis the individual model may be ascribed to lower statistical power in the team-level analyses.
To sum up, the findings of this study imply that social capital in the workplace is associated with important outcomes in contemporary industrial workplaces. Efforts to enhance the social capital in the workplace may contribute toward improving both job performance and psychological well-being of workers. The study also suggests that the larger the improvements are in the social capital, the higher levels in job performance, work engagement, and psychological well-being can be expected over time.
This study has both strengths and limitations. One limitation of the study may be that the study is based on a small sample of teams and that we are unable to adjust for other potential confounders than age and sex. The analyses reported in Table 2 are prospective with a 2-year follow-up, which is a strength of the study as this temporal separation between dependent and independent variables reduces the risk of common methods biases inflating the observed associations.39 The results reported in Table 3, however, are not entirely prospective as we investigate associations between the difference in social capital between T2 and T1 and the three outcomes at T2. Therefore, it cannot be ruled out that the stronger associations between the predictors and the three outcomes observed in Table 3 vis-á-vis Table 2 may, to some extent, be ascribed to common methods biases.39 Another limitation of the study may be that participants changed work teams during follow-up. However, the sensitivity analysis including participants who did not change from one work team to another during follow-up showed a similar pattern of results as in the main analysis, which supports the robustness of the findings. Finally, analyses of nonresponse indicate that respondents lost to follow-up have lower scores on work engagement, psychological well-being, and two types of social capital, whereas we found no other statistically significant differences between the two groups on the other main variables of the study. These differences must be taken into account in the interpretation of the results of the study. Strengths of the study are that it is based on a workplace intervention study in the industrial sector, which implies that the results may characterize associations between social capital and important outcomes as they are observed in concrete workplaces. The finding of significant results in spite of the relatively small study population and when adjusting for outcomes measured at T1 is also considered a strength of the study.
The study showed that social capital in the workplace predicted self-reported job performance and psychological well-being in employees. The study furthermore showed that the magnitude of the changes in social capital predicted self-reported job performance, work engagement, and psychological well-being in employees at both individual and aggregated levels.
To sum up, the results imply that social capital is an important job resource and that it may well be worth the effort to implement interventions aimed at enhancing social capital as the size of the improvement in the four types of social capital is associated with outcomes that are highly relevant in work organizations.
The financial support from the Danish Work Environment Research Fund (26-2014-09) is gratefully appreciated. In addition, the authors would like to thank Torben R. Jans and Paw S. Jensen from the Danish Dairy Cooperative Forum for their support of the study. Finally, the authors would like to thank the participating dairies for their time and effort.
1. Schaufeli WB, Bakker AB. Job demands, job resources
, and their relationship with burnout and engagement: a multi-sample study. J Organ Behav
2. Bakker AB, Demerouti E. The job demands-resources model: state of the art. J Manag Psychol
3. Mauno S, Kinnunen U, Ruokolainen M. Job demands and resources as antecedents of work engagement: a longitudinal study
. J Vocat Behav
4. Meng A, Clausen T, Borg V. The association between team-level social capital and individual-level work engagement: differences between subtypes of social capital and the impact of intra-team agreement. Scand J Psychol
5. Halbesleben JRB. Bakker AB, Leiter MP. A meta-analysis of work engagement: relationships with burnout, demands, resources and consequences. Psychology Press, Work Engagement: A Handbook of Essential Theory and Research
. New York, NY: 2010.
6. Theorell T, Hammarström A, Aronsson G, et al. A systematic review including meta-analysis of work environment and depressive symptoms. BMC Public Health
7. Daniels K, Watson D, Gedikli C. Well-being and the social environment of work: a systematic review of intervention studies. Int J Environ Res Public Health
2017; 14: pii: E918.
8. Halbesleben JRB, Wheeler AR. The relative roles of engagement and embeddedness in predicting job performance and intention to leave. Work Stress
9. Harter JK, Schmidt FL, Hayes TL. Business-unit-level relationship between employee satisfaction, employee engagement, and business outcomes: a meta-analysis. J Appl Psychol
10. Xanthopoulou D, Bakker AB, Demerouti E, Schaufeli W. Work engagement and financial returns: a diary study on the role of job and personal resources. J Occup Organ Psychol
11. Kouvonen A, Oksanen T, Vahtera J, et al. Low workplace social capital as a predictor of depression: the Finnish Public Sector Study. Am J Epidemiol
12. Oksanen T, Kouvonen A, Vahtera J, Virtanen M, Kivimaki M. Prospective study of workplace social capital and depression: are vertical and horizontal components equally important? J Epidemiol Community Health
13. Theorell T, Jood K, Järvholm LS, et al. A systematic review of studies in the contributions of the work environment to ischaemic heart disease development. Eur J Public Health
14. Clausen T, Burr H, Borg V. Do psychosocial job demands and job resources
predict long-term sickness absence? An analysis of register-based outcomes using pooled data on 39,408 individuals in four occupational groups. Int Arch Occup Environ Health
15. Török E, Clark AJ, Jensen JH, et al. Work-unit social capital and long-term sickness absence: a prospective cohort study of 32 053 hospital employees. Occup Environ Med
16. Coleman JS. Social capital in the creation of human capital. Am J Sociol
17. Almedom AM. Social capital and mental health: an interdisciplinary review of primary evidence. Soc Sci Med
18. Kawachi I, Berkman LF. Social ties and mental health. J Urban Health
19. Nahapiet J, Ghoshal S. Social capital, intellectual capital, and the organizational advantage. Acad Manag Rev
20. Putnam RD. The strange disappearance of civic America. Policy J Public Policy Ideas
21. Kouvonen A, Kivimäki M, Vahtera J, et al. Psychometric evaluation of a short measure of social capital at work. BMC Public Health
22. Szreter S, Woolcock M. Health by association? Social capital, social theory, and the political economy of public health. Int J Epidemiol
23. Borg V, Cayuelas-Mateu N, Clausen T. Udvikling af en ny metode til undersøgelse af social kapital på arbejdspladsen [Development of a new method for investigating social capital in the workplace]. København: Det Nationale Forskningscenter for Arbejdsmiljø [National Research Centre for the Working Environment]; 2014.
24. Strömgren M, Eriksson A, Bergman D, Dellve L. Social capital among healthcare professionals: a prospective study of its importance for job satisfaction, work engagement and engagement in clinical improvements. Int J Nurs Stud
25. Westlund H, Adam F. Social capital and economic performance: a meta-analysis of 65 studies. Eur Planning Studies
26. Hobfoll SE. The influence of culture, community, and the nested-self in the stress process: advancing conservation of resources theory. Appl Psychol Int Rev
27. Fujita S, Kawakami N, Ando E, et al. The association of workplace social capital with work engagement of employees in health care settings: a multilevel cross-sectional analysis. J Occup Environ Med
28. Kowalski C, Driller E, Ernstmann N, et al. Associations between emotional exhaustion, social capital, workload, and latitude in decision-making among professionals working with people with disabilities. Res Dev Disabil
29. Meng A, Borg V, Clausen T. Enhancing the social capital in industrial workplaces: developing workplace interventions using intervention mapping. Eval Program Plann
30. van Mierlo H, Vermunt JK, Rutte CG. Composing group-level constructs from individual-level survey data. Organ Res Methods
31. Schaufeli WB, Salanova M, Gonzalez-Roma V, Bakker AB. The measurement of engagement and burnout: a two sample confirmatory factor analytic approach. J Happiness Studies
32. Clausen T, Madsen IE, Christensen KB, et al. The Danish Psychosocial Work Environment
Questionnaire (DPQ): development, content, reliability and validity. Scand J Work Environ Health
33. Bech P, Olsen LR, Kjoller M, Rasmussen NK. Measuring well-being rather than the absence of distress symptoms: a comparison of the SF-36 Mental Health subscale and the WHO-Five Well-Being Scale. Int J Methods Psychiatr Res
34. Cohen A, Keren D. Individual values and social exchange variables: examining their relationship to and mutual effect on in-role performance and organizational citizenship behavior. Group Organ Manag
35. Diez Roux AV. The study of group-level factors in epidemiology
: rethinking variables, study designs, and analytic approaches. Epidemiol Rev
36. Carmeli A, Ben-Hador B, Waldman DA, Rupp DE. How leaders cultivate social capital and nurture employee vigor: implications for job performance. J Appl Psychol
37. Susanne LB, Kowalski C, Wirtz M, et al. [Work engagement of hospital physicians: do social capital and personal traits matter?]. Psychother Psychosom Med Psychol
38. Gonzalez-Roma V, Hernandez A. Climate uniformity: Its influence of team communication quality, task conflict, and team performance. J Appl Psychol
39. Podsakoff PM, MacKenzie SB, Lee JY, Podsakoff NP. Common method biases in behavioral research: a critical review of the literature and recommended remedies. J Appl Psychol
Keywords:Copyright © 2019 by the American College of Occupational and Environmental Medicine
epidemiology; job resources; leadership; longitudinal study; occupational health; psychosocial work environment; social support; working conditions