International Journal of School and Cognitive Psychology
Open Access

The Effects of Stressors on Work Life Balance: The Case of World Vision Ethiopia Head Office Employees, Addis Ababa, Ethiopia

Hunde Benti Gudina¹ and Dagne Tafa Gemechu^{2* *}Correspondence: Dagne Tafa Gemechu, Department of Psychology, Ambo University, Oromia, Ethiopia, Email:

Author info »

Introduction

Human capital is believed to be driving force to economic development. Effective utilization of man power to boost growth and development requires management of wellness of human resources. Jim Young Kim, World Bank President (2013) in his speech advised countries the need to build human capital through investments in health, education and social protection for all their citizens to drive growth. Thus, staff wellness has to be focus of human resource management of every organization. Keeping the balance between work and other aspects of life is one aspect of ensuring staff wellness.

Work Life Balance (WLB) in today’s dynamic economic world become strategic wellbeing issues of employees in any organization. Increasing number of studies has endorsed the importance of maintaining work life balance for effectiveness and efficiency of career and life of individuals as well as for realizations of organizational goal. Poulose S et al. argue that, WLB and employee perception of wellbeing have come to be recognized to be vital for the organizational growth and effectiveness [1]. It is natural that employees seek work and home environment where they can enjoy balanced life. According to Frone MR et al. increasing demands and concerns of employees of balancing work and family life can be a major cause of stress in an individual’s life; leading to a decrease in satisfaction for both domains and result in destructive effects on health of employees and their wellbeing [2]. Particularly, at this age of globalization, stress affect balance of work life of employees, this in turn hampers success and satisfaction in both work and life.

Stressors, especially hindrance ones, significantly disturb work life balance of employees in all aspects. Widmer, Semmer, Kälin, Jacobshagen, Meier explain that this hindrance stressors result in stressful situations and obstruct work accomplishments. Role overload, role ambiguity, work family conflict are among stressors negatively affecting well beings of employees. These stressors distract work life balance by causing hassle to time management, involvement balance and finally lead to dissatisfaction in work and life in general. Greenhaus JH et al. elaborate life balance dimensions as equal time devotion, equal psychological effort and presence invested, and equal satisfaction expressed across work and family roles [3]. Kim HK et al. further explains balance of work life, as balance between work and a variety of roles to be considered as individuals perform multiple roles at the same time through their complex relationships with neighbors, friends, and community [4].

Among WVE employees, such stressors and their impacts on life balance are observable. Ashforth, Kreiner and Fugate describe that, there is a recognition that individuals may be actively participating in one role while simultaneously feeling distracted by thoughts, emotions, or demands that are tied to another role. From the daily observable behavior and actions of WVE head office employees, echoes of worry and fretting about some problem at work in the organization or something that would demand their attention at home, are reflected. Therefore, the consequence of such negatively affecting stressors is hurting overall wellbeing of the employees as well as organizational achievements finally put meaningfulness or satisfaction of life in jeopardy. However, the main problem in our country at large and in WVE HO in particular, there is gap of systematic study findings to support the reality with tangible evidences. Therefore, this study is primarily to shed light on this rift with evidences empirically tested to bridge the gap. Identifying the occurrence and extent of impact of stressors on life balance is critical to address the problems for healthy and productive employees and effective organizational development program management. To this end the following research hypotheses were designed.

H1: There is significant effect of stressors on employee work life balance

H2: Employees social support moderated the relationship between the stressors and work life balance.

H3: The influence of stressors on work life balance is not the same across department

H4: The influence of stressors on work life balance is not the same over service years of the employees

Materials and Methods

Design and population

The study has utilized quantitative cross sectional research design which enables to make statistical association between dimensions of stressors, work life balance and interactive effects of social supports. The target population of this study was three hundred seventy three (N=373). The sample size (N=193) of the study was determined using Yemane’s formula. The participants of the study systematic sampling technique were selected using.

Instruments of data collection

As mentioned above, the study has used standardized psychometric devices developed by different scholars. Modification and adaptation works were done to all scales so that it fits contextual reality. The scales were revised and made context appropriate for its content and language appropriateness to the target study participants.

Peter Gröpel developed the work life balance index and these measures have been extended to how one’s personal life is enhanced by work (or) vise-versa. The scale consists of 20 items each having six alternatives with 5 point Likert scale (1=strongly disagree to 5=strongly agree). The scale has sub measurable constructs that were asked to identify work/achievements which have to do with time balance, contact/relationship that deals with the involvement balance issues and life/meaningfulness or satisfaction balance of work life. The time balance such as the amount of time employees spend for work and family or friends, the involvement balance in work and life and the satisfactions employees get in their work life were treated as dependent variables.

Independent variables were independent variables are stressors and its indicators: Work overload, role ambiguity and workfamily conflicts that influence life balance of employees [5] . Five items with five point Likert-scales that range from (1=strongly disagree to 5=strongly agree). Work Overload (WOL), by Spector and Jex, were used to measure the constructors. In this investigation, role ambiguity items were taken from Rizzo, House, and Littman’s and, similarly, measured using five-point Likert-scale.

It is used to measure the level of ambiguity about roles authority and responsibility, work objective, necessary information about the job, and the expectation of others from them. Work family conflict was also measured by twelve items with five point Likert scales that range from (1=strongly disagree to 5=strongly agree). This was originally developed by Kopelman, Greenhaus and Connolly (1983).

Moderating variables were supervisory support, spousal supports, coworker support and family/relative supports were used as moderating variables of the study. Supervisory support was measured by six items with five scales point Likert-scale, the first four items were developed by Greenhaus JH et al. the fifth one from Mor Barak and the last one from Scarpello and Vandenberg [6]. Items for spousal support measures were taken and adapted from House. Five questions were also asked for spousal support.

Instrument validity and reliability

Validity refers to the extent to which an instrument measures what it is supposed to measure. There are various instrument validity for the variables in terms of content validity, external validity and construct validity (convergence and discriminant). Regarding content validity, the questions were evaluated by other university professors. For inferential purpose the researcher also used adequate sample size which fulfills the external validity implication to infer the result to whole population in the study area. Further, convergence and discriminant validity also assessed.

Reliability related to the consistency of the measurement regardless of what it measure whether or not a test produced the same results on different occasions [7]. Hence, the measure is reliable when respondents give the same responses in different situations. This study used the following three criteria to evaluate reliability of the instruments. First, Cronbach‟s alpha ought to be above 0.70 [8]. Second, corrected item-total correlations ought to be retained if the value not less than 0.35 [9]. Correlated item-total correlations should not be less than 0.3 Bernstein. This value revealed the extent to which, within a scale, an item correlated with the other items. It was employed to determine the items which ought to be retained in a scale to support construct validity. For better reliability, this study used 0.35 as cutoff point. Third, inter item correlation should not exceed 0.8 for all pairs of items.

As it can be seen in Table 1 the measure of WOL began with 5 items. One item was dropped because its correlated items total correlation was below 0.35. Hence, using 4 items Cronbach alpha for WOL was 0.896. RA began with 6 items, of which 1 was dropped because their correlated items total correlations were below 0.35 as a result; Cronbach alpha for RA was 0.847. Over all, the following Table 1 also shows the proposed number of items, number of items dropped because of its total item correlated below 0.35 and number of items retained and their corresponding Cronbach alpha value respectively.

Table 1: KMO and Bartlett’s test.

Method of data analysis

The collected responses were coded and analyzed by using the latest Statistical Package For Social Science (SPSS) and AMOS version (23). SEM is used for two purposes: First, similar to factor analysis, it provides summary of GOF indices to test the interrelationships among variables and second, similar to path analysis, SEM test hypothesized relationships among constructs. Both mean that the SEM method can simultaneously assess the properties of the underlying measurement model and test the theoretical propositions. Structural Equation Model (SEM) analysis was thus, performed to examine the effect of stressors: Composites of role overload, role ambiguity and work-family conflict on work life balance of employees. In addition, the path coefficient was also used to predict the moderating role of social supports between stressors and work life balance. The study also tested if the score of the stressors are different across departments and duration of stay in the organization on their life balance using path analysis. The data quality assessment with procedural assumptions is conducted as part of analysis.

Results

Factor analysis

Before the detail analysis of the data the factorability of the variables was tested by using the KMO and Bartlett’s test for Sphericity to measure the overall adequacy of sample to assess the appropriateness of indicators for factor analysis. According to Kaiser HF et al. a KMO value is between 0 (Factor analysis is likely to be inappropriate) and 1 (Factor analysis yield reliable factors) [10]. As it is indicated in the following table the KMO test value was 0.813 which lies within great range (between 0.8 and 0.9). Hence, the sample size taken for this factor analysis was quite adequate.

Factor extraction

The factor extraction was also used to determine the smallest appropriate number of factors that can represent the interrelationship between among set of factors considered. The two common means of extraction factors used in factor based investigations are communality and confirmatory factor analysis [11].

In most research works communality of the data is explained in common variance which is the variance that exists between specific variable and with other variables after factors extraction. The cut point for communality for retained factor after extraction is 0.5 for which common variance below 0.5 leads to the omission of the variable [12]. Accordingly, we found that all factors result in communality test of 0.5 is high communality. Hence, no items dropped because of lower factor loading result.

To determine the number of factors that most explain the variance in the data Eigen values was used. Hence, out of the initial 63 items with 63 Eigen values are 69.4% total explained variance of 14 factors by using Varimax rotation factor for the analysis of these data.

Confirmatory factor analysis

According to Byrne (2001) CFA is the measurement model which assesses the linkage of factors and its measurement variables. Confirmatory Factor Analysis (CFA) through Structural Equation Modeling (SEM) was used to assess construct validity through model fit indices Tabachnick and Fidell. It is statistical means by which the measurement model evaluated for its “goodness of fit” to sample data. The researcher used SEM for two purposes. Firstly, similar to factor analysis, SEM provides a parsimonious summary of the interrelationships among variables. Expanding on the potential of Extracted Factor Analysis (EFA), SEM can include CFA that can test specific hypotheses about the structure of the factor loadings and intercorrelations Holmes and Smith. Secondly, similar to path analysis, SEM can test hypothesized relationships among constructs with a linear equation system Weston and Gore. Both applications mean that the SEM method can simultaneously assess the properties of the underlying measurement model and test the theoretical propositions. For analytical purposes, the SEM method can be separated into two models: The measurement model and the structural model Byme. The measurement model is concerned with the variables that are supposed to test the concept or, in other words, the measurement model represents the CFA model. These are the stressors, work life balance and moderators. And shows how the latent variables, or constructs, are represented by their respective indicators. These latent variables are work overload work family conflict, role ambiguity, time balance, involvement balance, satisfaction balance, supervisory support, wife support, coworkers support and friends support. Moreover, it is aimed to test the relationship between multiple observation variables explained by latent variables and independent variable by assessing the suitability or fitness of hypothetical construct to sample data. Hair JF further explained hypothetical construct validity assesses which set of measured items actually reflects the underlying factors model through convergent and discriminant validity [10].

Within the following Table 2 the examination of some selected measures of Goodness of Fit (GOF) summary with its criteria which was selected and adopted from Asmare E et al. were within the acceptable range in terms of independency test, absolute, incremental and parsimonyfit indices [13]. Hence the full structural model shown in Figure 1 was supported and accepted based on these literature fit indices summary criteria.

Table 2: GOF Summary indices.

Figure 1: Full cofactor analysis of the measurement model (source: Amos output).

Source: Research amos out put

The convergent validity of the variables which is within the threshold of Squared Multiple Correlation (SMC) more than 0.4, except IB2, IB4, RA1, SB1, SB2, SB3, SB4, TB1,TB3, TB4, WFC1 and WOL1 which were deleted because they lack indicating the acceptability range. On the other hand the discriminant validity of the latent variables was also examined by using Pearson’s factors correlationof less than 0.8 or 0.9 which are acceptable range for model to be fit.

After the validity of the factors had been checked the internal consistency of the measurement instruments were evaluated, using coefficient of internal consistency (Cronbach’s Alpha). The recommended and widely accepted range of coefficient of internal consistency is 0.7. Likewise the following table indicates the Cronbach’s Alpha greater than 0.7 which asserted the reliability of measurement instruments (Table 3).

Table 3: Final instrument reliability.

The other means of assessing the structural model validity is to check the size, direction and significance of the structural path estimates. Accordingly, this shows the significance of the path estimates in which three of the total paths are statistically supported for hypothesis testing Figure 2.

Figure 2: Structural equation model.

Finally, the research questions hypothesized were examined for their sign of direct effect and indirect effect whether they were statistically supported or rejected (Table 4). Thus, except the interaction effect of moderator which was indicated to be less significant and all research hypothesis direct effects signals were supported and accepted Figure 3.

Table 4: Summary of hypothesis.

Figure 3: Factor scores from latent factors.

Discussion

Discussion of empirical findings

This section deals with the major objective of the research which is “the influence of stressors on work life balance, and the moderating effects of social supports to the life balance of employees” in the selected organizations. The factors were grouped in to three major factors as stressors (with three sub factors), social supports as moderating factor (with four sub factors), and work life balance (three sub parts) with their respective observable variables. Finally, the direct effect of stressors on work life balance and the moderation of the relationship between stressors and work life balance were tested. Further, the researcher also tested if there were significant differences of stressors based on workers’ services year and departments.

Hypothesis 1: There is significant effect of stressors on employee work life balance

The output of structural equation model (Amos version 23) indicated that all parts of stressors (WOL, RA, WFC) are statistically significant. This result shows work and family related stressors affect the work life balance of employees. These stressors are also the main actors which explained 79% (Figure 4) of the variation in work life balance of the works and only 21% of variation is due to other variables. Hence, the first hypothesis is accepted.

Figure 4: The standardized beta estimate in path stressors to work life balances for program department (A) and sponsorship department (B).

There is statistically significant and negative relationship between stressors and work life balances. Stressors have strong effect on work life balance of employees which can be explained by the standardized regression coefficient or path coefficient of 0.96 at p<0.01. This implies that organizational work over load, role ambiguity and work related family conflicts decreases the workers capability of balancing their work life (time, satisfaction and involvement). This is substantiated with the finding of similar researches by Razak MI et al. that stressor factors,workload are significant at p<0.01 (0.000) and role conflict also significant at p<0.01 (0.002) and both correlate with work life balance.Research result by (Mohd 2011) with periodic time test results on the effects stressors have also concluded that there are significant impact and causing strain, disturbance, to life balance of workers [14]. According to this resultrole overload and role ambiguity cause work life imbalance as the result of strain, empirical evidence by Ock WH et al. also shows that work overload and role ambiguity both in quality and in quantity are negatively related to job satisfaction aspect of work life balance [15]. Stressors are directly related to work life balance dimensions. As the professionals have more stress, their work interferes with personal life.

Hypothesis 2 a: The social support to workers in an organization has significant effects on their work life balance

The analysis result showed there is positive and significant relationship between social supports and employee’s work life balance. In this case, the hypothesis that the causal effects of social support on work life balance is supported. The strength of relationship resulted in the path coefficient (β=0.31 at p<0.01). Hence, the supports from spouse, supervisor, coworker and relatives/friends enable employees to balance their work life. Research result by Seulki L and Soo-Young L confirm similar results that social supports, particularly supervisory and spousal supports have significant positive relationship with job satisfaction dimension of work life balance.

Hypothesis 2 b: The stresses social supports moderates the relationship between stresses and work life balance

To get the extent that social supports moderate (interaction effect) the relationship between stressors and work life balance of the employees, the researcher used structural equation of creating factor scores from latent factors by using inputting factor scores in AMOS. Accordingly, factor scores from latent factors (Figure 5) were derived from the full measurement structural model. The result of path analysis for latent factor score for moderation effect is β=-0.18 at p=0.091 which implies the hypothesis that the moderating effects of Stress Social support on work life balance are significant and supported at 90% confidence level. Here, the significance level is much less when we compare with the other variables acting as independent variables. Since the hypothesis for the main effect is still significant after the moderator enters the model the type of moderation that occurs in this case is partial moderation.

Figure 5: Factor score in path stressors to work life balances, program department (A) and sponsorship department (B).

Further, the regression coefficient of product term (stress social support) on work life balance is negative, which indicates that the moderating variable (social supports) weakens the causal Effects of Stressors (STR) on individual Work Life Balance (WLB). The study result by Seulki L et al. still confirms the interactive effects of social supports between stressors job satisfaction in work life. However, they indicate that social support is likely to improve employees’ job satisfaction without interacting with stressors. It has been also demonstrated by some scholars Beehr TA et al. that social support interacts with job stress and has a buffering effect [16]. Similarly, indicated in the result of partial moderation effect of this hypothesis above.

Hypothesis 3: The influences of stressors across department (program and sponsorship)

The standardized parameter estimate for “Respondents from Program Department” is -0.71, while the same estimate for “Respondents from Sponsorship Department” is -0.15. Thus, one can conclude that the effect of stressors on work life balance is more pronounced in “Respondents from Program Department” compared to “Respondents from Sponsorship Department” (Table 5).

Table 5: Hypothesis summary.

Now the researcher wants to determine the type of moderation that occurs in the social support and work life balance relationship. The results show that the type of moderation is partial moderation since the standardized estimate for respondents from Program Department, while the standardize estimate for respondents from Sponsorship Department is full and significant. If both estimates are significant, then partial moderation occurs.

Hypothesis 4: The influence of stressors across service year (high and low)

To address the research question in which group (high service year group or low service year group) the stressors are more pronounced? Before taking these groups, the researcher calculated the average of overall respondents’ service years as cut point. Hence, those responses of service years below average were taken as low and the above average as higher. Then researcher determined the standardized path coefficients of both groups as follows Figures 6A and 6B.

Figure 6: The effect of stressors on work life balance across high (A) service year and Low (B) service year groups.

As it can be seen from the above Figures 6A and 6B the stressors significant and negatively affected in the case of high service year group (β =-0.75 at p<0.001, Z score =-4.244) and the effect is very low (β =-0.14) which was statistically not significant in low service year group. This also showed that the effect of stressors on work life balance is very high and strong in the case of high service year group (more experienced) than low service year employees in the organizations. Hence, the hypothesis of the stressors had high influence on work life balance of workers was accepted [17].

Recommendations

Based on the empirical finding of this study the following recommendations were done:

To enhance growth and productivity of both employees and the organization, considering installation of systematic role tracking and clarification is advantageous.

Job grading and role distribution should take into account the practical quantity and quality of work assigned to employees and the time it takes to accomplish. This would help decreasing or avoiding the scenarios of misuse, overuse or abuse of HR management [18]. Appropriate HR management is beneficiary both for employees holistic life as well as to contribute to growth and development of nations.

Employees need to make behavioral self-adjustment in role sharing, specially learn to avoid gender role stereotype belief to assist mismatch between family responsibilities like care for children and other home chores to increase work life balance for all.

Employees should get spouse supports like home work and child caring as well as supervisor and peer supports to get direction on work methods and practices when they are encountered with work related stressors.

Supervisors are advised to give more supports for employees who are working in program departments than the sponsorship to increase their capability of their work life balance [19].

Organizations have to initiate and enhance peer supports and team works to settle role ambiguities, work overload and work family conflict that may happen.

Either the employees face stresses due to work or family, it is advisable if the organization prepare joint employee and family wellness counseling program which enable them reduce their stresses level and support them healthy life style.

Flexible work schedules and relaxation activities play great role in reducing the stresses that might be emerged from personal life or organizational work environment [20]. Therefore, the organizations may arrange such as appropriately flexible work schedule, adopt Rest and Relax (RR) policy, facilitate working at home or telecommuting to accomplish their duties.

Conclusion

The study used the extent that literatures are employed as instrument to evaluate the work life balance of employees in their work. Based on the results, work over load, work family conflicts and role ambiguity were the major causing factors which were scored latent factors for stressors (STR). On the other hand, the study employed supporting factors (spouse supports, supervisor supports, coworker supports and family/ relative supports) generally as Social Supports (SoS) to moderate the relationship between stressors and work life balance. In this investigation Work Life Balances (WLB) were also scored latent composite factors of time balance, satisfaction balance and involvement balance. Furthermore, the measurement items (observed variables) were separately used for each latent variables mentioned. In addition to direct and moderation effects, the result also included the extent that the influence (strength) of stressors across employees department and service years.

Based, on the empirical finding this investigation was concluded as follows:

The structural equation model analysis showed that the stressors had strong and negative (β=-0.96) effect on the work life balances of the employees. This clearly indicates that employee’s work overload, employee work family conflict and role ambiguities decreases the individual work life balances.

The path analysis beta coefficient of social support (β=0.31) is significant and positively related to the work life balance. This implies that in place where social supports are available for workers, there would be the probability that the individual able to increase their work life balance.

Social supports (spouse supports, supervisor supports, coworker supports and family/relative supports) partially moderate the relationship between stressors and work life balances of the employees. The negative interaction effect (β=-0.18) showed social supports weaken the negative correlation relationship between the independent variable (STR) and dependent variable (WLB). Therefore, this implies that as workers get more supports from spouse, peers, families and supervisors on their work related issues may have enabled those individuals to balance their time, involvement and satisfaction.

The effect of stressors was stronger on employees who were working in program department than that of sponsorship. The effect (β=-0.71 at p<0.001) is negative and statistically significant. This indicates that the intensity and influence of stressors were different across department.

The moderation effect across departments was also pronounced more on respondents from sponsorship than that of program department. In fact funding source of WV is mainly child sponsorship business. Sponsorship staff paly integrated and added roles both in meeting standards of sponsorship business to maintain the funding sources as well as the actual program implementation to ensure child well wellbeing. Therefore, this implies that more social supports markedly required and assisting for employees who are working in sponsorship sections.

The result also showed, that the direct effect of stressors was positive and strong in the case of high service year group of the respondents, as far as the service year of the employees concerned. This might be due to the fact that high service year employees are experienced group which indicates the existence of multiple roles to be assumed, may be as the result of family, social and organizational position responsibilities.

References

1. Poulose S, Sudarsan N. Work life balance: A conceptual review. Int j adv manag economics. 2014;3(2):1-7.

   [Google Scholar]

2. Frone MR, Quick JC, Tetrick LE. Handbook of occupational health psychology. American Psychological Association. 2003.

   [Google Scholar]

3. Greenhaus JH, Collins KM, Shaw JD. The relation between work-family balance and quality of life. J Vocat Behav. 2003;63(3):510-531.

   [Crossref] [Google Scholar]

4. Kim HK. Work-life balance and employees’ performance: The mediating role of affective commitment. Global business and management research: An international journal. 2014;6(1).

   [Google Scholar]

5. Greenhaus JH, Parasuraman S, Collins KM. Career involvement and family involvement as moderators of relationships between work–family conflict and withdrawal from a profession. J Occup Health Psychol. 2001;6(2):91-100.

   [Crossref] [Google Scholar] [PubMed]

6. Adams GA, King LA, King DW. Relationships of job and family involvement, family social support, and work-family conflict with job and life satisfaction. J Appl Psychol. 1996;81(4):411-420.

   [Crossref] [Google Scholar]

7. Bordens KS, Abbott BB. Research design and methods: A process approach. McGraw-Hill; 2002.

   [Google Scholar]

8. Hair JF, Matthews LM, Matthews RL, Sarstedt M. PLS-SEM or CB-SEM: Updated guidelines on which method to use. J Multivar Anal. 2017;1(2):107-123.

   [Crossref] [Google Scholar] [PubMed]

9. Netemeyer RG, Bearden WO, Sharma S. Scaling procedures: Issues and applications. Sage publications; 2003 Mar 12-37.

   [Google Scholar]

10. Kaiser HF. An index of factorial simplicity. psychometrika. 1974;39(1):31-6.

    [Crossref] [Google Scholar]

11. Pallant J. SPSS survival manual: A step by step guide to data analysis using IBM SPSS. Routledge. 2020.

    [Crossref] [Google Scholar]

12. Thompson B. Exploratory and confirmatory factor analysis: Understanding concepts and applications. American Psychological Association. 2004.

    [Crossref] [Google Scholar]

13. Asmare E. The effect of business process reengineering on public sector organisation performance: A developing economy context. RMIT University.2012.

    [Google Scholar]

14. Razak MI, Yusof NM, Azidin RA, Latif MM, Ismail I. The impact of work stress towards work life balance in Malaysia. Int j economics commerce manag. 2014;2(11):1-6.

    [Google Scholar]

15. Ock WH, Kim SY. A study on job stress and job satisfaction, and organizational commitment of local public service. Korean Public Administration Review. 2001;35(4):355-73.

    [Google Scholar]

16. Beehr TA, King LA, King DW. Social support and occupational stress: Talking to supervisors. J Vocat Behav. 1990;36(1):61-81.

    [Crossref] [Google Scholar]

17. Iis MK. Over time effects of role stress on psychological strain among Malaysian public university academics. Int j soc sci bus. 2011;2(9).

    [Google Scholar]

18. de Cuyper N, de Witte H. Autonomy and workload among temporary workers: Their effects on job satisfaction, organizational commitment, life satisfaction, and self-rated performance. Int J Stress Manag. 2006;13(4):441-459.

    [Crossref] [Google Scholar]

19. Lee S, Lee SY. Assessing the effect of job stress and social support on job satisfaction: Analysis on Korean police officers. University of Wisconsin-Madison. 2013.

    [Google Scholar]

20. Yamane T. Statistics: An introductory analysis-3. Harper and Row. 1973.

    [Google Scholar]