Analysis of The Effect of Workload, Role Conflict, Work Stress on Exit Intentions and Work Burnout

Arwansa Wahana¹, Iwan Kresna Setiadi², Sri Mulyantini³

^1,2,3University of National Development Veterans Jakarta, DKI Jakarta, Indonesia

Email: [email protected]¹,[email protected]², [email protected]³

Abstract

This quantitative research aimed to assess how workload, role conflict, and work stress impact turnover intentions and work fatigue among employees of Bank Syariah X in the Jabodetabek area. Using a sample of 75 employees selected through simple random sampling and data collected via questionnaire surveys, analysis was conducted employing Partial Least Square (PLS) analysis. The study found that both workload and role conflict significantly contribute to work stress, subsequently influencing turnover intentions and job burnout. These findings emphasize the importance of addressing these issues to reduce turnover rates and improve employee well-being. The results offer valuable insights for academia to expand research literature, researchers to advance knowledge, and Bank Syariah X management to develop effective workforce management policies and practices.

Keywords:

Job Burnout, Role Of Conflict, Turnover Intention, Work Stress, Workload.

INTRODUCTION

One of the challenges faced by the banking sector, including Bank Syariah X, is the high employee turnover rate. Compared to other sectors, the banking sector has a higher turnover rate. The survey conducted by Tower Watson from 2012 to 2014 showed that the turnover rate in the banking industry continued to increase year on year, reaching 13.92% in 2014 (Fan & Wang, 2022). Similarly, a 2015 salary survey by Mercer Talent Consulting & Information Solution noted that the banking sector had the highest talent turnover rate, reaching 16% (DANG, 2021). Bank Syariah X, one of the banks in Indonesia, also faces similar problems, especially in branch offices in the Greater Jakarta area, with employee turnover rates reaching more than 10% (Sartika & Akhmad, 2023).

High turnover rates are primarily caused by employees' intentions to leave their current positions and workplace burnout. Burnout is influenced by two main factors: job demands and inadequate resources. "Job demands" refer to the aspects of a job that require physical or mental effort (Thangal et al., 2022). Conversely, "job resources" are the physical or psychological factors that help employees achieve their goals while reducing stress and burnout ((Surachartkumtonkun et al., 2023);(Bakker & de Vries, 2021)). Based on the research of other scholars, job satisfaction plays a vital role in both attracting and retaining a highly skilled workforce. Previous scholarly viewpoints suggest a strong connection between employee satisfaction, customer satisfaction, and overall corporate performance (Stamolampros et al., 2019).

Although unhappiness at work can be a cause, it is not always the case. A pertinent inquiry arises regarding whether unhappiness can merely be considered the opposite of happiness, given the cautioned observation regarding the relative independence of positive and negative affect (Brzozowski & Coniglio, 2021). To illustrate, the lack of paid employment showed a more pronounced correlation with a decline in positive emotions compared to an uptick in psychological symptoms like depression and anxiety (Blanchflower, 2020). Many other factors influence an employee's decision to leave an organization, such as better career opportunities, higher rewards, or more attractive benefits ((Mahadi et al., 2020);(Ali & Anwar, 2021);(Al-Suraihi et al., 2021)). Excessive workload, role conflict, and work stress are factors that influence employee exit intentions and work burnout. Discomfort, suboptimal contribution, and decreased work motivation are the impacts that can arise from these factors (Salama et al., 2022).

The aim of this study is to examine how workload, role conflict, and work stress impact employees' intentions to leave and their experience of work fatigue at Bank Syariah X. Additionally, it is anticipated that this research will offer various benefits, including enriching the research literature in Indonesia for academics, advancing knowledge development for researchers, and providing Bank Syariah X management with insights to enhance workforce management policies and practices, ultimately leading to positive contributions to the organization.

RESEARCH METHODS

The research method used in this study is Partial Least Square (PLS). The population of this study is organic employees of Bank Syariah X who serve in supervising branch offices in Jakarta, Bogor, Depok, Tangerang, and Bekasi. Sampling was done using a simple random sampling technique, where 75 respondents were randomly selected from the population.

Data for this study were collected through questionnaires containing questions related to the variables studied, namely workload, role conflict, work stress, exit intentions, and work fatigue. The questionnaire is designed on the basis of pre-established structural models and measurement models.

Data analysis is carried out using the Partial Least Square (PLS) method, which is a powerful analysis method because it does not require certain assumptions about the data, both in terms of measurement scale and the number of samples required. PLS can cope with data with relatively small samples and is not tied to a specific distribution.

The analysis process begins with designing structural models and measurement models. Path diagrams are used to illustrate relationships between variables and the conversion of path diagrams to systems of equations is done to formulate mathematical models to be estimated. Estimation is done using the Smart PLS program with a resampling method using Bootstrapping.

A measurement model fit test is performed to check convergent validity and discriminant validity, while a goodness of fit evaluation is performed to measure structural model fit. Hypothesis testing is carried out using t statistics, where the significance of the probability value determines whether the hypothesis is accepted or rejected. Through employing this methodology, it is anticipated that this study will offer enhanced insights into the determinants of exit intentions and work fatigue among Bank Syariah X employees. Furthermore, it aims to provide relevant policy suggestions to address these challenges effectively.

RESULTS AND DISCUSSION

Convergent Validity Test

The initial step involves assessing whether the model demonstrates convergent validity, which examines whether the loading factor of each indicator for the constructs meets the required criteria. Generally, an indicator is considered valid if it exhibits a correlation value above 0.7, although during the research development phase, loading scales between 0.50 to 0.60 are still deemed acceptable (Purwanto, 2021). In this study, all instruments corresponding to workload, role conflicts, work stress, exit intentions, and work fatigue have been deemed qualified and valid, as each correlation value demonstrates a loading factor exceeding 0.50. The smallest loading factor is observed in the BK18 statement instrument at 0.629, while the largest loading factor is found in the IK12 statement instrument at 0.938. This indicates that the statement instruments for the utilized indicators in this study are valid and meet the requirements of the convergent validity test. Detailed numerical descriptions are provided in the appendix alongside the convergent validity test table.

Top of Form

Discriminant Validity Test

To assess discriminant validity, the square root value of the average variance extracted (AVE) is examined to determine the validity of each indicator in the study. The recommended threshold is typically above 0.5. According to the output from Smart-PLS 3.0 software in this study:

Table 1. Average Variance Extracted (AVE)

Variable	AVE
Workload	0,623
Exit Intentions	0,789
Work Fatigue	0,789
Role Conflict	0,617
Work Stress	0,718

The provided table indicates that all constructs within the research model have AVE values exceeding 0.50. Specifically, the workload variable has a value of 0.623, exit intention variable stands at 0.789, work fatigue variable is also 0.789, role conflict variable measures at 0.617, and work stress variable is 0.718. With these outcomes, it can be concluded that the questionnaire utilized to measure workload, exit intention, work fatigue, role conflict, and work stress is considered valid as a measurement tool for these variables.

PLS Reliability Test

After confirming the validity of each variable question instrument, the next step involves conducting reliability testing. This test assesses the reliability of the variables using the Composite Reliability and Cronbach's Alpha values obtained from the Smart-PLS 3.0 software output:

Table 2. Composite Reliability dan Cronbachï¿½s Alpha

Variable	Composite Reliability	Cronbachï¿½s Alpha
Workload	0,967	0,964
Exit Intentions	0,989	0,989
Work Fatigue	0,987	0,986
Role Conflict	0,965	0,961
Work Stress	0,976	0,974

The provided table illustrates that both Cronbach's Composite reliability and Alpha values for all constructs surpass 0.7, indicating that all constructs in the estimated model meet the required criteria for reliability. Specifically, the workload variable exhibits a composite reliability value of 0.967 and a Cronbach's alpha value of 0.964. Similarly, the exit intention variable displays a composite reliability value of 0.989 and a Cronbach's alpha value of 0.989. Moreover, the work fatigue variable demonstrates a composite reliability value of 0.987 and a Cronbach's alpha value of 0.986, while the role conflict variable shows a composite reliability value of 0.965 and a Cronbach's alpha value of 0.961. These results indicate that all variables possess excellent reliability within each construct, meeting the requirements as per Table 8. Reliability testing affirms that the instruments utilized in the research for data collection can be relied upon to provide accurate information and insights from the field.

Model Struktural (Inner Model)

The model was assessed by examining the values of R-Square, QSquare, path analysis coefficients (Path Coefficients), and t-statistic.

Original Sample Value measurement results

The relationship between the independent and dependent variables is represented by a line indicating the influence between the variables. This line of influence constitutes the inner model, illustrating the magnitude of influence from the independent variable to the dependent variable.

Table 3. Original Sample Value Table

Relationship Between Variable	Original Sample
Workload -> Work Stress	0,484
Role Conflict -> Work Stress	0,852
Work Stress -> Intention to Exit	0,461
Work Stress -> Work Fatigue	0,438

Sumber: Hasil Output Smart-PLS, 2019.

The table indicates positive relationships between various variables. Firstly, there is a positive correlation between the Workload variable and the Work Stress variable, with an Original Sample value of 0.484. This suggests that when workload increases, work stress also tends to increase, and conversely, when workload decreases, work stress tends to decrease. Similarly, a positive relationship is observed between the Role Conflict variable and the Work Stress variable, with an Original Sample value of 0.852. This implies that higher levels of role conflict are associated with higher levels of work stress, while lower levels of role conflict correspond to lower levels of work stress. Furthermore, there exists a positive correlation between the Work Stress variable and the Exit Intention variable, with an Original Sample value of 0.461. This suggests that higher levels of work stress are linked to higher intentions to leave the job, and lower levels of work stress are associated with lower intentions to leave.

Finally, a positive relationship is found between the Work Stress variable and the Work Fatigue variable, with an Original Sample value of 0.438. This indicates that higher levels of work stress tend to result in higher levels of work fatigue, while lower levels of work stress lead to lower levels of work fatigue.

Sample Mean Measurement Results

The mean sample values in this study indicate a robust relationship between the workload variable and the work stress variable, the role conflict variable and the work stress variable, the work stress variable and the exit intention variable, as well as the work stress variable and the work fatigue variable. This strong influence relationship is evidenced by positive sample mean results.

Table 4. Sample Mean Values

Relationships Between Variables	Sample Mean
Workload -> Work Stress	0,700
Role Conflict -> Work Stress	0,825
Work Stress -> Intention to Exit	0,453
Work Stress -> Work Fatigue	0,434

Sumber: Hasil Output Smart-PLS, 2019.

R Square Value Measurement Results

Once all statement items per variable are deemed valid and all variables are confirmed reliable, the subsequent step involves testing the structural model of the research through the R Square test. The output results from Smart-PLS 3.0 software pertaining to the R Square test are as follows:

Table 5. Square R-Value Table

Variable	R Square
Exit Intentions	0,213
Work Fatigue	0,192
Work Stress	0,785

Sumber: Hasil Output PLS, 2019.

According to the table provided, the R Square (R2) value for exit intentions is 0.213, indicating that work stress accounts for 21.3% of the variance in exit intentions. The remaining 78.7% of the variance is influenced by other factors or variables. Similarly, the R Square value for work fatigue is 0.192, indicating that work stress explains 19.2% of the variance in work fatigue, with the remaining 80.8% influenced by other variables. Furthermore, the R Square value for work stress is 0.785, suggesting that workload and role conflict collectively explain 78.5% of the variance in work stress, while the remaining 21.5% is influenced by other variables.

Q Square

The Q-Square measure assesses the predictive relevance of structural models, indicating how well observed values are generated by the model and its parameter estimations. A Q-Square value greater than 0 signifies that the model has predictive relevance, while a value less than or equal to 0 indicates a lack of predictive relevance. The calculation of Q-Square is performed using the formula: ≤ 0 indicates insufficient predictive relevance in the model.

The Q2 value is calculated using the formula: Q2 = 1 ï¿½ (1 ï¿½ Ra12) (1 ï¿½ Ra22) ... (1 - RP2), where R12, R22, ..., RP2 represent the R-square values of the endogenous variables in the equation model. The Q2 value ranges between 0 and 1, where a value closer to 1 indicates a better model fit.

Q2 = 1 ï¿½ (1 ï¿½ R12) (1 ï¿½ R22)

= 1 ï¿½ (1 ï¿½ 0,213) (1 ï¿½ 0,192) (1 ï¿½ 0,785)

= 1 ï¿½ (0,787) (0,808) (0,215)

= 1 ï¿½ 0,0135364

= 0.986

The obtained Q2 result is 0.986, which aligns with the condition that the Q2 value ranges between 0 and 1, where a value closer to 1 indicates a good model fit. Therefore, with 0 < 0.986 < 1, the model is considered good as it is closer to 1.

t-Statistical Test

Once the statement items per variable are confirmed valid, and all variables are deemed reliable, the structural model fitness is established, allowing for the continuation of the research by conducting t-statistical tests. In this study, the t-statistical test, or partial test, was utilized to ascertain whether there exists an influence between workload and role conflict on work stress, as well as the influence of work stress on exit intentions and work fatigue. According to (Wan et al., 2020) , to determine the value of t-table, one should consider a significance level of 0.05 and calculate the t-table value using the formula: t-table = 1.994, obtained from df = N-K or df = 75-5 = 70, associated with an error degree of 5% or 0.05.

Based on the results of data processing for significance tests (t-test), the following outcomes were obtained:

Table 6. T-Statistical Test Results

Relationships Between Variables	T Statistics	P Values
Workload -> Work Stress	4,582	0,002
Role Conflict -> Work Stress	10,097	0,000
Work Stress -> Intention to Exit	3,249	0,001
Work Stress -> Work Fatigue	3,010	0,003

Sumber: Hasil Output PLS, 2019.

Workload and Work Stress

In this study, it was found that Workload had a significant influence on Work Stress, evidenced by a T-statistic value of 4.582. This resulted in a P-Value of 0.002, which is considerably smaller than the alpha value of 0.05 at a 95% confidence level. These findings suggest that as the workload increases, work stress among Bank Syariah X employees in the Jabodetabek area is likely to increase as well, and conversely, when the workload decreases, work stress tends to decrease. The original sample value of 0.484 further confirms this relationship, indicating a positive correlation between workload and job stress.

Role Conflict and Work Stress

In this study, it was found that Role Conflict had a significant influence on Work Stress, as indicated by a T-statistic value of 4.582. This resulted in a P-Value of 0.002, which is considerably smaller than the alpha value of 0.05 at a 95% confidence level. These results imply that as role conflicts increase, the work stress experienced by Bank Syariah X employees in the Jabodetabek area is likely to increase as well, and conversely, when role conflicts decrease, work stress tends to decrease. The original sample value of 0.484 further supports this interpretation, indicating a positive relationship between role conflict and work stress.

Work Stress and Exit Intentions

In this study, it was observed that Work Stress exerted a significant influence on Exit Intentions, as evidenced by a statistical T-value of 4.582. This resulted in a P-Value of 0.002, which is significantly smaller than the alpha value of 0.05 at a 95% confidence level. These findings suggest that as work stress increases, the likelihood of exit intentions among Bank Syariah X employees in the Jabodetabek area also increases, and conversely, when work stress decreases, exit intentions tend to decrease. The original sample value of 0.484 further confirms this relationship, indicating a positive correlation between work stress and exit intentions.

Work Stress and Work Burnout

In this study, it was found that Work Stress had a significant influence on Work Fatigue, with a T-statistic value of 4.582. This resulted in a P-Value of 0.002, which is considerably smaller than the alpha value of 0.05 at a 95% confidence level. These findings suggest that as work stress increases, work fatigue among Bank Syariah X employees in the Jabodetabek area is likely to increase as well, and conversely, when work stress decreases, work fatigue tends to decrease. This interpretation is further supported by the positive original sample value of 0.484, indicating a positive correlation between work stress and work fatigue.

CONCLUSION

Higher workload has a significant and positive impact on work stress, with a coefficient of 0.484, accounting for 3.06% of the variance. This implies that an increase in workload will notably elevate work stress among Bank Syariah X employees in Jabodetabek. Specifically, for every unit increase in workload, work stress is expected to increase by 0.484, assuming other factors remain constant. Likewise, higher role conflict significantly and positively influences work stress, with a coefficient of 0.852, explaining 75.41% of the variance. This indicates that elevated role conflicts substantially contribute to increased work stress among Bank Syariah X employees in Jabodetabek. For every unit increase in role conflict, work stress is anticipated to increase by 0.852, assuming other factors remain unchanged. Moreover, work stress has a significant and positive effect on exit intentions, with a coefficient of 0.461, explaining 21.29% of the variance. This suggests that heightened work stress significantly raises exit intentions among Bank Syariah X employees in Jabodetabek. For every unit increase in work stress, exit intentions are projected to increase by 0.461, assuming other factors remain constant. Additionally, work stress has a positive and significant impact on work fatigue, with a coefficient of 0.438, accounting for 19.19% of the variance. This implies that increased work stress significantly contributes to higher levels of work fatigue among Bank Syariah X employees in Jabodetabek. For every unit increase in work stress, work fatigue is expected to increase by 0.438, assuming other factors remain constant.

REFERENCES

Al-Suraihi, Walid Abdullah, Samikon, Siti Aida, & Ibrahim, Ishaq. (2021). Employee Turnover Causes, Importance and Retention Strategies. European Journal of Business and Management Research (EJBMR).

Ali, Bayad Jamal, & Anwar, Govand. (2021). An empirical study of employeesï¿½ motivation and its influence job satisfaction. Ali, BJ, & Anwar, G.(2021). An Empirical Study of Employeesï¿½ Motivation and Its Influence Job Satisfaction. International Journal of Engineering, Business and Management, 5(2), 21ï¿½30.

Bakker, Arnold B., & de Vries, Juriena D. (2021). Job Demandsï¿½Resources theory and self-regulation: New explanations and remedies for job burnout. Anxiety, Stress, & Coping, 34(1), 1ï¿½21.

Blanchflower, David G. (2020). Unhappiness and age. Journal of Economic Behavior & Organization, 176, 461ï¿½488. https://doi.org/https://doi.org/10.1016/j.jebo.2020.04.022

Brzozowski, Jan, & Coniglio, Nicola. (2021). International migration and the (un) happiness push: Evidence from Polish longitudinal data. International Migration Review, 55(4), 1089ï¿½1120.

DANG, Nhan Truong Thanh. (2021). Talent conceptualisation and talent management practices within the banking sector in Vietnam. RMIT University.

Fan, John Hua, & Wang, Zhenping. (2022). Portable Beta And Total Portfolio Management.

Mahadi, Nomahaza, Woo, Norliyana M. F., Baskaran, Shathees, & Yaakop, Azizu Yadi. (2020). Determinant factors for employee retention: Should I stay. International Journal of Academic Research in Business and Social Sciences, 10(4), 201ï¿½213.

Purwanto, Agus. (2021). Partial least squares structural squation modeling (PLS-SEM) analysis for social and management research: a literature review. Journal of Industrial Engineering & Management Research.

Salama, Wagih, Abdou, Ahmed Hassan, Mohamed, Shaimaa Abo Khanger, & Shehata, Hossam Said. (2022). Impact of work stress and job burnout on turnover intentions among hotel employees. International Journal of Environmental Research and Public Health, 19(15), 9724.

Sartika, Mila, & Akhmad, Farida. (2023). Determinant Factors Analysis of Turnover Intention in Y Generation at Bank Syariah Indonesia, Semarang City. Velocity: Journal of Sharia Finance and Banking, 3(1), 1ï¿½21.

Stamolampros, Panagiotis, Korfiatis, Nikolaos, Chalvatzis, Konstantinos, & Buhalis, Dimitrios. (2019). Job satisfaction and employee turnover determinants in high contact services: Insights from Employeesï¿½Online reviews. Tourism Management, 75, 130ï¿½147. https://doi.org/https://doi.org/10.1016/j.tourman.2019.04.030

Surachartkumtonkun, Jiraporn, Ngo, Liem Viet, & Shao, Wei. (2023). The crisis cloudï¿½s silver linings: The effects of hope and gratitude on employee burnout and engagement. Journal of Retailing and Consumer Services, 74, 103421. https://doi.org/https://doi.org/10.1016/j.jretconser.2023.103421

Thangal, Thahira Bibi T. K. M., Shafie, Nani Ilyana, Yunos, Nurhazirah Mohamad, Ab Hamid, Mohd Halimi, Sidik, Norbaiti, & Azman, Hairul Amiza. (2022). Work Burnout: Unbalanced Job Demands and Job Resources?

Wan, Suxin, Yi, Qingjie, Fan, Shibing, Lv, Jinglong, Zhang, Xianxiang, Guo, Lian, Lang, Chunhui, Xiao, Qing, Xiao, Kaihu, & Yi, Zhengjun. (2020). Characteristics of lymphocyte subsets and cytokines in peripheral blood of 123 hospitalized patients with 2019 novel coronavirus pneumonia (NCP). MedRxiv.

Copyright holder:

Arwansa Wahana¹, Iwan Kresna Setiadi², Sri Mulyantini³(2024)

First publication right:

Journal of Management, Ekonomic and Financial

This article is licensed under: