An Analysis of Work Stress and Family-Conflict Work on Organizational Commitment and Turnover Intention: A Case Study on Employee of PT. Apparel One Indonesia

This study departs from findings that show that the turnover rate of employees at PT. Apparel One Indonesia in 2018 is 3.715% which exceeds the threshold set by the company by 2% each year, and the findings that the variables that explain the phenomenon of turnover intention are still different which then leads to the formulation of the problem regarding "how to reduce turnover". Based on these problems, this study developed a research model that was empirically tested to reexplain the variables that could affect turnover intention by using data obtained from interviews using a questionnaire to PT. Apparel One Indonesia's employees. The test was conducted using the Structural Equation Modeling (SEM) approach. The test results showed that work stress and family conflict had a significant adverse effect on organizational commitment. This study also shows that work stress and work-family conflict had a significant positive effect on turnover intention. In contrast, organizational commitment had a significant negative effect on turnover intention.


A. Introduction
Human resources are one of the crucial factors and aspects that determine the direction of a company's progress. The existence of reliable and quality human resources will improve the company's performance. Therefore, the company must be able to manage human resources in the company well to compete with other companies. PT. Apparel One Indonesia is a textile production company located in Semarang, Central Java. PT. Apparel One Indonesia has 2,673 employees as of December 2018. Based on data from HRD of PT. Apparel Semarang, the average turnover rate at PT. Apparel One Indonesia in 2018 was 3.71%. This has exceeded the maximum employee turnover limit set by PT. Apparel One Indonesia by 2% per year.
Turnover in a company means that the company will lose the talents and abilities of its employees, so the company must be able to find replacements for employees who leave the company, where this will require new time and costs in recruiting new employees (Hartono & Setiawan, 2018).
(2014) is an employee's plan in a company to intend to leave the job or fire the employee. Furthermore, the expressed turnover intention can be categorized into two groups, namely voluntary turnover and involuntary turnover.
Voluntary and forced turnover require reimbursement costs related to recruitment and selection (Hartono & Setiawan, 2018).
Turnover intention can be caused by several factors, one of which is employee job satisfaction will be influenced by several factors, one of which is work-family conflict. Work-family conflict is a condition of a person who has difficulty dividing his role in work and the family or vice versa (Latifah & Rohman, 2014). This can occur because of the demands of a difficult time or too many roles to do, so it can cause stress.
Job stress is one of the factors that can lead to turnover. Work stress can be interpreted as a negative feeling due to the inability of a person or individual to deal with a workload that exceeds his capacity or facing various other pressures from the workplace (Basri, 2012).
Several previous studies examined in this study found that the problem of turnover intention that occurred in the company was caused by organizational commitment. Someone who has a high commitment will have an identification of the organization, the leader is faced with a commitment to entrust the duties and responsibilities to subordinates. Employees who have a solid affective commitment will stay with the organization because they want to stay, and employees who have a solid continuous commitment because they have to stay with the organization. Employees' perceptions of high organizational commitment will result in low employee intention to leave.
Previous studies related to the effect of organizational commitment on turnover intention have been conducted by Nasution (2017); Utama and Sintaasih (2015), which show that organizational commitment has a significant negative effect on turnover intention. However, not so, the results of a study conducted by Firdaus (2017), Wahyuni, Zaika, and Anwar (2014) showed a significant positive effect of commitment to turnover intention.
The mapping results in previous research conducted in this study also found that work stress and work-family conflict influence organizational commitment and turnover intention. However, the results of these studies show different findings. The following previous studies examined in this study show the interplay between variables and research gaps that are still found in empirical testing between these variables.

B. Method
This research is explanatory research with a causality approach, namely research that wants to find an explanation in the form of a causeeffect relationship between several concepts or some variables or strategies developed in management (Ferdinand, 2006). The population is the whole object of research (Arikunto, 2009). The population in this study was all PT. Apparel One Indonesia employees, amounting to 2028 people.
In contrast, the sample is a portion of the population with relatively the same characteristics and is considered representative of the population (Singarimbun, 1991). The sample was obtained from the slovin formula with 10% of alpha. The sampling method used simple random sampling.
The researcher randomly chose the respondent, and based calculation sample used in this study were 170 employees.

512}
The data collection used is the census method in which the questionnaire is distributed to all populations that are the object of research through direct interview techniques to the employees of PT. Apparel One Indonesia in the hope that they will give a positive response to the list of questions asked (Hussein, 1999).
SEM (Structural Equation Modeling) analysis with the AMOS (Moment Structure Analysis) program is a hypothesis test proposed through an analysis technique. SEM modelling can present answers to dimensional and regressive questions, wherein dimensionally; conceptual indicators can be measured. At the same time, the regressive influence between factors identified can also be measured, so the degree of relationship between factors is presented.
Whereas the description of causality relationships between constructs in the model built by researchers through structural equations is useful to follow the following guidelines: Endogen variable = eksogen variable + endogen variable + error η1 = λ1. Conformity and a statistical test determine whether modelling can be accepted or rejected. The validity index and the test cut-off test apply.

C. Result and Discussion
Description of Respondent Characteristics is carried out to provide an overview of the characteristics of PT. The following is a description of the respondent's characteristics. The descriptive analysis of the sex of the respondents showed that the majority of respondents, namely some 144 people (84.7%), were women. This is to the characteristics of the work carried out at PT Apparel One Indonesia, a garment company that requires precision and accuracy.

514}
Descriptive analysis results showed that most respondents were in the age range ≤ 30 years, with 137 people (80.6%). This shows that the employee respondents at PT Apparel One Indonesia are at a productive age, so it is expected to produce optimal performance. Descriptive analysis results showed that as many as 132 people (77.6%) had worked at PT Apparel One Indonesia for 1-3 years. Referring to the data presented in the table above, it can be seen that the majority of respondents' education is high school, with a total of 133 people (78.2%).
The description of the research variables was carried out to obtain a description of the respondents' answers regarding the variables studied in this study which were carried out using the index analysis technique.
Because the scoring technique used in this study is a minimum of 1 and a maximum of 10, the calculation of the index for respondents' answers is carried out using the following formula (Ferdinand, 2006): Where: F1 = frequency of respondents who answered 1 F2 = frequency of respondents who answered 2 Etc., F7 = frequency of respondents who answered 7 By using the three box method criteria, the basis for interpreting the index value is as follows: 14.28 -42.86 = low > 42.86 -71.43 = moderate > 71.43 -100.00 = high The results of calculating the index value for each research variable are described below.
The following is a description of the respondents' answers to the indicators of work stress variables measured using indicators that include restlessness/anxiety, irritability, fatigue, lack of focus, and many work mistakes. The results of the calculation of the index value show that the average index value for the work stress variable is 57.3, which means that the work stress experienced by respondents is perceived as moderate. Of the five indicators of working stress, the X3 indicator of fatigue has the highest index of 60.7. This means that it is easy to become an indicator that shows the work stress experienced by many respondents.
The following is a description of the respondents' answers to the work-family conflict variable indicators, which are measured using indicators The results of the calculation of the index value show that the average index value for the turnover intention variable is 54.4, which means that the respondent's turnover intention is perceived as moderate.
Of the five indicators of turnover intention, indicator X13 about the tendency of individuals to think about leaving the organization is the indicator with the highest index, which means that respondents often think of leaving the company. From the model proposed and related to the data, it will be known how the causal relationship between work stress, work-family conclusions, organizational commitment, and turnover intention. The results of the processing of the proposed model are described below.

An Analysis of Work Stress and Family-Conflict Work on Organizational Achmad Junaidi, Febri Elsa Adelia & Yoga Nindhita
JIP-The International Journal of Social Sciences {517

Figure 1. SEM Test Results on the Research Model
A goodness-of-fit test was performed to determine the model's accuracy with research data. The index of test results is compared with the critical value to determine whether or not the model is summarized in the following table. After testing the SEM assumptions and the suitability of the model (model fit), the hypothesis testing the causality relationship of the research variables is then performed. Hypothesis test results for the relationship between variables are shown from the value of Regression Weight in the CR column (identical to the t-count) which is compared with the critical value (identical to the t-table). The critical value for the significance level  -table). If the CR value> critical value, then the research hypothesis will be accepted; conversely, the research is rejected if the CR value <critical value. The CR capability ratio is used to summarize the estimated spread of the system compared to the spread of the specification limits (upper and lower). The lower the Cr value, the smaller the output spread.
The estimated parameter for testing the effect of work stress on organizational commitment shows a CR value of -2.012 with a probability of 0.044. Because the CR value generated from the calculation is more significant than the critical value at the 0.05 (5%) significance level of 1.998 and the resulting probability value (0.044) is <0.05, it can be concluded that the work stress variable is statistically proven to have a negative effect and significant to organizational commitment. Employees experience work-family conflict due to long working hours, overwork, inflexible work operations, and tight banking policies regarding holidays and office time (Moqsood et al., 2012). Work-family conflicts can affect organizational commitment before the employee finally leaves the job. However, a direct relationship shows that employees will continue to enjoy their work despite experiencing workfamily conflict. However, because family pressure is strong enough, it can force them to quit their jobs (Agustina, 2008).
The estimated parameter for testing the effect of work stress on turnover intention shows a CR value of 2.413 with a probability of 0.016.
Because the CR value generated from the calculation is more significant than the critical value at the 0.05 (5%) significance level of 1.998 and the resulting probability value (0.016) is <0.05, it can be concluded that the work stress variable is statistically proven to have a positive effect and significant on turnover intention. Research conducted by Junaidi et al.
(2020) examines the effects of overtime, work stress, and workload on turnover intention. The results of his research stated that the variable work stress has a significant positive effect on turnover intention.
Work stress is a psychological response felt by employees by the exposure that exceeds the limits of their ability as an employee. Exposure to work stress that occurs continuously makes employees need to respond. One response that arises when employees do not want to be continually exposed to work stress is the emergence of a desire or interest to resign from a job perceived by employees to cause work stress.
Previous researchers have widely studied the influence of work stress on turnover intention, even though the results have not been conclusive. The following summarises the results of previous studies that have been reviewed in this study.
Work-family conflicts can occur in men and women, but several studies have shown that women have a more considerable intensity of work-family conflicts (Apperson et al., 2002). This conflict level will worsen if women work formally because they will be bound by organizational rules, which include working hours, assignments, and targets in completing tasks. This work-family conflict will be felt even more if female employees are married because the traditional role that has been unavoidable is taking care of the household and their children  Widyaningrum, 2013). Work-family conflict pressure that occurs continuously will trigger women to resign.
The estimated parameter for testing the effect of organizational commitment on turnover intention shows a CR value of -2.245 with a probability of 0.025. Because the CR value generated from the calculation is more significant than the critical value at the 0.05 (5%) significance level of 1.998 and the resulting probability value (0.025) is <0.05, it can be concluded that the organizational commitment variable is statistically proven to have a negative effect and significant on turnover intention. Research conducted by Chalim, AS (2018) with the study results stated that organizational commitment has a negative and significant effect on turnover intention. It means that turnover intention will decrease if organizational commitment increases. Conversely, the turnover intention will increase if organizational commitment decreases.
Someone who has a high commitment will have an identification of the organization, the leader is faced with a commitment to entrust the duties and responsibilities to subordinates. Employees who have a solid affective commitment will stay with the organization. They want to stay (because of the want to), employees who have a solid continuous commitment because they have to stay with the organization. Employees' perceptions of high organizational commitment will result in low employee intention to leave. Vice versa, when employees lack organizational commitment, the intention to leave will be higher (Tumbelaka et al., 2016).

D. Conclusion
This study departs from the findings which show that the turnover rate of employees at PT. Apparel One Indonesia in 2018 is 3.715% which exceeds the threshold set by the company by 2% each year, and the findings that the variables that explain the phenomenon of turnover intention are still different which then leads to the formulation of the problem regarding "how to reduce turnover". Regarding "how to reduce turnover", developing a research model tested empirically to re-explain the variables that can

JIP-The International Journal of Social Sciences {521
affect turnover intention obtained the following findings: 1) Job stress has a significant negative effect on organizational commitment. If the work stress experienced by employees increases, it will cause a decrease in organizational commitment. 2) Work-family conflict has been shown to have a significant adverse effect on organizational commitment. If the work-family conflict experienced by employees increases, it will cause a decrease in organizational commitment.
3) The study results found that job stress had a significant positive effect on turnover intention. That is, if the work stress experienced by employees increases, it will cause an increase in turnover intention. 4) The study results found that work-family conflict proved to have a significant positive effect on turnover intention. If the work-family conflict experienced by employees increases, it will cause an increase in turnover intention. 5) The study results found that organizational commitment proved to have a significant adverse effect on turnover intention. If the organizational commitment experienced by employees increases, it will cause a decrease in turnover intention.
From the research results obtained, to overcome a reasonably high turnover rate, various actions can be taken, such as increasing employees' organizational commitment by reducing work stress from employees, which can be done through recreation, reducing workload, etc. In addition, increasing organizational commitment can also be done by reducing workfamily conflicts experienced by employees. The company can do this by monitoring and understanding the conditions experienced by employees.
The exogenous variables studied did not separate internal and external factors. Therefore, developing a more comprehensive research model involving external and internal factors that can explain turnover intention is necessary.