Compassion in an organizational Context is described as an emotional response to others suffering, accompanied by a sincere desire to help alleviate it (Simpson et al., 2014) . Compassion plays a critical role in driving organizational success by enhancing essential human skills and emotional intelligence in the workplace. Compassion supports the development of positive emotions such as gratitude, empathy, and pride, and strengthens employees’ emotional connection and commitment to their organization (Dutton et al., 2006) . This support not only impacts health outcomes, but also fosters authentic behavior, empathy, and stronger interpersonal relationships in the workplace.

Research by (Lilius et al., 2008) shows that an organizational culture that fosters compassion plays a role in strengthening employee emotional attachment, improving healthy social relationships, and creating a positive and productive work environment. (Guinot et al., 2020) also explains that compassion can strengthen emotional attachment, build trust between employees, and increase loyalty to the organization. In addition, compassion plays an important and significant role in employee retention. A loving relationship needs to be built in the organization, because compassion socialized by managers to employees can minimize turnover and make employees feel at home in the workplace (Sabil et al., 2024) .

Work-life balance (WLB) refers to an individual's ability to balance the demands of work and personal life, such as family, health, and leisure time. According to Murdiana et al. (2024) WLB is an individual's perception that work and non-work activities can be in harmony and support personal development according to the life priorities being lived. This definition is reinforced by subsequent studies that emphasize that WLB is not just a fair division of time, but also includes emotional balance, role satisfaction, and support from the work environment (Dousin et al., 2019).

Female employees working in factories often face challenges in achieving work-life balance because they have to play dual roles as workers and managers of domestic responsibilities. This is reinforced by a study by Adisa and Adekoya (2021) which shows that Gen Z women in shoe factories often experience physical and emotional stress due to lack of work flexibility and minimal institutional support, such as childcare services or adequate leave. In such conditions, an imbalance between work and personal life demands can lead to stress, fatigue, and decreased productivity. Therefore, the implementation of flexible work policies, gender- friendly work environments, and psychosocial support are essential to help women in the manufacturing sector maintain their work-life balance.

Workload is the total tasks or responsibilities that a person must complete within a certain time limit, both physically, mentally, and emotionally (Watson et al., 1996) ; (Wickens, 2008) . According to (Minata Ardi et al., 2024) , excessive or unbalanced workload can cause work stress, reduce job satisfaction, and have a negative impact on productivity. Meanwhile, (Depari, 2024) explains that workload occurs when there is a mismatch between an individual's capacity and the demands of the tasks given, which can cause disruption to work performance. In addition, research across sectors shows that high workload is positively correlated with turnover intention among Gen Z, where burnout is the main mediator. In the context of Gen Z women Z, the presence of domestic pressure also adds to their psychological burden, making the need for strategic and gender-sensitive workload management crucial to avoid extreme fatigue and high rates of early resignation (Subagyo et al., 2024).

However, workload does not always have a negative impact. Research conducted by (Suprapto et al., 2024) shows that workload has a positive relationship with employee retention. This means that when the workload is managed well according to individual capacity and with adequate organizational support, this can actually increase employee loyalty and motivation to stay in the company.

Employee well-being is a comprehensive condition that reflects the positive physical, mental, and social state of workers, which is influenced by the work environment, workload, interpersonal relationships, and support from the organization (Grawitch & Kohler, 2014) . Well-being does not only include physical aspects, but also psychological conditions such as work involvement, satisfaction, and perceived meaning in work. Grant et al. (2007) explains that well-being at work is closely related to positive psychological aspects, which ultimately contribute directly to increased performance and productivity.

In the context of a labor-intensive industry such as a shoe factory, well- being becomes a crucial issue, especially for Gen Z female employees who often carry dual responsibilities. both in professional and domestic roles. The pressure of this dual role increases vulnerability to stress, burnout, and conflict between work and personal life. This is supported by findings from (Allen et al., 2000) which state that conflict between work and family is more often experienced by women because of the dual roles they hold, thus potentially causing chronic stress and having a negative impact on psychological well- being. As a result, paying attention to welfare is no longer just a personal matter, but has become a strategic need for companies.

Doef et al. (2010) emphasizes that a balance of moderate work demands, high levels of control, and social support from the workplace will improve psychological well-being and minimize stress levels. This approach is in line with the needs of Gen Z women who tend to prioritize work-life balance, healthy interpersonal relationships, and inclusive and supportive work values.

Compassion and Well-being of Gen Z Female Employees

The level of compassion in the workplace, namely genuine concern and concrete action from colleagues and superiors when someone faces difficulties, is predicted to increase the well-being of female employees Gen Z. This emotional support fosters a sense of psychological safety and strengthens an inclusive work climate, something that is highly valued by the digital-native generation (Dutton et al., 2014). In addition, compassion also strengthens emotional attachment to the organization, which according to Lilius et al. (2008), is key to building a healthy and supportive work environment.

H1: The compassion variable has a positive and significant effect on the welfare of female Gen employees Z

work-life balance (WLB) is a condition where the demands of work and personal life are balanced and is considered a major determinant of mental health, life satisfaction and work motivation of female employees Z. This generation tends to reject the “always on” work culture and prefers flexibility and space for self-development outside the office (Deshmukh, 2016) . Organizations that provide equitable time arrangements, such as hybrid work or flexible schedules, report lower stress, improved sleep quality, and greater social ties (Sirgy & Lee, 2023). Rantanen et al., (2011) also found that high work-life balance strengthens psychological resilience and prevents emotional exhaustion, especially in the challenging early stages of a career.

H2: The work-life balance variable also has a positive and significant influence on well-being. female employees Gen Z

On the other hand, excessive workload in terms of both task volume and deadlines is projected to damage the well-being of female Gen employees. Z. This pressure can trigger chronic stress, physical and mental exhaustion, and decreased life satisfaction (Kulikowski, 2017) . In the context of remote work that blurs the boundaries between office and home, unmanageable workloads increase the risk of role conflict and erode energy recovery time (Woerkom & Warren, 2021) . Moreover, the double pressure of high expectations at work and lingering domestic responsibilities makes Gen women Z is more vulnerable to burnout and long-term mental health disorders (Swanberg et al., 2020) .

H3: The workload variable has a negative and significant effect on well- being female employees Gen Z

Figure 1. Theoritical Framework

Validity indicates the extent to which an instrument is able to measure the intended construct accurately. According to (Sartori & Pasini, 2007) , validity reflects the suitability between the measuring instrument and the theoretical concept it represents. In this study, validity was tested using the Pearson Product Moment technique, by comparing the calculated r and table r values at a significance level of 5%. An item is declared valid if the calculated r > table r (0.155), which means that the item contributes to the measured construct.

Table 2. Validity Test

SPSS 25 output results

Based on the validity test results, all items in the variables attachment, work-life balance, workload, and wellbeing of female Gen Z employees had a computed r value larger than the r table. This means that each item in each of these variables is legitimate and appropriate for use in research.

Reliability measures the internal consistency of an instrument the extent to which items -in a variable produce stable and interrelated results. According to (Edelsbrunner et al., 2025) , Cronbach's Alpha ≥ 0.70 is considered to indicate reliability Which good in research base. Meanwhile (Cortina, 1993) explains that Alpha measures the uniformity between items and is commonly used in the evaluation of quantitative instruments .

Table 3. Reliability Test

SPSS 25 output results

In this study, reliability tests were performed on each variable, which included work supervision, competence, motivation, and work discipline. The test findings showed that all variables had a Cronbach's Alpha value more than 0.70, indicating that all instruments were reliable. This demonstrates that the items in each variable are very consistent and appropriate for measuring the target construct.

Normality is checked to see if the regression residual data is regularly distributed, which is a key assumption for accurate model estimation. According to (Mishra et al., 2019) , the Kolmogorov–Smirnov test can be used as a goodness -of fit test for a normal distribution whose parameters (mean and variance) have been estimated . Data is considered normal if the significance value of p-value> 0.05 which means there is no evidence of rejection of the normality hypothesis.

Table 4. Normality Test

One-Sample Kolmogorov-Smirnov Test

The test findings yielded a significance value of 0.200, which is greater than the threshold of 0.05. As a result, we can conclude that the residual data is normally distributed and fits the basic linear regression assumptions.

Multicollinearity is tested by looking at the Tolerance and VIF values: Tolerance > 0.10 and VIF < 10 indicate that the independent variables are not highly correlated. As explained by Garc (2020) , multicollinearity can cause the coefficient variance to increase, the coefficient sign to be wrong, or the model interpretation to be unreliable . In this study, all variables meet these criteria, as a result, it is possible to conclude that the regression model does not exhibit any multicollinearity.

Table 5. Multicollinearity Test

SPSS 25 output results

Based on the test results:

Since all variables show Tolerance values > 0.10 and VIF < 10, it can be concluded that there is no multicollinearity in this regression model, and the data meets the classical assumptions regarding multicollinearity.

The F test is used to assess whether the independent factors simultaneously have a significant effect on the dependent variable. In this study, the independent variables tested were affection, work balance, and workload, while the dependent variable was well-being.

Table 6. Simultaneous Hypothesis Test Results

SPSS 25 output results

Based on the results of the F test, the calculated F value was 140.844 with a significance level of 0.000. Because the significance level is less than 0.05, it may be argued that the regression model utilized is also significant. Thus, it can be concluded that the variables of attachment, work balance, and workload have a major impact on the well-being of female generation Z employees.

The t-test is used to test how far the influence of one independent variable is. individual in explaining the variation of the dependent variable (Ronald A. Fisher, 1992) . If the significance value <0.05, then the variable is considered to have a significant partial influence on the dependent variable. The results of the SPSS 25 calculation are as shown in table 7 below.

Table 7. Multiple Linear Regression Test Coefficients^a

Output: SPSS 25

Multiple linear regression tests show that affection, work-life balance, and workload have a significant impact on employee welfare. The regression model obtained is:

Wellbeing = 14.442 + 0.293(Compassion) + 0.289(Work-Life Balance) – 0.344(Workload).

These findings suggest that affection and work-life balance have a beneficial effect on well-being, implying that the more affection and balance employees experience between work and personal life, the better their well- being. In contrast, workload has a negative influence, meaning that the higher the workload, the lower the level of employee well-being. All factors in this model are statistically significant (p-value < 0.000), implying that attachment, work-life balance, and workload all have an impact on employee well-being.

The determination test is used to examine how well the regression model can explain the variation of the dependent variable caused by the independent variable. The R Square (R²) value represents the proportion of the dependent variable's variance that the model can explain.

Table 8. Determination Test

Output: SPSS 25

According to the table output above, the R Square value is 0.730, indicating that attachment, work-life balance, and workload account for 73% of changes or fluctuations in employee welfare variables. The remaining 27% is influenced by factors other than this model. The Adjusted R Square score of 0.725 reflects the results after adjusting for the number of independent variables and samples, indicating that the model has excellent explanatory power. As a result, this regression model is quite good at predicting employee welfare based on the three independent variables.