Work-family enrichment as a mediator effect of supervisor support, self-esteem, and optimism on job sat- isfaction

Article history: Received: February 2, 2020 Received in revised format: March 7 2020 Accepted: March 7, 2020 Available online: March 9, 2020 This study aims to examine the work-family enrichment (WFE) model as influence mediator of supervisor support, self-esteem, and optimism on job satisfaction. The population in this study is 263 permanent employees of Bank BPD DIY, who occupy the positions as staff, and are married, in 7 branch offices in the Special Region of Yogyakarta (DIY). This research uses census method, while data analysis uses AMOS. The results of this study show that work-family enrichment model acts as influencer mediator of supervisor support, self-esteem, and optimism on job satisfaction fit with the empirical data. Hypothesis testing results show that supervisor support implies positive effect on job satisfaction, self-esteem implies positive effect on job satisfaction, and optimism implies positive effect on job satisfaction. Supervisor support also implies positive effect on workfamily enrichment, self-esteem implies positive effect on work-family enrichment, and optimism implies positive effect on work-family enrichment and work-family enrichment implies positive effect on job satisfaction. Work-family enrichment works significantly as influence mediator of supervisor support on job satisfaction, work-family enrichment works significantly as influence mediator of self-esteem on job satisfaction, and work-family enrichment works significantly as influence mediator of optimism on job satisfaction. © 2020 by the authors; licensee Growing Science, Canada


Introduction
Work and family are the two most important domains for most people's lives (Michel et al., 2011). Both of these domains have considerable influence on human life and behavior (Ghafoor, et al., 2014). Although the work and family systems are different, they are interconnected (Clark, 2000). To fulfill the responsibilities both at work and family is a challenge for men and women. This relationship between work and family can be mutually supportive, but it can also interfere with each other (Greenhaus & Beutell, 1985). Research on the relationship between work and family is largely dominated by conflict or negative perspectives (Fung et al., 2014;Marais et al., 2014;Parasuraman & Greenhaus, 2002;Siu et al., 2013;Stoddard and Madsen, 2007). This negative perspective is then called work-family conflict (WFC). The results of the meta-analysis by Allen et al. (2000) show that WFC had negative consequences or outcomes related to work and non-work settings and caused stress. More contemporary view holds that the relationship between work and family is not always negative, but it is also positive (Greenhaus & Powell, 2006). WFE is one of constructions representing how work and family can mutually beneficial. The basic idea of enrichment is that both work and family give resource to individual, such as award, salary, and many other benefits which will be able to help the individual perform better performance in all domain of his life (Burhanudin et al., 2016). Research on the relationship between work and family taken from positive perspective still receives a little attention (Carlson et al., 2011;Fung et al., 2014;Greenhaus & Powell, 2006;Stoddard & Madsen, 2007).
WFE is an important concept to be studied for two reasons. First, theoretically, it would be incomplete to understand the relationship between work and family without considering WFE as a possibility that the roles of work and family could be mutually beneficial. Research on this subject is still relatively neglected. Scholars are therefore calling for a more balanced approach in examining the relationship between work and family, recognizing the positive effects of combining work and family roles (Greenhaus & Powell, 2006). Second, practically, WFE is associated with positive organizational outcomes such as job satisfaction, family satisfaction, and individual well-being. However, compared to WFC, WFE is empirically and conceptually underdeveloped (Carlson et al., 2006). In addition, studies on the relationship between work and family were initially developed in Europe and the United States (Lu et al., 2009;Srivastava & Srivastava, 2014), with different context in developing countries such as Indonesia. Antecedents of WFE can originate from the work environment and aspects of personal disposition or characteristics (Carlson et al., 2006). Antecedents from the work environment include supervisor support. Besides, WFE is also influenced by various dispositional aspects or personal characteristic which is generally less-discussed. Meta-analysis by Allen et al. (2012) found that some personal characteristics such as extraversion, positive affect, internal locus of control, self-esteem, optimism, and self-efficacy can help individuals avoid conflicts in work and family demands. Some of these positive personal characteristics influence individuals to experience enrichment between work and family roles. Considering the lack of attention to the positive side of the relationship between work and family, Allen et al. (2000Allen et al. ( , 2012, and Michel et al. (2011), recommended the importance of examining WFE antecedents originating from the work environment and dispositional aspects or personal characteristics (Baral & Bhargava, 2011b), and their consequences to job satisfaction. To fill the gap of empirical research on WFE, this study aims to test the WFE model as the influence mediator of supervisor support, self-esteem, and optimism on job satisfaction.

Supervisor Support and Job Satisfaction
Supervisor support is an important resource for employees to improve their functions in the organization. An employee who perceives that her supervisor pay attention or care for the employees and provide socio-emotional support will initiate a positive assessment of the environment. Supports to the employees given by the supervisors will help improve job satisfaction among them. Supervisor support is also identified as a factor that reduces the negative consequences of stress in various work contexts. Meanwhile, the lack of supervisor support might increase the rate of burnout and employee dissatisfaction. Supervisor support is an important element for the organization because it relates to job satisfaction and employee intentions to remain working in the organization (Ibrahim, 2012). The results of meta-analysis by Ng and Sorensen (2008) showed that the perception of supervisor support is related to positive work attitudes (job satisfaction, affective commitment, and intention to stay) compared to the perception of colleague support. Ibrahim (2012) argued that the perception of supervisor support had a significant effect on job satisfaction. Kula and Guler (2014) also found that the perception of supervisor support was positively related to job satisfaction. Tang et al. (2014) showed that work support (supervisor support, coworker support, and organizational support) is positively related to job satisfaction. Fung et al. (2015) argued that supervisor support is positively related to job satisfaction. Meanwhile, Gok et al. (2015) found that the perception of supervisor support is positively related to job satisfaction. Therefore, the hypothesis proposed in this study is: H1: supervisor support has a positive effect on job satisfaction.

Self-Esteem and Job Satisfaction
Self-esteem is one of the essential factors that influence how individuals perceive and respond to their lives. Individuals with high self-esteem are controlled by motives for self-improvement. Conversely individuals with low self-esteem are governed by the motive to protect themselves. They tend to expect negative results and tend to psychologically prepare themselves for disappointment. Their dispositional is based on their retrospect on negative experiences, thus they are vulnerable to feelings of negative emotions. Locke et al. (1996) in Judge and Bono (2001) noted that an individual with high self-esteem will view the job as a decent challenge he can master and benefit from. Whereas, an individual with low self-esteem views the job as an inappropriate opportunity and chance of failure. An individual with high self-esteem also has confidence in developing her ability to overcome difficulties. Ahmed (2012) found that self-esteem has a significantly positive correlation with job satisfaction. The results of meta-analysis by Judge and Bono (2001) show that self-esteem was positively related to job satisfaction. Chang et al. (2010) pointed out that self-esteem is related to job satisfaction. Cleare (2013) found that self-esteem has a positive correlation with the overall job satisfaction. Igbal (2012) show that self-esteem had a positive impact on job satisfaction, in which the development in self-esteem will increase job satisfaction. Nurullah (2010) pointed out that self-esteem has a positive relationship with job satisfaction. Therefore, the hypothesis offered in this study is: H2: self-esteem has a positive effect on job satisfaction.

Optimism and Job Satisfaction
Optimism is an individual's tendency to have positive expectations about the future, while pessimism is an individual's tendency to have negative expectations about the future (Ahmed, 2012). An individual with optimistic outlook adjusts better to important life transitions than pessimists. Optimistic individuals tend to hold positive expectations for their future, while a pessimistic individual tends to hold more negative expectations for the future (Scheier et al., 1994). An individual with highlevel of optimism shows greater perseverance during difficult times. In work setting, optimistic employees have higher job satisfaction and greater well-being, and are more likely to perform better in bad situations and avoid dangerous behaviors (Chang et al., 2010). Various studies showed that optimism is positively related to job satisfaction. Ahmed (2012) found that optimism has significantly positive correlation with job satisfaction. Chang et al. (2010) pointed out that optimism is related to job satisfaction. Kaplan and Bickes (2013) also found a positive relationship between optimism and job satisfaction. This study offers the following hypothesis: H3: optimism has a positive effect job satisfaction.

Supervisor Support and WFE
Social support in work setting can be sourced from colleague, supervisor, or the organization. Supervisor support is defined as the general view of the employees in how their supervisors appreciate and pay attention to their well-being (Eisenberger et al., 2002). Social support can be further explained by the theory of COR (conversation of resources) which stated that the society attempts to obtain, maintain, secure, and develop resources; and the threat arises from potential or actual loss of the resources (Hobfoll, 1989). Social support has two main functions: secure the available resource and allow individuals to obtain new resources (Siu et al., 2013). Supervisor support is one essential resource in work setting operates in several ways to facilitate the integration of work and family demands. Several empirical findings have shown positive relation between work support (work resources) and WFE. Supervisor support, colleague support, and organizational support are essential antecedents of WFE, as social support in work setting may ease the pressure in workplace (Tang et al., 2014). Baral and Bhargava (2011a); Lu (2011);and Fung et al. (2015) found that supervisor support has positive relation with WFE. Higher the supervisor support, higher the WFE. The hypothesis offered in this study is: H4: supervisor support has a positive effect on WFE.

Self-Esteem and WFE
According to COR theory, self-esteem is an essential personal trait as the center of well-being in an individual. An individual with high self-esteem is more likely to be motivated to obtain well-being. High self-esteem gives several positive effects, such as buffering towards stress and negative emotions, increase the ability to adapt, and develop healthy psychological condition. An individual with high self-esteem is more likely to be able to cope with anxiety and have healthier condition. Meanwhile, an individual with low self-esteem will experience psychological frigidity, prone to depression, anxiety, and feeling worthless (Rashid et al., 2011). An individual with high self-esteem is more able to resolve conflicts arising from roles in work space and family. She is more likely to see herself as a successful person in managing the balance between work and family lives. Rashid et al. (2011) found that self-esteem has positive relationship with WFE. Higher self-esteem an individual has, the higher the WFE. The offered hypothesis is: H5: self-esteem has a positive effect on WFE.

Optimism and WFE
According to COR theory, optimism is a personal characteristic resource that help individuals deal with stress and enrichment development (Hobfoll, 1989). Optimists are people who expect positive and desirable events in the future, whereas pessimists are people who constantly have negative thoughts and are certain that undesirable events will occur (Luthans et al., 2007). Optimism can be defined as a dispositional (individual traits) filled with hopeful disposition or a tendency to take a favorable view. Optimism can be defined as general expectations that good things will happen, or in other words general expectations for positive results. Optimism is also associated with the nature of extraversion and positive affect, while pessimism is related to neuroticism and negative affect (Mishra, 2013). Optimism is a cognitive construct (expectations about the future), which is also related to motivation. Optimists tend to exert effort, whereas pessimists tend to break away from efforts (Carver and Scheier, 2014). Personality factors often influence how individuals perceive their situation or environment and as a result this perception will affect the outcome of work-family relationships. Several studies have found that dispositional factors, individual differences or personality characteristics are related to WFC or WFE experiences (Ahmad & Ngah, 2011;Allen et al., 2012;Baral & Bhargava, 2011a). Allen et al. (2012) conducted a meta-analysis on the relationship between dispositional variables and WFC, and suggested to examine the relationship between dispositional variables and the positive side of workfamily relationships. Therefore, optimism variable can be relevant predictor for WFE. Washington (2006) pointed out that optimism is positively related to WFE and FWE and negatively related to WFC and FWC. More specifically, the higher one's optimism the higher the WFE. The hypothesis proposed in this study is: H6: optimism has a positive effect on WFE.

WFE and Job Satisfaction
According to Carlson et al. (2014) the relationship between WFE and job satisfaction can be explained in two perspectives: receiving domain view dan originating domain view. According to receiving domain view, the advantage and result of role accumulation happens with enrichment, especially in the domain of enrichment receiver. However, this point of view received inconsistent empirical support. Meanwhile, according to originating domain view, the domain where enrichment sourced is the domain which gained the main advantage. It shows positive attribute in the initial role of enrichment, and the job satisfaction will increase in roles which generate resources. Empirical study consistently supports this relationship, and another study on two-ways enrichment showed that WFE has greater effect on job satisfaction and FWE in family satisfaction (Carlson et al., 2014). It corresponds to the results of meta-analysis by McNall et al. (2010a), which found that WFE gives stronger impact to work setting. Meanwhile, FWE gives stronger impact to the non-work setting. The results of meta-analysis by Shockley and Singla (2011) found that WFE has relatively weak impact on family satisfaction and relatively strong impact on job satisfaction. FWE implies weak impact on job satisfaction and stronger impact on family satisfaction. This relationship proves that enrichment reaction can happen, especially in origin domain, known as source attribution perspective. Akram et al. (2014) found that WFE is a significant predictor of job satisfaction, and Carlson et al. (2010) found that WFE has positive relationship with job satisfaction. Fung et al. (2014) pointed out that WFE implies positive relationship with job satisfaction. Michel and Michel (2015) showed that WFE implies positive and significant relationship with job satisfaction. The results of meta-analysis by McNall et al. (2010a) found that WFE has stronger impact to job satisfaction, while FWE has stronger impact to family satisfaction. In addition, Tang et al. (2014) pointed out that WFE has positive relationship with job satisfaction. This study offers the following hypothesis: H7: WFE has a positive effect on job satisfaction.

WFE as Mediator
According to COR theory, in order to reduce stress, an individual tends to conserve and acquire resources which produce positive gains spiral (Hobfoll, 2001). COR theory can be understood as stress theory or motivational theory. Stress happens due to the threat of losing resources, actual loss of resources, and when resource investment does not bring the expected fortune. As a motivational theory, COR explains that individuals are in their attempts to retain, protect, and build resources (Hobfoll, 1989). COR theory is based on the principle which holds that individuals are motivated to secure their resources and obtain new resources (Halbesleben, 2014). Resources include object, personal characteristics, condition, energy, and social support (Hobfoll, 1989). Supervisor support, self-esteem, and optimism are the resources which are able to reduce stress and other negative consequences which later influence the enrichment process. Through enrichment process, COR theory can be used to understand the relationship between WFE and job satisfaction as the outcome. Supervisor support, self-esteem, and optimism can both directly or indirectly affect WFE. Higher the supervisor support, self-esteem and optimism, higher the WFE and it will eventually increase job satisfaction. The role of WFE as mediator is first reported by McNall et al. (2010b). Baral and Bhargava (2010) pointed out that WFE mediates the relationship between job characteristics and job outcomes (job satisfaction, organizational citizenship behavior, and affective organizational commitment) by participating 216 managerial staff from four organizations in India as research sample. Study by McNall et al. (2010b) with 220 staff members as research sample showed that WFE mediated the relationship between the availability of flexible work arrangements and job satisfaction, and WFE mediated the relationship between flexible work arrangement and turnover intentions. Study by Tang et al. (2014) with 543 staff members in China as research sample showed that WFE mediated the relationship between supervisor support and job satisfaction, while WFE mediated the relationship between organizational support and job satisfaction. Study by Fung et al., (2015) with 280 teachers in Malaysia showed that WFE mediated the relationship between supervisor support and job satisfaction. This research offers the following hypothesis: H8: WFE as the mediator of the effect of supervisor support on job satisfaction. H9: WFE as the mediator of the effect of self-esteem on job satisfaction. H10: WFE as the mediator of the effect of optimism on job satisfaction.

Research Population
This study included 263 staff of bank BPD DIY (Regional Development Bank) in seven branch offices in Special Region of Yogyakarta (DIY): branch office Utama, Senopati, Sleman, Bantul, Wates, Wonosari, and Syariah. This study implements census.

Operational Definition and Variable Measurement
Job satisfaction/KK (Y): Job satisfaction can be defined as the level of positive feelings of employees towards work as a result of evaluation of its characteristics (Robbins & Judge, 2008). Job satisfaction is measured by 5 indicators developed by Brayfield & Rothe (1951) in Judge and Klinger (2008). Responses are made using Likert scale of 1 (strongly disagree), 2 (disagree), 3 (neutral), 4 (agree), and 5 (strongly agree).
Work-family enrichment/WFE (M): WFE is defined as the extent of the transferability of experience from work role to family role (Greenhaus and Powell, 2006). WFE is measured by questionnaire developed by Carlson et al. (2006) consisting of 9 indicators. Responses are made using Likert scale of 1 (strongly disagree), 2 (disagree), 3 (neutral), 4 (agree), and 5 (strongly agree).
Self-esteem/SE (X2): Self-esteem is defined as an assessment of an individual towards herself in relation to her role in the organization. Self-esteem is measured by questionnaire developed by Rosenberg (Greenberger et al., 2003) with 10 indicators. Responses are made using Likert scale of 1 (very low), 2 (low), 3 (neutral), 4 (high), and 5 (very high). The average of responses obtained from the respondents is counted by the highest score (5) minus the lowest score (1) divided by two (2). Based on the average score, self-esteem is then categorized into low self-esteem (category 1) for score 1 to 3 and high selfesteem (category 2) for score from 3.1 to 5.
Optimism/OP (X3): Optimism is defined as the level of positive expectations regarding future events or outcomes (Bastianello et al., 2014). Optimism is measured by Life Orientation Test-Revised (LOT-R) questionnaire developed by Scheier et al. (1994) with 9 indicators. Responses are made using Likert scale of 1 (strongly disagree), 2 (disagree), 3 (neutral), 4 (agree), and 5 (strongly agree). Data analysis in this study used SEM with AMOS program.

Respondents' Characteristics
Respondents in this study were 263 permanent employees in seven branches of Bank BPD DIY, who occupy positions as staff and are married. The respondents' characteristics in this study are presented in the following table.

Descriptive Statistics
Descriptive statistics provide data description from the minimum score, maximum score, average (mean), and standard deviation values. The results of descriptive statistical analysis are presented in the following table.

Confirmatory Factor Analysis (CFA)
CFA aims to test the multidimensionality of a theoretical construction. The validity test criteria used in this study: if all indicators are statistically significant and have standard loading factor value of > 0.50. Based on the CFA test results, it turns out that the indicators/items of supervisor support, self-esteem, optimism, WFE, and job satisfaction variables are all significant and have standard loading factor of > 0.50. Therefore, all indicators/items in this study are valid.

Structural Model Analysis
At this full model stage, suitability and statistical tests are conducted. Based on the chi-square value and the probability of the goodness of fit criteria, it turns out that the model is not yet fit, so it needs to be modified. The results of the modified model are as follows.

Fig. 1. Full Model Modification Result
The result of full model modification shows the model's fitness. The criteria of goodness of fit full model is shown in the table below.

Structural Model Evaluation of the Modified Model
Test for Assumption of Normality. The data of assumption of normality is performed using criteria ratio (c.r.) of ± 2.58 at a significance level of 0.01. Based on multivariate data normality test results, it turns out that the critical ratio value (c.r.) of -1.088 is between ±2.58. Therefore, the data used in this study has multivariate normal distribution.
Assumption of Multicollinearity. The analysis shows that there is no "warning" in the data used in this study. Therefore, there is no multicollinearity and singularity in the data used in this study.
Construct Reliability Test. The construct reliability test in this study uses construct reliability, with the minimum cut-off value of construct reliability of 0.70. The results of construct reliability test showed that the variables of supervisor support, optimism, work-family enrichment, and job satisfaction have cut-off value of > 0.70, and declared reliable.

Interpretation of Modified Model
Chi-square test results showed a value of 358.414 and probability of 0.112, which is greater than 0.05. Likewise the other criteria, namely CMIN/DF indicated the probability value of 1.096 which is smaller than 2.00, RMSEA indicated the probability value of 0.019 which is smaller than 0.08, GFI indicated the probability value of 0.912 which is greater than 0.90, and TLI indicated the probability value of of 0.992 which is greater than 0.90. The results showed that the model of WFE as an influence mediator for supervisor support, self-esteem, and optimism on job satisfaction in this study is fit with empirical data or data samples.

Hypothesis Testing
The results of hypothesis testing from 1st to 10th hypothesis are shown in the following table. The results of the analysis showed that supervisor support has a positive and significant effect on job satisfaction with regression coefficient of 0.196. Therefore, the first hypothesis which states that supervisor support has a positive effect on job satisfaction is supported in this study. The results of the analysis show that self-esteem had a positive and significant effect on job satisfaction with regression coefficient of 0.149. Therefore, the second hypothesis which states that self-esteem has a positive effect on job satisfaction is supported in this study. The results of the analysis show that optimism had a positive and significant effect on job satisfaction with regression coefficient of 0.125. Therefore, the third hypothesis which states that optimism had a positive effect on job satisfaction is supported in this study. The results of the analysis showed that supervisor support has a positive and significant effect on WFE with regression coefficient of 0.433. Therefore, the fourth hypothesis which states that supervisor support has a positive effect on WFE is supported in this study. The results of the analysis showed that self-esteem has a positive and significant effect on WFE with regression coefficient of 0.164. Therefore, the fifth hypothesis which states that self-esteem had a positive effect on WFE is supported in this study. The results of the analysis showed that optimism has a positive and significant effect on WFE with regression coefficient of 0.421. Therefore, the sixth hypothesis which states that self-esteem had a positive effect on WFE is supported in this study. The results of the analysis show that WFE had a positive and significant effect on job satisfaction with regression coefficient of 0.585. Therefore, the seventh hypothesis which states that WFE had a positive effect on job satisfaction is supported in this study. The results of the 8th to 10th hypothesis testing using Sobel test show that WFE had significant role as a mediator. The results of Sobel test in this study are summarized in the following table.

Discussions
The results of this study indicate that supervisor support had a positive and significant effect on job satisfaction. The results of this study support the previous studies by Ng and Sorensen (2008), Ibrahim (2012), Kula and Guler (2014), Tang et al. (2014), Fung et al. (2015), and Gok et al. (2015), which found that supervisor support is positively related to job satisfaction. According to the theory of conservation of resources (COR), individuals try to obtain, maintain, protect, and develop resources (Hobfoll, 1989;Hobfoll, 2002). This finding is in accordance with COR theory, that supervisor support is a form of social support and is a resource coming from the workplace. Research results show that self-esteem had a positive effect on job satisfaction. This result supports the previous studies by Ahmed (2012), Judge and Bono (2001), Chang et al. (2010), Cleare (2013), Igbal (2012), and Nurullah (2010), which found that self-esteem has a positive effect on job satisfaction. According to COR theory, personal characteristics, in this case is self-esteem, is a resource that can help someone to fight stress. People must invest resources to protect against loss of resources, recover from losses, and obtain resources (Hobfoll, 2011).
The results show that optimism had a positive effect on job satisfaction. It corresponds with Ahmed (2012), Chang et al. (2010), and Kaplan and Bickes (2013), who found that there was a positive relationship between optimism and job satisfaction. According to COR theory, optimism is a personal resource (Ten Brummelhuis & Bakker, 2012). This resource, according to Hobfoll (2002), is a basic component of an individual's ability to adapt. Individuals with a high level of optimism will have lower levels of psychological pressure, less stress, and the ability to reduce depression and loneliness, and obtain greater social support compared to less-optimistic individuals. Research results showed that supervisor support has a positive effect on WFE. The results of this study support previous studies by Baral and Bhargava (2011a), Lu (2011), andFung et al. (2015), which found that supervisor support is positively related to WFE. According to COR theory, people are trying to protect and obtain new resources. One resource that comes from the workplace is supervisor support.
Research results show that self-esteem had a positive effect on WFE. The results of this study support previous studies by Rashid et al. (2011), which found that self-esteem is positively related to WFE. According to COR theory, self-esteem an important personal resource as the center of an individual well-being. An individual with high self-esteem will be motivated to achieve better welfare. People with high self-esteem will produce significant effects by buffering stress, negative emotions, increasing personal adjustment, and being psychologically healthy and happy. Individuals with high self-esteem are more able to deal with anxiety, improve coping and physical health. Conversely, individuals with low self-esteem will have psychological frigidity, prone to depression, anxiety, and disability (Rashid et al., 2011). The results showed that optimism has a positive effect on WFE. The results of this study support the findings of Washington (2006) that optimism is positively related to WFE and FWE and negatively related to WFC and FWC. According to COR theory, optimism is one of the personal resources that help individuals deal with stress and develop enrichment (Hobfoll, 1989).
According to theory by Friede and Ryan (2005) in Fung et al. (2012) which integrates personality factors with the outcomes of work and family relationships, individuals with different personalities can interpret the same situation, namely as a conflict or as enrichment in the relationship between work and family. In this case, personality variable has an important role in the perception of enrichment. Grzywacz and Marks (2000) also found that certain personality characteristics were related to WFE. Wayne et al., (2004) examined personality as a factor contributing to the occurrence of WFE. Research results showed that WFE has a positive effect on job satisfaction. The results of this study support previous research conducted by Akram et al. (2014), Carlson et al. (2010), Carlson (2014), Fung et al. (2014), Michel and Michel (2015), McNall et al. (2010a), andTang et al. (2014), who found that WFE was positively related to job satisfaction. This finding supports the originating domain view theory which states that the domain from which enrichment originates is the domain that gets the main benefit. The results of this study indicate that WFE acts as an effect mediator of supervisor support on job satisfaction. This is indicated by Sobel test statistic value of 5.11111720 and the two-tailed probability value of 0.00000032 < 0.05.
According to COR theory, people try to conserve and acquire new resources to produce a positive gains spiral. One of the resources obtained from the workplace is supervisor support. In accordance with COR theory, supervisor support will produce other resources namely WFE, which in turn will produce job satisfaction. This mediating effect provides support for the antecedents and consequences of WFE, that supervisor support is one of the antecedents of WFE, while job satisfaction is the consequence of WFE. The results showed that WFE acts as an effect mediator of self-esteem on job satisfaction. This is indicated by Sobel test statistic value of 3.17771886 and the two-tailed probability value of 0.00148439 < 0.05. In accordance with COR theory, self-esteem as personal resource will produce other resources, namely WFE, which in turn will produce job satisfaction. This mediating effect provides support for the antecedents and consequences of WFE, that self-esteem is one of the antecedents of WFE, while job satisfaction is a consequence of WFE. The results also showed that WFE acts as an effective mediator of optimism on job satisfaction. This is indicated by the value of Sobel test statistic for 4.98075336 and the twotailed probability value of 0.00000063 < 0.05.
The results of this study support the research by McNall et al. (2010b), Baral and Bhargava (2010), Tang et al. (2014), and Fung et al. (2015). In accordance with COR theory, optimism as personal resource will produce other resources, namely WFE, which in turn will produce job satisfaction. This mediating effect provides support for the antecedents and consequences of WFE, that optimism is one of the antecedents of WFE, while job satisfaction is the consequence of WFE. The results of this study support the COR theory which states that to reduce stress, individuals tend to conserve and acquire resources to produce a positive gains spiral (Hobfoll, 2001). COR theory is based on the principle that individuals are motivated to protect their current resources and obtain new resources (Halbesleben, 2014). Supervisor support, self-esteem, and optimism are resources that can reduce stress and other negative consequences, which in turn have an impact on the enrichment process. Through this enrichment process, COR theory can be used to understand the relationship between WFE and the outcome, in this case is job satisfaction. The influence of supervisor support, self-esteem, and optimism on job satisfaction can occur directly or indirectly through WFE. The higher the support of supervisors, self-esteem, and optimism, it will increase WFE, which in turn will increase job satisfaction.

Conclusions
The result of this study is expected to contribute both theoretically and empirically. Research on the relationship between work and family from a positive perspective is still underdeveloped, whereas it is important to understand that the relationship between work and family does not always lead to conflict, but can also have a positive or mutually beneficial impact. Therefore, this study seeks to fill this gap by providing an understanding of the relationship between work and family from a positive perspective. Theoretically, the results of this study indicate that the WFE model as an influence mediator of supervisor support, self-esteem, and optimism on job satisfaction fit with the empirical data. This study provides a theoretical contribution in the form of WFE model as an influence mediator of supervisor support, self-esteem, and optimism on job satisfaction. The results of this study indicate that WFE acts significantly as an influence mediator of supervisor support on job satisfaction. This can be interpreted that the influence of supervisor support on job satisfaction is mediated by WFE. WFE acts significantly as an influence mediator of self-esteem on job satisfaction. This can be interpreted that the effect of self-esteem on job satisfaction is mediated by WFE. WFE acts significantly as an influence mediator of optimism on job satisfaction. This can be interpreted that the effect of optimism on job satisfaction is mediated by the WFE.
This study is served as a contribution to the low number of the research on the relationship between work and family from a positive perspective, especially in developing countries like Indonesia. It is important to understand that the relationship between work and family does not always lead to conflict, but can also have a positive impact or mutual benefit. Therefore, this study seeks to fill this gap by providing an understanding of the relationship between work and family from a positive perspective. This study aims to examine WFE model as an influence mediator of supervisor support, self-esteem, and optimism on job satisfaction. Second, theoretically, the results of this study indicate that the WFE model as an influence mediator of supervisor support, self-esteem, and optimism on job satisfaction, fit with empirical data.
This study provides a theoretical contribution in the form of WFE model as an influence mediator of supervisor support, selfesteem, and on job satisfaction. Third, the results of this study indicate that supervisor support has a positive effect on job satisfaction, self-esteem has a positive effect on job satisfaction, and optimism has a positive effect on job satisfaction. The results of this study support the results of previous studies, so the support of supervisors, self-esteem, and optimism are antecedents or predictors of job satisfaction. Fourth, the results of this study indicate that supervisor support has a positive effect on WFE, self-esteem has a positive effect on WFE, and optimism has a positive effect on WFE.
The results of this study support the results of previous studies, so the support of supervisors, self-esteem, and optimism are antecedents or predictors of WFE. Fifth, the results of this study indicate that WFE has a positive effect on job satisfaction. This study supports the originating domain view regarding the relationship between WFE and job satisfaction. According to this view, the domain where enrichment originated is the domain which obtained the main benefits. In this case, job satisfaction will increase because work setting becomes the origin of enrichment. Limitations of this study are: (1) the data in this study are cross-sectional data, and use longitudinal data is recommended for further research; (2) data collection techniques in this study using self-report questionnaires, so it depends on the respondents' perceptions; and (3) this study only tests the WFE model, and testing WFE as a mediator is recommended.