Predictors of Workplace Well-Being for Dual-Earner Couples During Covid-19 Pandemic

The Covid-19 pandemic has changed the life order of entire community including employees from dual-earner families. Overlapping roles experienced can cause tension and trigger work-family conflicts that can impact workplace well-being. Previous studies had often focused on well-being in general and there was a lack of research related to well-being in dual earner couples during pandemic. This study investigated the predictor role of core selfevaluation and spousal support through mediating role of work family conflict (WFC) on workplace well-being. Based on the conservation of resources theory, we need to identify essential resource to fulfill workplace well-being. Data were collected from 200 employed Jakarta Metropolitan area (Jabodetabek) dual-earners who had at least one child below the age of 13. According to the mediation with two predictors analysis, the mediating effect of WFC can only predict the linkage between core self-evaluation and workplace well-being (β = 0.02, SE =0.02; 95% CI [0.005, 0.061]) but cannot predict the role of spousal support through WFC on workplace well-being (β = 0.00, SE =0.00; 95% CI [-0.002, 0.007]). This finding explained the importance of self-evaluation during pandemic to enhance workplace well-being and buffered negative effect of work and family related burdens.

office have to deal with the possible virus transmission while being out of the house.
Threats to one's well-being can be a source of despair especially when the threats are inexplicable, uncertain, ambiguous, and when the person thinks that they have low control over the stressor (Taha et al., 2014). In addition, to adapt to the new work pattern, employees who are raising school-age children also have to accompany their children as well as being fully responsible for child-rearing. For dual-earner couples, work responsibilities must be fulfilled alongside their domestic tasks. It causes the line between work and personal domains to be more blurred during this pandemic. Various role demands that must be faced altogether become a threat to one's well-being.
Well-being is generally divided into two perspectives namely hedonic which focuses on the presence of positive affect, the lack of negative affect, and cognitive evaluation of individual's life satisfaction as well as eudaimonic which focuses on the fulfillment of one's potentials through self-actualization, meaningfulness, and personal development (Deci & Ryan, 2008;Fisher, 2014;Ryan & Deci, 2001). Hobfoll (1989Hobfoll ( , 2002 explained that when someone is experiencing an increase of resources, they will experience positive well-being while when the person is unable to obtain resources then they will experience stress and lack of well-being. Rothausen (2013) said that resources can be categorized into primary and secondary resources. Primary resources encompass self-esteem, mastery, intimacy, protection, and self-improvement (Hobfoll, 1989(Hobfoll, , 2002. This is aligned with the aspects of psychological well-being that are eudaimonic-oriented, namely self-acceptance, environmental mastery, positive relationships, autonomy, life purpose, and personal growth (Ryff, 1989). Secondary resources help someone to achieve the primary resources (Rothausen, 2013) and it can include objects, relations, conditions, and personal characteristics that serve as the means to an end (Hobfoll, 1989(Hobfoll, , 2002. In line with that, Santrock (2011) described that family and work are the two main domains that become an adult's focus. Thus, the measurement of well-being particularly for adults needs to incorporate the two domains to be more comprehensive. Zheng et al. (2015) introduced a more contextual employee well-being concept for workplace situations compared to other general well-being concepts. The measurement of well-being in a specific context is needed to capture the details, complexities, and variations of an employee's cognitive and affective experiences at the workplace.
Moreover, the employee well-being concept developed by Zheng et al. (2015) is oriented to the Eastern culture which has a difference from the well-being concept applied in Western culture. Specifically, the Western culture emphasizes environmental mastery, autonomy, and personal emotions in well-being. On the other hand, Eastern culture emphasizes harmony and social values in achieving a higher level of well-being (Zheng et al., 2015). The concept of employee well-being is also a combination of hedonic and eudaimonic perspectives that measures both satisfaction in life and employment as well as aspects of psychological well-being at the workplace. The measurement of employee EFI & PARAHYANTI || WORKPLACE WELL-BEING 35 well-being should not only consider the work and health of an employee but also measure their familial relationship and life satisfaction (Zheng et al., 2015). Zheng et al. (2015) stated that employee well-being involves not only employee perception and feeling about their work and life satisfaction but alson their psychological experience and the level of satisfaction exhibited in both their work and personal lives.
Employee well-being is construed of three components, i.e.: life well-being (LWB) which focuses on personal emotion reflection, life problems, and family; workplace well-being (WWB) that includes aspects related to work such as compensation and benefits, employment protection, managerial style, and the work arrangements; as well as psychological well-being (PWB) that emphasizes learning, personal development, work achievements, and self-actualization (Zheng et al., 2015).
The present study would deeply explore the predictive factors needed to achieve workplace well-being (WWB) that encompass job satisfaction and things affecting work (Zheng et al., 2015) amidst the pandemic. The shift in work arrangement and the unpredictable economic situation can have impacts on job continuity. Covid-19 pandemic that is happening can be a threat to the essential resources of an individual and thus threaten workplace well-being. It is aligned with the conservation of resources theory (Hobfoll, 1989(Hobfoll, , 2002 which describes that a person would strive to obtain, maintain, keep, and protect things that are valuable to them. According to the theory, stress occurs when a) essential resources are threatened with loss, b) essential resources are lost, or c) failing to obtain essential resources following significant effort (Hobfoll, 1989(Hobfoll, , 2002. Thus, it is important to identify the essential resources required by employees especially in dualearner couples who raise children amidst pandemics to achieve workplace well-being.
One of the secondary resources that can facilitate well-being attainment is personal characteristics (Rothausen, 2013). In addition, social support can be seen as a resource that provides or facilitates other valuable resources for the individual (Hobfoll, 1989(Hobfoll, , 2002. Therefore in this study, personal characteristics, core self-evaluation, and spousal support become the essential resources that could be used to predict workplace well-being in dual earner couples which is mediated by work-family conflict. Previous studies correlated core self-evaluation with well-being about employment, such as burnout and job satisfaction (Hsieh et al., 2019;Wu & Griffin, 2012). Furthermore, core self-evaluation is often correlated with subjective well-being concept that is hedonicoriented (Liu et al., 2016;Montasem et al., 2013;Rosopa et al., 2016;Sudha & Shahnawaz, 2013) and psychological well-being which is eudaimonic-oriented (Ding et al., 2020;Gibson & Hicks, 2018). Specifically, a study that directly links core self-evaluation and workplace well-being has yet to be done. Core self-evaluation is a person's fundamental evaluation of oneself, others, and the world; which is comprised of four aspects namely self-esteem, generalized self-efficacy, locus of control, and emotional stability (Judge et al., 1998). All aspects of core self-evaluation simultaneously reflect individual's positive belief in their own capacity and it motivates them to proactively face stressors so that they can EFI & PARAHYANTI || WORKPLACE WELL-BEING 36 deal with the strains that arise (Greaves et al., 2017). A person with high core selfevaluation will assess their own self as having a positive attitude in various situations and as a capable and valuable individual who has control over their life (Judge et al., 2005).
Spousal support is one of the important factors that affect employee well-being (Peeters & Le Blanc, 2001). Support in the form of advice, understanding, and alikes provided by a spouse (Aycan & Eskin, 2005) can be distinguished into emotional and instrumental spousal supports (King et al., 1995). Both emotional and instrumental supports can strengthen employee well-being (Irak et al., 2020;Karapinar et al., 2019;Rasool et al., 2019).
The impact of changes during the pandemic period presents new challenges to dualearner couples to be able to decide the boundary between family and work which demands several transitions altogether (Restubog et al., 2020). Goode (1960) explained that overlap in fulfilling several roles at the same time often triggers role strain. Work to family conflict (WFC) is a conflict between roles where demands, time spent, and strains formed due to work issues can affect the responsibilities related to the family (Netemeyer et al., 1996). Core self-evaluation influences work to family conflict that life satisfaction and individual's job will be affected as well (Boyar & Mosley, 2007).
For countries with Eastern culture, harmony and social values significantly affect relationships within the family and thus social support, especially spousal support, is important (Karapinar et al., 2019). A study by Zhang and Ma (2020) showed that amidst Covid-19 pandemic, respondents experienced an increase in support and felt more caring towards family members. According to buffering hypothesis (Cohen & Wills, 1985), Covid-19 pandemic becomes a potential event that triggers, where individual assessment towards this event would lead to psychological response or adaptive behavior that affects their well-being and is influenced by social support. Researchers assumed that individuals with high spousal support would be protected from the negative effects of work to family conflict so that they can achieve better well-being.
This study could describe the essential resources and mechanisms to achieve workplace well-being, particularly in dual-earner couples with children in the current pandemic era. The findings could become considerations for organizations in taking strategic steps to formulate policy for dual-earner couples with children so they can maintain productivity during the pandemic. Moreover, this study can contribute to enriching the discourse of well-being within the context of the organization as well as about dual-earner couples in Indonesia during such a trying period. Based on that, researchers propose several hypotheses: 1) there is a direct correlation between core selfevaluation and workplace well-being, 2) there is a direct correlation between spousal support and workplace well-being, 3) work to family conflict mediates the correlation between core self-evaluation and workplace well-being, and 4) work to family conflict mediates correlation between spousal support and workplace well-being.

Method
This study was a quantitative study with a correlational design. The sampling technique used was non-probability sampling, specifically accidental sampling. Researchers utilized this sampling technique because the size of the study population was not precisely known and consideration of ease to find respondents participants whilst still considering the representation of respondents based on the specific criteria. The study involved 200 participants with the following characteristics: dual-earner couples who worked full-time and had at least one child aged under 13 years old in the Jakarta Metropolitan area (Jabodetabek). The decision to include parents with children aged 0 to 13 years old was based on the demand of time and energy from parents which is so high in a family with an infant to elementary school children, and thus the possibilities to be exposed to higher psychological stress (Lopata, 1966;Barnett, 1993;in Aycan & Eskin, 2005).

Research instruments
There were four research variables in the present study, namely: workplace well-being as the dependent variable, core self-evaluation and spousal support as independent variables, and work to family conflict as mediator. This study tested the role of mediator on the two independent variables in predicting the dependent variable. It utilized several measurements, namely: a) Employee Well-being Scale (EWB) developed by Zheng et al. (2015) to measure employee well-being, specifically in the workplace well-being (WWB) dimension, using 6 items with Cronbach's alpha coefficient = 0.87; b) Core Self-Evaluation Scale (CSES) developed by Judge et al. (2003) to measure core self-evaluation using 12 items with Cronbach's alpha coefficient = 0.84; c) Work Family Conflict Scale (WFCS) developed by Netemeyer et al. (1996) to measure work to family conflict using 5 items with Cronbach's alpha coefficient = 0.88; 4) Family Support Inventory (FSI) developed by King et al. (1995) to measure spousal support using 44 items reported to have Cronbach's alpha coefficients for emotional support scale = 0.97 and instrumental support scale = 0.93.
All of the measurements used a 6-point Likert scale where 1 was for "Strongly Disagree" and 6 for "Strongly Agree". The measurements were adapted to the Indonesian language following the guideline of cross-cultural adaptation by Beaton et al. (2000): 1) initial translation; 2) translation synthesis; 3) back-translation to the original language; 4) review from the expert committee; and 5) pilot study to 28 respondents. After undergoing scale adaption and considering item factor loading in the pilot study, researchers selected 10 items from CSES and 39 items from FSI.

Data analysis
Researchers conducted descriptive analysis and assumption tests using SPSS 24.0 version and measurement model testing with confirmatory factor analysis using Lisrel 8.80 EFI & PARAHYANTI || WORKPLACE WELL-BEING 38 version. Hypothesis testing was done using mediator analysis with two predictors using MPlus 7.0 version.

Result
Model testing using confirmatory factor analysis was conducted on the four research variables. Workplace well-being, core self-evaluation, and work to family conflict are unidimensional variables; thus measured with 6-item, 10-item, and 5-item scales. Then, the model testing for spousal support consists was first done to test spousal emotional support (27 items) and spousal instrumental support (12 items). According to the analysis done to core self-evaluation, work to family conflict, and spousal support some items should be eliminated as they had low factor loading. There were item reductions in several variables for further analysis and the addition of error covariance for inter-item measurement of similar construct thus the fit indices of the revised measurement model became satisfactory. The measurement model was determined to fit to several indicators suggested by Meyers et al. (2005)   The preliminary analysis; i.e., descriptive statistic, reliability, Product Moment correlational analysis; from the instruments used and presented in Table 1 described the correlation between workplace well-being, core self-evaluation, spousal support, work to family conflict, and the demographic data of respondents. According to Table one, it was found that all of the research variables showed adequate Cronbach's alpha α > 0.70 (Nunnally & Bernstein, 1994) which proved that all of the instruments have good reliability.
According to the correlation between demographic data and research variables, it was found that the respondent's age, child's age, and work pattern are not correlated with all of the research variables. Meanwhile, gender is correlated with core self-evaluation and work to family conflict. The distribution of respondents based on gender was 75% females and 25% males. The age ranges were 25 to 44 years old (94% of the respondents), 45 to 52 years old (5% of the respondents), and 24 years old (1% of the respondents). During the pandemic, respondents were categorized into several work pattern options: 20% implemented work from home (WFH), 20% implemented work from office (WFO), 50% implemented a combination of WFH and WFO, and 10% implemented other work patterns. The age of respondents' children was as follows: 54% 0 to 2 years old, 29% 3 to 6 years old, and 17% within the range of 7 to 12 years old.
According to Table 2, the response distributions of the four variables were dominated by the "high" category. All of the respondents even showed high core self-evaluation amidst the current pandemic situation. In the current pandemic, half of the respondents also perceived high work to family conflict. According to hypothesis testing, there is a direct relationship between core self-evaluation and workplace well-being (b = -0.54, t = 8.75; p < 0.05) but there is no direct relationship between spousal support and workplace well-being (b = -0.01, t = -.0.99; p =0.32). It can be concluded that the first hypothesis is accepted while the second one is rejected. Based on the mediator analysis of two independent variables, it was found that workplace well-being can only be predicted by core self-evaluation mediated by work to family conflict. As seen in Table 3, core self-evaluation has negative effect on work to family conflict (b = -0.17, t = -2.20; p < 0.05) and work to family conflict has negative effect on workplace well-being (b = -0.14, t = -2.17; p < 0.05). The researchers used bootstrapping to test the significance of the non-direct effect from work to family conflict (Hayes, 2013).
Using 5000 bootstrapping samples with a 95% confidence interval, the indirect effect from work to family conflict in the relationship between core self-evaluation and workplace well-being was found significant (β = 0.02, SE =0.02; 95% CI [0.005, 0.061]). However, the indirect effect of work to family conflict in the relationship between spousal support and workplace well-being was not found significant (β = 0.00, SE =0.00; 95% CI [-0.002, 0.007]).
Thus, it can be concluded that the third hypothesis is accepted and the fourth one is rejected.

Discussion
Hobfoll (1989) defined resources as objects, personal characteristics, conditions, or energy that are valuable for a person or that serve as facilitators to obtain other objects, personal characteristics, conditions, or energy. According to the conservation of resource theory (Hobfoll, 2002) and buffering hypothesis (Cohen & Wills, 1985), researchers examined the role of mediation by work to family conflict on core self-evaluation and spousal support in predicting workplace well-being among dual-earner couples who have at least one child under 13 years old in the Jakarta Metropolitan area during Covid-19 pandemic. In this study, it was found that core self-evaluation can be a significant essential resource to achieve workplace well-being for dual-earner couples but not spousal support.
In the first hypothesis, there is a significant correlation between core self-evaluation and workplace well-being. This aligned with previous studies that linked core selfevaluation with subjective well-being and emotional fatigue which indirectly affects wellbeing (Greaves et al., 2017;O'Donoghue et al., 2016;Rosopa et al., 2016). An individual with high core self-evaluation are known to have higher life and job satisfaction as well as more sensitive in reading signs of problems, thus having advantages in facing future stress (Greaves et al., 2017;Judge et al., 2005;Judge et al., 2008;Piccolo et al., 2005).
In the third hypothesis, it was proven that work to family conflict mediated the relationship between core self-evaluation and workplace well-being. An individual with high core self-evaluation is known to experience lesser work to family conflict (Boyar & Mosley, 2007). Someone with high core self-evaluation evaluates their ability to solve issues in a more positive light than a person with low core self-evaluation who assesses such situation as a threat (Judge et al., 2005). Thus, this core self-evaluation can help someone to perform proactive coping, which is a form of anticipation or detection towards possible stressors, and act first to prevent or minimalize the effect (Aspinwall & Taylor, 1997).
Aligned with buffering hypothesis (Cohen & Wills, 1995) and the conservation of resource theory (Hobfoll, 2002), in the present study, core self-evaluation had a role in reducing the negative effect of work to family conflict that can influence an individual's workplace well-being. Judge et al. (1998) explained that there are two mediation mechanisms in core self-evaluation, situation judgment, and action. In the present study, core self-evaluation drove someone to take action and manage work to family conflict more effectively to achieve better workplace well-being. A person with higher core selfevaluation will perceive the dual roles in their life more positively; as something that is not too demanding, not conflicting with each other, and tend to provide more EFI & PARAHYANTI || WORKPLACE WELL-BEING 42 opportunities for evaluation (Westring & Ryan, 2010). Therefore, dual-earner couples with higher self-evaluation can integrate job roles and family roles, and able to manage work to family conflict better (Karatepe & Azar, 2013).
A previous study showed that work to family conflict is a mediator between social support, namely supervisor support and family support, and the output of well-being did not show consistent results (Blanch & Aluja, 2012). Social support can function as a supporting variable in several conditions and it is necessary to link the source of support with the direction of conflict or disturbance (O'Driscoll et al., 2004). Spousal support is expected to be more significant in influencing the family to work conflict, while support from supervisor and coworker is more relevant to work to family conflict (Aycan & Eskin, 2005;Byron, 2005;Drummond et al., 2017). This is further reinforced by findings from Aryee et al. (1999) and Burkee and Greenglass (1999) which showed that spousal support does not affect work to family conflict but can lower the effect of the family to work conflict. Those research findings can explain the insignificance in the relationship between spousal support and workplace well-being through work to family conflict. The finding in the present study provides an overview that spousal support in work to family conflict is not as relevant compared to previous findings during the pandemic situation.
One of the limitations in this study was not measuring the two-way conflict between work and family comprehensively and only focused on measuring work to family conflict. In a meta-analysis by Byron (2005), the directions of conflict between work and family domains, as well as the sources of support received by a person have their own specifications. Additionally, the uneven distribution of demographic data such as age and gender needs to be taken into consideration as they are found to have an impact on wellbeing research in dual-earner couples. (de Wet et al., 2012;Drummond et al., 2017).

Conclusion
The conclusion that can be drawn from this study is that core self-evaluation can predict work to family conflict in workplace well-being for dual-earner couples who have children under 13 years old amidst the ongoing pandemic situation. On the other hand, spousal support's role was found to be unable to predict well-being, directly or mediated by work to family conflict. Thus, it is necessary for dual-earner couples with young children to have a better core self-evaluation so that they can tone down the negative effects from pressures of strains related to work and family, as it can incite work to family conflict and affect workplace well-being.

Suggestion
According to the limitation of the present study, it is suggested that future replications of this study should consider several issues. Future studies can measure both work to family conflict and family to work conflict. In addition, it is necessary to include other sources of support such as spousal support, supervisor support, and co-worker support in predicting workplace well-being. The demographical factors of the participants, such as age and number of children, also need to be analyzed.
For human resource practitioners and government, this study's finding can be a basis to formulate policy and development programs for married employees with children during the pandemic. Furthermore, the company can develop employee development programs, such as counseling and training that focus on the aspects of core self-evaluation namely self-esteem, generalized self-efficacy, locus of control, and emotional stability to achieve workplace well-being.