Research Article Work vs. Family and Marital Satisfaction among Malaysian Chinese Dual-earner Couples: An Actor-Partner Interdependence Model

Work-family conflict is frequently studied because they are the two main life domains in a person’s life. The present study examines how work-family conflict influenced marital satisfaction among Malaysian Chinese dual-earner couples using a dyadic approach. Results from the Actor-Partner Interdependence Model (APIM) indicated that work-family conflict (work-to-family and family-to-work) were associated with a lower level of marital satisfaction. Work-family conflict has a significant negative impact on marital satisfaction for both husbands and wives (actor effect). However, no significant results were found for the partner effect. Both actor and partner effects were reported and discussed. The present findings provide empirical evidence on the spillover and crossover effects of work-family conflict on marital satisfaction and add to the literature by utilizing advanced data analysis technique, APIM by collecting dyadic data. The implications of the findings were discussed as well as the suggestions for future research. and crossover effects from Taiwanese sample, focusing on how work-to-family conflict mediates the relationship between emotional exhaustion and marital satisfaction. The study was done by recruiting 226 employees and partner dyads from different work settings such as manufacturing, electronics, banking, service industries, and public organizations through convenience sampling. The mean age for males was 39 while for females, it was 38. Both spillover and crossover effects were found. Emotional exhaustion at work was positively correlated to work-to-family conflict and negatively correlated to marital satisfaction. The spillover effect of work-to-family conflict on marital satisfaction was significant. As for crossover effect, an employees’ emotional exhaustion was positively correlated to their partner’s family-to-work conflict and was

The experiences in a marriage are closely related to various aspects of life satisfaction and physical health (Diener & Chan, 2011;Robles, 2014). Marital interactions have been found to be closely related to individuals' psychological and physical health, where marital functioning has a direct effect on depression and health habits .
Moreover, relationship health has begun to be considered a public health issue in recent years (Cordova et al., 2014) due to the negative implications resulting from poor marital satisfaction .
Marital satisfaction is defined as individuals' global assessment and feelings towards their marriage (Karney & Bradbury, 1995). In addition, marital satisfaction refers to how an individual perceives his or her marriage and the level of happiness being in the marriage.
Marital satisfaction has always been an area of interest for many researchers and a metaanalysis has suggested that wives are 1.6 times more likely to be dissatisfied with their marriage than husbands on average among Asian samples (Jackson et al., 2014). Some of the identified factors contributing to marital dissatisfaction were unbalanced distribution of shared responsibilities such as child care and household chores (Jackson et al., 2014). In addition, wives have additional emotional responsibilities of providing emotional support to their husbands (Jackson et al., 2014).
Marriage seems to be both a protective factor and a risk factor for our health. Married individuals have been found to experience less psychological distress as compared to individuals who are single, divorced or widowed (Barrett, 2000;Waite & Gallagher, 2001).
On the other hand, unsatisfying relationships have been observed to have many negative consequences which may affect an individual's health psychologically and physically. Some examples are eating disorders (Van den Broucke et al., 1997), depression (Beach et al., 1998), anxiety (Lebow et al., 2012), specific diseases such as cancer, cardiac disease and chronic pain Schmaling & Sher, 1997), as well as the risk of mortality Psychological Thought South-West University "Neofit Rilski" 2022, Vol. 15(1), 53-80 https://doi.org/10.37708/psyct.v15i1.614 . Aside from these negative side effects, it has been also reported that difficulties in relationships are one of the most common reasons people seek psychotherapy (Beach et al., 1998). Jackson et al. (2014) called for more research involving racial minority groups in order to contribute additional knowledge in the field of marital satisfaction. This is consistent with the prevailing idea that more studies involving non-WEIRD (Western, educated, industrialized, rich, and democratic) samples are needed (Henrich et al., 2010). The meta-analysis' finding is not conclusive due to the limited number of samples involving Chinese ethnic minority groups. There have been no studies at all involving Malaysian Chinese in the meta-analysis.
Past studies investigating marital satisfaction among Malaysians were mainly based on demographic variables such as gender and race. In addition, individual factors such as attachment styles and couple types were also examined. However, situational factors such as stress experienced at work were not included as a contributing factor to marital satisfaction. Uniformly, all of the studies (e.g. Mustafa et al., 2013;Ng et al., 2009;Ng et al., 2013) were cross-sectional surveys and only one partner was surveyed. Yet, none of the studies compared the marital satisfaction differences between the spouses. Marital satisfaction deserves more attention, especially among Malaysian Chinese as there have previously been limited efforts investigating marital satisfaction among working couples in Asian countries (Hassan et al., 2010;Ng et al., 2009).

Malaysian Chinese: Context, Values, and Marriage
The Malaysian Chinese is an important segment of the 40.3 million overseas Chinese residing outside China, making up the third-largest group after Indonesia and Thailand (Poston & Wong, 2016). As of 2019, Malaysian Chinese (6.7 million) form 22.8% of the population in Malaysia, making them the largest minority group in the country (Department of Statistics Malaysia [DOSM], 2019a).
According to Zaimah et al. (2013), dual-earner families are on the rise in Malaysia.
Malaysia's overall female labour force was 55.5% as compared to the male labour force at 80.4%. It has been observed that the percentage of women participating in the labour force decreases according to age group. Women aged between 25 to 34 make up the highest percentage of the labour force at 73.6%, followed by 67.9% for the 35 to 44 age group, and 58.8% for the 45 to 54 age group (DOSM, 2019b). When compared to 68.3% of overall female labour force participation in the United States (U.S.) (U.S. Bureau of Labor Statistics, 2018) and 73.1% in the United Kingdom and 66.5% in the European Union (EU) (Eurofound, Psychological Thought South-West University "Neofit Rilski" 2022, Vol. 15(1), 53-80 https://doi.org/10.37708/psyct.v15i1.614 2019), Malaysia's overall female labour force participation is relatively low. However, no labour force information on gender by ethnicity is available for comparison. Families in urban areas are more likely to be dual-earner families (Ng et al., 2009) and 46.8% of Malaysian Chinese wives reported that they believed that their husbands would prefer them continuing to work full-time after marriage as this helps with the family's income (Noor, 1999).
The importance of family prosperity -a value influenced by Confucianism, is deeply rooted in Malaysian Chinese culture (Noor, 1999;Tan, 2004). Malaysian Malay women tend to work for financial gain mainly to contribute towards the family and not so much for career advancement. As for Malaysian Chinese women, identity, a sense of self-worth and recognition from work contribute as much towards the family's prosperity as income does (Noor, 1999). Both Malaysian Chinese and Malay women see family roles as a top priority (Noor, 1999). This finding is aligned with the Confucian value which propounds that a man should take care of the family's external affairs while the woman is in charge of household matters (Tan, 2004).
Malaysian Chinese are generally raised with traditional Confucian ideologies (Ng et al., 2009). The Confucian value system is in line with social role theory which emphasizes the importance of socialization in shaping gender role attitudes and expectations towards appropriate male and female behaviours (Eagly & Wood, 2012). Based on the theory, individuals develop gender role attitudes based on early socialization views of how males and females should behave, especially within marriage (Eagly & Wood, 1999, 2012. There are two main types of gender role attitudes: traditional and egalitarian (Kaufman, 2000).
Traditional gender role attitude is defined as the perception of the husband as the family breadwinner and in charge of finances while the wife is associated with household duties and childcare. Egalitarian gender role attitude perceives both men and women to be equally in charge at work and at home, with social roles being neither defined nor segregated by gender (Kaufman, 2000). Hence, men with an egalitarian gender role attitude are more likely to share household responsibilities. Past research has also found that when both husband and wife share the household chores, the burden does not fall on the wife alone, promoting marital satisfaction among these husbands who hold an egalitarian gender role attitude (Amato & Booth, 1995).
Globalization changes in the world have exposed Malaysian women to Western culture and ideas. This has led to a tension between traditional Confucian ideologies and the more Psychological Thought South-West University "Neofit Rilski" 2022, Vol. 15(1), 53-80 https://doi.org/10.37708/psyct.v15i1.614 liberal Western ones which impacts gender role expectations (Edwards & Roces, 2000).
Depending on the gender role attitude one holds, it will influence the way individuals perceive marital roles (King & King, 1997), which in turn contributes to marital satisfaction.
Nonetheless, studies conducted in establishing gender role attitudes as a significant predictor of marital satisfaction are mainly based in the U.S., and evidence from outside of the U.S. are limited (Taniguchi & Kaufman, 2014). Egalitarian gender role attitude is positively correlated with marital quality among men (Amato & Booth, 1995;Ogolsky et al., 2014;Oshio et al., 2013) and marital stability (Olah & Gahler, 2014). Differences between spouses in perception of gender role attitude and marital satisfaction were also identified (Taniguchi & Kaufman, 2014). Although egalitarian gender role attitude is positively correlated with men's marital satisfaction, it is negatively correlated with women's marital satisfaction (Taniguchi & Kaufman, 2014). In a similar study, egalitarian gender role attitude was found to be negatively correlated with marital satisfaction among Korean wives (Yoo, 2020). All the previous studies discussed have examined the relationship between gender role attitudes and marital satisfaction utilising samples from China, Japan, and Korea.
Findings among Malaysian Chinese dual-earner respondents will add new knowledge to the existing literature.

Work-Family Conflict -Spillover and crossover effects
The construct of work-family conflict can be explained through the concept of role conflict in role theory (Biddle, 1986). Role theory defines role conflict as having two or more mismatched expectations towards one's performance simultaneously (Biddle, 1986). Based on Biddle's (1986) definition, when a person is performing a "work" role, it might interfere with the performance of another role, such as the "family" role. Based on this definition, role theory suggests that work and family roles are associated with different expectations and behaviours deemed appropriate to that role. Thus, contradictory demands from the work and family domains may create role conflict (Kahn et al., 1964). Role conflict is considered the antecedent of work-family conflict where work-family conflict is defined as "a form of interrole conflict in which the role pressures from the work and family domains are mutually incompatible in some respect. That is, participation in the work (family) role is made more difficult by virtue of participation in the family (work) role" (Greenhaus & Beutell, 1985, p.77).
The work-family conflict can be regarded as a bidirectional construct, in which work can interfere with family (i.e. work-to-family conflict [WTF]), and family can interfere with work (i.e. family-to-work conflict [FTW]) Frone, 2003;Michel et al., 2010). The operational definition of work-to-family conflict (WTF) is "the degree to which participation in the family role is made more difficult from participation in the work role" and family-to-work Psychological Thought South-West University "Neofit Rilski" 2022, Vol. 15(1), 53-80 https://doi.org/10.37708/psyct.v15i1.614 conflict (FTW) is operationally defined as "the degree to which participation in the work role is made more difficult from participation in the family role" (Michel et al., 2010, p.691).
In addition to the work-to-family and family-to-work conflicts, work-family conflict can be understood by the spillover and crossover effects. The spillover effect is defined as the stress experienced in one domain overflowing into another within the same individual, while the crossover effect refers to the transmission of stress from one person to another (Bolger et al., 1989). Thus, the spillover effect (or actor effect) is an intrapersonal experience while the crossover effect (or partner effect) is an interpersonal transmission of stress (Westman, 2001). To avoid confusion, the spillover effect aligns with the actor effect where the work-to-family and family-to-work conflict experienced individually has a negative effect on their individual marital satisfaction. The crossover effect aligns with the partner effect in the APIM model where a spouse's work-to-family or family-to-work conflict affects their spouse's marital satisfaction. The independent variables are either work-to-family or family-to-work conflict.
The dependent variable is always marital satisfaction.

Psychological Thought
South-West University "Neofit Rilski" 2022, Vol. 15(1), 53-80 https://doi.org/10.37708/psyct.v15i1.614 Undeniably, work and family are two primary domains in life between which conflict is often experienced (van Steenbergen et al., 2011). Due to the limited resources we have, when we spend time and energy in one role, we will have less time and energy for another role (Greenhaus & Beutell, 1985). Family is inseparable from the work domain. Malaysia has shown a trend towards an increasing number of women in full-time employment, which in turn has a direct impact on the work-family interface (Hassan et al., 2010;Zaimah et al., 2013). A new dual-earner household family model has emerged making the traditional-malebreadwinner-family model the exception rather than the norm (Macinnes, 2006). It is said that Malaysia has moved away from the sole breadwinner paradigm towards a dual breadwinner paradigm (Bakar & Abdullah, 2013;Zaimah et al., 2013).
With women's involvement in the work force, structural buffers must exist in order to balance the work-family role system (Pleck, 1977). To explain the spillover effect further, Pleck (1977) suggested that there is an "asymmetrically permeable boundaries between work and family roles for men and women" in the work-family role system (p. 423). The term 'permeable' is defined as the influence of one role upon another, for example, in the case of women, when the family role is allowed to intrude into the work role; or men, when the family role has to give way to or accommodate the work role. Thus, permeability is often asymmetrical for men and for women (Pleck, 1977). Considering that men are expected to provide for the family while women are expected to take care of household duties in social role theory (Eagly & Wood, 1999, 2012, the boundaries of the dominant domain for the respective genders (i.e. work domain for men and family domain for women) are less permeable to their secondary domain based on Pleck's definition (1977). Combining both social role theory and role theory, it is likely that men experience a higher level of work-to-family conflict while women experience a higher level of family-to-work conflict. Opposed to Pleck's (1977) idea, no significant gender differences were found in terms of the permeability between family and work domains, although family boundaries were more permeable than work boundaries (Eagle et al., 1997;Frone et al., 1992;Ngo & Lau, 1998). However, these claims have been dated and more recent examination on the phenomenon is deemed important.
Although the spillover effects of work-family conflict have consistently been found in past research, the effect of work-to-family conflict is found to be more prominent than the effect of family-to-work conflict Hassan et al., 2010). This means that the work-to-family conflict is more frequently experienced in higher intensity and family domain is more frequently affected by work. On the other hand, no conclusions can be drawn on the direction of crossover effect due to limited scientific evidence and some studies have tested Psychological Thought South-West University "Neofit Rilski" 2022, Vol. 15(1), 53-80 https://doi.org/10.37708/psyct.v15i1.614 only one direction of crossover effect Liu & Cheung, 2015). By collecting dyadic data from both spouses, the crossover effect can be examined and serves to add another level of analysis to spillover effect.
In addition, Liu and Cheung (2015) report that studies investigating the crossover effect in Chinese dual-earner couples are rare and under-investigated compared to the spillover effect. Malaysian Chinese women are also the most vulnerable group to experience workfamily conflict as compared to Malaysian Malay and Indian women (Mahpul & Abdullah, 2011). Recognizing the need for empirical evidence of crossover effect within Chinese dualearner couples, Liu and Cheung (2015) conducted a study investigating crossover effect of work-to-family conflict on spouses' psychological strain, as well as life, marital, and job satisfaction among 361 dual-earner couples in China. Actor-partner interdependence modelling (APIM) was used to analyse the dyadic data. Actor effect was significant where work-family conflict was positively correlated to psychological strain and negatively correlated to life, marital, and job satisfaction. No significant partner effect was found for work-family conflict on psychological strain, and life and job satisfaction. This means that although husbands and wives' work-family conflict was positively correlated to their own psychological strain, the work-family conflict had no significant correlation to their partner's psychological strain. The partner effect for husbands' work-family conflict had a significant negative correlation with wives' marital satisfaction. However, no partner effect for wives' work-family conflict was found. It is speculated that women seem to be better able to compartmentalise their family roles which allows them to maintain emotional boundaries and not allow their stress to affect their partners. On the other hand, it may be that men are less sensitive to their partner's stressors, which justifies the findings of this study (Liu & Cheung, 2015). The spillover effect was not studied in this study.
There is also empirical evidence of spillover and crossover effects from Taiwanese sample, focusing on how work-to-family conflict mediates the relationship between emotional exhaustion and marital satisfaction. The study was done by recruiting 226 employees and partner dyads from different work settings such as manufacturing, electronics, banking, service industries, and public organizations through convenience sampling. The mean age for males was 39 while for females, it was 38. Both spillover and crossover effects were found. Emotional exhaustion at work was positively correlated to work-to-family conflict and negatively correlated to marital satisfaction. The spillover effect of work-to-family conflict on marital satisfaction was significant. As for crossover effect, an employees' emotional exhaustion was positively correlated to their partner's family-to-work conflict and was Psychological Thought South-West University "Neofit Rilski" 2022, Vol. 15(1), 53-80 https://doi.org/10.37708/psyct.v15i1.614 significantly correlated to their partner's marital satisfaction (Liang, 2015). This study reflects how a person's experiences at work can influence and interfere with their spouses' marital satisfaction in both husbands and wives.
The studies discussed above have several limitations, such as investigating only a unidirectional effect of work-family conflict upon marital satisfaction, testing either the spillover or crossover effects, involving only one partner in the study and not the couple, and actor-partner interdependence model (APIM) analysis not being conducted on dyadic data collected. Another limitation is that the various studies investigating gender differences in spillover and crossover effects on marital satisfaction produced conflicting results.
Furthermore, the methodologies used in previous research that examines marital and family therapy have limitations in that analysis at couple-level is lacking (Liu & Cheung, 2015;Parker et al., 2016). Considering the suggestions from past research, this study aims to use actor-partner interdependence model (APIM) by collecting dyadic data to examine the crossover effect of work-family conflict on the other partner's marital satisfaction using a sample of Malaysian Chinese dual-earner couples.

The Present Study
This study attempts to address the gaps by including measures of both directions in workfamily conflict (work-to-family and family-to-work) as well as dyadic data from both spouses.
Although the association between the spillover and crossover effects and marital satisfaction has been well established, few existing studies have integrated both directions of workfamily conflict and their link to marital satisfaction using dyadic data. Based on the existing theories and previous literatures, it is plausible to propose that gender role attitude is associated with marital satisfaction. This is because individuals with egalitarian gender role attitude see family as a shared responsibility between spouses. In addition, it may be further hypothesized that egalitarian gender role attitude may moderate the negative effects of work-family conflict on marital satisfaction by interacting on the link between them, thus serving as a protective factor. Again, this hypothesis is based on the argument that individuals holding an egalitarian gender role attitudes perceive equality in family responsibilities, which serves as a protective factor in mitigating the negative impact of workfamily conflict upon marital satisfaction. To summarize, it is hypothesized that egalitarian gender role attitudes will moderate the negative effects of work-family conflict (work-to-family and family-to-work) on marital satisfaction. Specifically, it is hypothesized that work-family conflict (work-to-family and family-to-work) will not significantly impact marital satisfaction for individuals holding egalitarian gender role attitude.

Method Procedure
To be eligible to participate, couples had to be Malaysian Chinese residing in Kuala Lumpur and Klang Valley, working full-time, in their first marriage, staying in the same house, and participating voluntarily in this study. Couples were recruited using purposive sampling (meaning that participants should fulfilled the inclusion criteria) and snowballing technique (which is to say couples were invited to provide contact number of potential candidates for this study if they know any other candidates who fulfilled the inclusion criteria). Participants were recruited via two different methods; through the researcher's personal contacts and through personal contacts of friends and families. Couples were recruited by recruiting one spouse first, and then the other. Couples who were willing to take part in the study were given two sets of consent forms and surveys to be completed at home as a paper-pencil survey. The researcher followed up with the couples after two weeks in order to collect the completed survey package. Participants were told to sign the consent form and to complete the survey independently without any discussion with their spouses. No remunerations were offered to the couples. On average, couples spent 30 minutes to complete the questionnaires. A total of 150 couples completed the surveys, 135 couples returned completed questionnaires from both spouses. Surveys from one couple were excluded from data analysis because there was a significant amount of missing data in the responses. The final sample size was 134 couples (268 individuals).

Participants
The sample in this study aged from 27 to 63 years and the mean age for husbands was 46.33 (SD = 7.9) and the mean age for wives was 43.53 (SD = 7.1). Compared to past studies (e.g. Liang, 2015;Sandberg et al., 2012), the sample is older and husbands are older than their spouses generally. The mean years of marriage was 14.2 years. More than half (58.82%) of the husbands and wives (68.33%) hold diploma and above qualifications.
The majority (90.70%) of the sample had at least one child. In terms of occupation, husbands are mostly working as professionals (24.17%) and in the services and sales industry (23.33%). On the other hand, wives in this sample are mostly working as clerical support staff (39.17%) and professionals (30.00%). On average, wives (M = RM5185.23) earn less than husbands (M = RM7016.86). Two-thirds of the wives are earning less than RM5000.00 per month (75.23%) while 53.98% of husbands earn more than RM5000.00 per month.
In the latest 2019 Household Income and Basic Amenities Survey Report, Malaysians are grouped into three different groups based on their monthly household income on a continuum of 100%; the B40 refers to the group of families who earn the least among the population; then, the M40 refers to the families with monthly household income in the middle 40%, while T20 refers to the top 20% of the population with the highest monthly household income. The income thresholds for B40 was RM4849.00 and below; the M40's income threshold was between RM4850.00 to RM10959.00; and the T20's earn more than RM10960.00 per month. In terms of the percentage, 36.45% of families in Malaysia fall into the T20 category, 52.34% of the families are categorized as M40, and only 11.21 % of the families fall into the B40 category (DOSM, 2020). The monthly household income of the sample is categorized accordingly to reflect the household income survey categories. Based on the education level and average household income, it is concluded that the sample is make up of educated, middle-and upper-class Malaysian Chinese.

Measures
Gender role attitude. Gender role attitude was assessed using a one item question as used in Bernhardt and Goldscheider's (2006) study. This question was also adopted in Olah and Gahler's (2014) study. The question, "what do you think would be the best arrangement for a family with preschool children?" was used. The four alternatives are either, "only the man works, and the woman takes main responsibility for home and children" or, "both work, but the woman works part-time and takes the main responsibility for home and children" or, "both work, but the man works part-time and takes the main responsibility for home and children" or, "both parents work roughly the same hours and share the responsibility for home and children equally". Participants who answered the first two options were categorized as holding "traditional" gender role attitudes while participants who answered the third and fourth option were categorized as holding "egalitarian" gender role attitudes. Thus, gender role attitude is a dichotomous variable.
Work-family conflict. Work-family conflict was assessed with the 10-item Work-family and Family-work Conflict Scales (Netemeyer et al., 1996) with two subscales measuring work-tofamily and family-to-work conflicts (5-item in each subscale). Participants were asked to think about how the demands and time from work affected their family life and vice-versa. A 7-point Likert scale (where 1 = strongly disagree and 7 = strongly agree) was used to indicated how much the participants agreed or disagreed with statements such as, "Things I want to do at home do not get done because of the demands my job puts on me" for the work-to-family conflict subscale and "I have to put off doing things at work because of demands on my time at home" for the family-to-work conflict subscale. This scale has been found highly reliable in past research. Similarly, this scale also showed high reliability in the current study (α = .910 for husband's work-to-family conflict subscale [5-item], α = .917 for wife's work-to-family conflict subscale [5-item], α = .898 for husband's family-to-work conflict subscale [5-item], and α = .918 for wife's family-to-work conflict subscale [5-item]).
Marital satisfaction. Marital satisfaction was assessed using the Quality Marriage Index (QMI; Norton, 1983), which measures a person's global evaluation of their marriage. This scale consists of six items and participants were required to rate evaluative item such as "My relationship with my partner makes me happy". The first five items were rated on a 7-point Likert scale from 1 (I do not agree at all) to 7 (perfect). The last item required participants to rate their degree of happiness in their marriage from 1 (very unhappy) to 10 (totally perfect).
A high score indicates a high level of marital satisfaction and happiness while a low score Psychological Thought South-West University "Neofit Rilski" 2022, Vol. 15(1), 53-80 https://doi.org/10.37708/psyct.v15i1.614 represents a low level of marital satisfaction and a low sense of happiness in the marriage. This scale has been found to be highly reliable for husbands and wives (α = 93 [6-item] and α = .96 [6-item] respectively; Dennison et al., 2014) in past research, as did the current study (α = .969 for husbands [6-item] and α = .965 for wives [6-item]).

Analyses
The moderating effect of egalitarian gender role attitudes in the association between workfamily conflict (work-to-family and family-to-work) and marital satisfaction was analysed using the Process Macro (version 3.5) (Hayes, 2018). The APIM was used to analyse the dyadic data with partial least squares in SmartPLS Version 3.0 software.

Moderating effect of Egalitarian Gender Role Attitude
Among husbands, 94 (70.1%) reported holding egalitarian gender role attitude, 37 (27.6%) reported holding traditional gender role attitude, and three did not report their gender role attitudes (2.2%). As for wives, 106 (79.1%) reported holding egalitarian gender role attitude and 27 (20.1%) reported holding traditional gender role attitude. One of the participants among wives did not report her gender role attitude (0.7%). To test the hypothesis that egalitarian gender role attitude moderates the relationship between work-family conflict (work-to-family and family-to-work) and marital satisfaction, moderation analysis was conducted. The first moderation analysis was conducted between work-to-family conflict and -.088, p = .930. Again, the interaction term of egalitarian gender role attitude and work-tofamily conflict was not significant, ΔR² = .000, F (1, 264) = .007, p = .930. Thus, it was concluded that Hypothesis 1 is not supported; egalitarian gender role attitude does not moderate the relationship between work-family conflict (work-to-family and family-to-work) and marital satisfaction.

Actor-Partner Interdependence Modeling (APIM)
The measurement model was established after verifying the variables' (work-family conflict and marital satisfaction) Cronbach Alpha and composite reliability (CR) for internal consistency, and average variance extracted (AVE) for convergent validity (Hair et al., 2016). Table 2 displayed the scale properties of the variables. The acceptable level for Cronbach Alpha is more than 0.8, CR of more than 0.8 and AVE of more than 0.5 (Hair et al., 2016).
Based on the results, the internal consistency and convergent validity values were acceptable and the measurement model was established. In addition, the discriminant validity according to the Fornell and Larcker (1981) was also established where the AVE was greater than the squared of correlation coefficients between the variables (Table 3). Results showed that the discriminant validity was accepted for all variables.

Psychological Thought
South-West University "Neofit Rilski" 2022, Vol. 15(1), 53-80 https://doi.org/10.37708/psyct.v15i1.614  Table 1 and 4 showed the descriptive statistics and correlations between the studied variables. Correlation analysis showed a significant correlation between work-to-family and family-to-work conflicts on marital satisfaction for both husbands and wives. However, no significant correlations between wives' work-to-family and family-to-work conflicts on husbands' marital satisfaction were found. On the other hand, husbands' work-to-family and family-to-work conflicts are significantly correlated to wives' marital satisfaction. Table 4.

Correlations between Work-Family Conflict (work-to-family & family-to-work) and Marital Satisfaction among Husband (H) and Wife (W) (N = 134 couples)
Variables 1 2 3 4 5 6 1. Work-to-family conflict (H) -2. Family-to-work conflict (H) .607** (p =.000) Subsequently, the structural model established is used to analyse the significance of the path coefficients through bootstrapping for actor and partner effects. The bootstrapping subsampling was set at 5000 (Hair et al., 2016) and the results of the basic APIM are presented in Table 5. For actor effect, husbands who experienced higher work-to-family conflict were less satisfied with their marriage (b = -.215, p = .023). After adjusting for partner effects, husbands' marital satisfaction decrease by .215 with each 1-point increase in workto-family conflict. However, wives' actor effect of work-to-family conflict on marital satisfaction was not significant. On the other hand, wives who experienced higher family-towork conflict were less satisfied with their marriage (b = -.227, p = .024). In other words, after adjusting for partner effect, wives' marital satisfaction decrease by .227 with each 1-point increase in family-to-work conflict. Unlike work-to-family conflict, the actor effect of family-to-

Discussion
The findings highlighted many critical facets of the relationships between the variables being addressed in the study. The moderating effect of egalitarian gender role attitude on workfamily conflict and marital satisfaction is not significant. This implies that egalitarian gender role attitude did not buffer the negative impact of work-family conflict on marital satisfaction among the participants. A possible explanation from the finding is egalitarian gender role attitudes do not necessarily translate into behaviours as Blaisure and Allen (1995) found that an ideology of marital equality does not necessarily translate into practice of marital equality.
Along these lines, Olah and Gahler (2014) found that some individuals may hold egalitarian gender role attitudes but do not share the household responsibilities. Even when wives are holding full time paid jobs, they still need to struggle with their family responsibilities, this is consistent with findings from Oshio et al. (2013). Additionally, as much as Malaysian women enjoy more freedom, they remained heavily burdened with household chores and reproductive roles (Hirschman, 2016). Women are still largely responsible for household chores despite being employed and contributing financially to the family (Noor, 2006;Oshio Psychological Thought South-West University "Neofit Rilski" 2022, Vol. 15(1), 53-80 https://doi.org/10.37708/psyct.v15i1.614 et al., 2013). Future studies may include number of hours spent in household chores in order to measure the consistency between attitudes and behaviours and how it contributes to the level of marital satisfaction.
The hypotheses were partially supported for the actor effects. Results from the APIM showed significant actor effect of work-to-family conflict on marital satisfaction for husbands but not for wives. As for the family-to-work conflict, there was a significant actor effect on wives' marital satisfaction but not for husbands. Hypothesis 2 is partially supported. The contrary actor effects of husbands and wives of work-family conflict on marital satisfaction entail interesting discussion. Past studies suggested work domain boundary is more permeable than the family domain boundary for women (i.e. women are more likely to allow family responsibilities matters to seep into their work domain) while the family domain boundary is more permeable for men (i.e. men are more likely to allow work responsibilities to seep into their family domain) (Pleck, 1977). However, other studies reported no gender differences with regard to the permeability of work and family boundaries although family boundaries are generally more permeable as compared to work boundaries (Eagle et al., 1997;Frone et al., 1992). In this study, consistent findings were found in the gender differences on the permeability of work and family boundaries where the family domain was more permeable for husbands while the work domain was more permeable for wives.
Despite the permeability of the domains were consistent with Pleck's (1977) suggestion, but the results also suggested that husbands were less satisfied with their marriage when their work interferes with their family and wives were less satisfied with their marriage when their family interferes with their work. It might seem that husbands are prioritizing their family over work more while wives are prioritizing their work over family. This is inconsistent with the traditional-male-breadwinner-family model. As it was proposed that despite the Malaysian society is moving towards a dual breadwinner paradigm (Bakar & Abdullah, 2013;Zaimah et al., 2013), but Malaysian Chinese dual-earner couples still hold traditional Confucian values where men are expected to provide for the family while women are expected to take care of the family (Ng et al., 2009;Noor, 1999). The findings from this study do not support past studies. These inconsistencies may be explained by a shift in gender role attitudes. The rise of dual-earner families in Malaysia contributed to the emergence of changes in gender roles, marital conflict, and family financial decisions (Zaimah et al., 2013). The changes are likely consistent with the general trend happening in East Asia. Boehnke (2011)  more egalitarian gender role attitude over the past 50 years. This supported the notion of gender role attitudes shift across time (King & King, 1985). Thus, it is plausible to suggest that Malaysian Chinese are shifting towards a more egalitarian gender role attitude based on the findings. These findings also supported the notion where a new work-family role system with different boundaries between the work and family domain (Pleck, 1977) where family needs are prioritized is required in order to manage the increased equality in the sharing of work and family roles by women and men.
In terms of the spillover effect, findings were consistent with past studies where the spillover effect of work-family conflict on marital satisfaction was significant (Liang, 2015;Liu & Cheung, 2015;Sandberg et al., 2012). As much as work-to-family conflict is more frequently reported as compared to family-to-work conflict (Frone et al., 1992;Hassan et al., 2010), but the negative association for both work-to-family and family-to-work conflict on marital satisfaction remained significant. The crossover effect of work-family conflict on marital satisfaction was not significant. Therefore, hypothesis 3 was not supported. The results were partially consistent with Liu and Cheung's findings where no crossover effect of wives' workfamily conflict on husbands' marital satisfaction was found. However, the husband's workfamily conflict was negatively correlated with wives' marital satisfaction (Liu & Cheung, 2015). Liu and Cheung (2015) justified their findings to women' better ability to compartmentalising their family and work roles where they are able to maintain emotional boundaries.
Subsequently, they did not allow their stress to affect their partners' marital satisfaction. In addition, it may also because men are less sensitive to their partner's stressors (Liu & Cheung, 2015). Similar findings were found in this study where men also showed a similar pattern. Wives' work-to-family conflict is not negatively correlated to husbands' marital satisfaction; the crossover effect was not significant as well. It could be that both husbands and wives used different coping strategies to cope with the stress at an individual level and tried not to allow their personal stress to affect their spouses.

Limitations and implications
The cross-sectional design used is unable to display the causal relationship between workfamily conflict and marital satisfaction and the results cannot be generalized to the other samples. The sample in this study was recruited using purposive sampling and snowballing technique which may result in sampling bias. As it is not realistic to recruit the participants for dyadic data via random sampling, future studies may choose stratified sampling and draw careful conclusions based on the sample recruited due to the limitation in generalizability of the findings. Furthermore, the sample of this study's age range is large and the marriage Psychological Thought South-West University "Neofit Rilski" 2022, Vol. 15(1), 53-80 https://doi.org/10.37708/psyct.v15i1.614 length is longer as compared to previous studies. The sample in this study were mainly middle and upper class Malaysian Chinese dual-earner couples with two to three children.
For couples completed the questionnaire at home, additional steps should be taken to make sure that husbands and wives completed the questionnaire independently. Future studies may consider having the couples complete the questionnaire in front of the researcher or in a lab rather than providing the option of completing the questionnaire at home. The correlations between variables may be explained by other factors such as the presence of young children at home which increased the demands of the family domain. Furthermore, types of work such as shift work and highly stressful career may have different demands and affect marital satisfaction. Future studies may consider studying specific workgroups and how work-family conflict have an impact on marital satisfaction. Moreover, self-reported data collected in this study are vulnerable to common method variance. However, common method variance was argued to be overestimated (Edwards, 2008). Another limitation of this study design is the socially desirable responding which is inevitable in questionnaire studies (Paulhus, 2002). Also, the sample size might limit the power to detect possible partner effects of work-family conflict.
Despite the limitations discussed, the present study adds to the literature by using the APIM approach which accommodates non-independence dyadic data and estimates both actor and partner effects simultaneously. The bidirectional effects of work-family conflict (work-tofamily and family-to-work) on marital satisfaction among Malaysian Chinese dual-earner couples were examined in this study. The data partially supported previous findings on the spillover effect of work-family conflict on marital satisfaction at both the actor and partner level. This study also contributed to the work family literature gap of Western focus by examining spillover and crossover effects among Malaysian Chinese dual-earner couples.
Future research can further explore these issues by testing possible mediators of this process continuing utilizing dyadic data analysis and to use a longitudinal design to examine the long-term effect of work-family conflict on marital satisfaction. At larger scale, by understanding the work family interface, mental health practitioners will be able to use the findings from this study in their clinical practice when trying to understanding marital satisfaction and providing family and marriage therapy. This is because a family is a small social unit and it is important to note that family life is not isolated from individuals' work life. Couples might benefit from intervention programs which enhance their skills in prioritizing or adjusting the hierarchy of importance between work and family to mediate the negative impact of work-family conflict on marital satisfaction and to preserve marital stability.

Psychological Thought
South-West University "Neofit Rilski" 2022, Vol. 15(1), 53-80 https://doi.org/10.37708/psyct.v15i1.614 The findings from this study can also be used and served as guidelines when companies are developing their policies especially when they are establishing and/or operating their companies in Asian countries. Furthermore, due to globalization, there are more and more multinational companies establishing their companies in Malaysia, therefore, by understanding how the work and family interface is experienced among Malaysian Chinese employees, the information is particularly useful for the management to ensure the company's smooth daily operation by designing and implementing more family-friendly work environment.

Conclusion
Overall, the findings highlighted the complex relationship between work-family conflict and marital satisfaction with the actor-partner interdependence model. The issues involved in the work-family conflict are likely to continue with the emerging of dual-earner couples. More research from different cultural backgrounds is needed to provide a clearer picture on the relationship between work-family conflict and marital satisfaction among dual-earner couples.

Funding/Financial Support
The authors have no funding to report

Other Support/Acknowledgement
The authors have no support to report.