The authors thank the Center For Creative Leadership for providing the multisource data used in this research. We also thank the Editors of the Special Issue and Kristen M. Shockley for their helpful comments on earlier versions; Cristina Arroyo Rodriguez and Christine R. Smith for their help as Research Assistants; and Phillip Braddy, John Fleenor, and Michael Campbell from the Center For Creative Leadership for their support and help with the data files. Portions of this study were presented at the International Conference on Work–Life: Cross-National Conversations and Context Theorizing in Work–Life Research, that took place in May 2011 in Paris, France; and at the Annual Meeting of the Society for Industrial and Organizational Psychology that took place in May 2009 in New Orleans, LA, USA.
Gender Egalitarianism and Work–Life Balance for Managers: Multisource Perspectives in 36 Countries
Article first published online: 11 JUN 2013
© 2013 International Association of Applied Psychology
Special Section—International Perspectives on Work and Family By Casper, W.J., Allen, T.D. & Poelmans, S.A.Y.
Volume 63, Issue 1, pages 96–129, January 2014
How to Cite
Lyness, K. S. and Judiesch, M. K. (2014), Gender Egalitarianism and Work–Life Balance for Managers: Multisource Perspectives in 36 Countries. Applied Psychology:An International Review, 63: 96–129. doi: 10.1111/apps.12011
- Issue published online: 6 NOV 2013
- Article first published online: 11 JUN 2013
Work–life balance has important implications for both personal well-being and work-related outcomes. This study investigated gender differences in multisource ratings of work–life balance, based on self-reports and supervisors' appraisals of 40,921 managers in 36 countries. Based on a combination of theoretical ideas from social role theory (Eagly & Wood, 2012), prior work–life literature, and gender egalitarianism as a cultural dimension related to societal gender roles, the study tested gender egalitarianism as a moderator of cross-national variations in these gender differences. Based on multilevel (HLM) analyses, results showed more cross-national variation by ratee gender in supervisors' appraisals than self-reports, suggesting that supervisors' perceptions reflected greater influence of societal gender stereotypes. Supervisors rated women lower in work–life balance than men in low egalitarian countries, but similar to men in high egalitarian countries, and only appraisals of women varied depending on egalitarian context. Country gender egalitarian values explained the majority of variation in supervisors' appraisals of women's work–life balance, whereas women's self-reported balance was linked to objective gender inequalities. Taken together, the findings show that supervisors' perceptions of employees' work–life balance differed by ratee gender and country context, with important implications for work–life theory and practical implications for global employers.
There is increasing interest in work–life balance, based on the idea of a “balance between work and the rest of life” (Guest, 2002) among work–life scholars, as well as working women and men, employers, and policy makers. The focus on work–life balance reflects the value placed on overall quality of life as well as concerns that increased work demands are leading to “work–life imbalance”, with negative effects on the well-being of individuals and their families (Guest, 2002; Marks & MacDermid, 1996; Mayo Foundation for Medical Education and Research, 2012; Wada, Backman, & Forwell, 2010). Moreover, scholars have suggested that these issues are complex: women's experiences of balance are thought to differ from men's, and national cultural values and norms may affect these experiences, underscoring the need for more cross-national research to investigate these issues (Aycan, 2008; Greenhaus & Allen, 2011; Lero & Lewis, 2008). Accordingly, the present study used a 36-country sample to examine gender differences in work–life balance and cross-national variation in these gender differences as a function of national gender egalitarianism, a particularly relevant aspect of national culture.
Gender, gender differences, and gender role beliefs have long been considered important for understanding work–life issues (Eby, Casper, Lockwood, Bordeaux, & Brinley, 2005). In fact, Hofstede (1980) argued that one of the most important ways that societies differ is in the types of roles that are prescribed for women and men. This concept is captured in the cultural dimension termed “gender egalitarianism”, defined as “beliefs about whether members' biological sex should determine the roles that they play in their homes, business organisations, and communities” (Emrich, Denmark, & Den Hartog, 2004, p. 347). Low gender egalitarian cultures are characterised by beliefs in the traditional gendered division of labor, such that men are viewed as breadwinners and women are viewed as caretakers and mothers, whereas in high gender egalitarian cultures there is less adherence to these traditional gender roles and more similarity in women's and men's involvement in work and non-work domains (Emrich et al., 2004).
There is theoretical reason to suspect that the gender egalitarianism of a country may relate to the extent to which women's and men's work–life balance differs (Greenhaus & Allen, 2011; Lyness & Kropf, 2005); however, there is limited cross-national research testing this idea. For instance, traditional low egalitarian gender role beliefs have been associated with greater difficulty for women to achieve work–life balance due to inherent incompatibilities in carrying out their work and domestic responsibilities (i.e. greater work–family conflict) while men may have less difficulty achieving balance because their work involvement also fulfills their traditional family responsibility as breadwinners (Wada et al., 2010). The first goal of the present study was to more thoroughly assess the relationship between gender egalitarianism and gender differences in self-reported work–life balance across 36 culturally diverse countries.
Moreover, in addition to the implications of work–life balance for personal well-being (Marks & MacDermid, 1996), previous research suggests that it may also have implications for work-related outcomes. Specifically, Lyness and Judiesch (2008) found that supervisors' perceptions of their subordinates' work–life balance were positively related to appraisals of future career advancement potential. Similarly, although they did not directly measure supervisors' perceptions of subordinates' work–life balance, Carlson, Witt, Zivnuska, Kacmar, and Grzywacz (2008) found that supervisors gave higher performance ratings to subordinates whom they perceived to be higher in family-to-work enrichment, which the researchers and some other scholars (e.g. Frone, 2003) consider to be a component of work–family balance, a construct that is related to work–life balance. On the negative side, two studies found that supervisors' perceptions of their subordinates' family-to-work conflict were negatively related to appraisals of work performance and promotability (Carlson et al., 2008; Hoobler, Wayne, & Lemmon, 2009); and work–family conflict has also been considered a component of work–family balance (Carlson et al., 2008; Frone, 2003). Taken together, these studies suggest that not only do supervisors form impressions about their subordinates' non-work involvement, but also that supervisors' judgments about how well their subordinates balance work and non-work responsibilities are positively linked to appraisals of work performance and promotability.
Although there is some initial evidence to support this link, little research has been conducted to clarify the process by which supervisors make assessments of their subordinates' work–life balance. In one of the few exceptions, Hoobler et al. (2009) compared US supervisors' and employees' ratings of the employees' work–family conflict. Their results revealed gender differences in supervisors' appraisals that are consistent with traditional gender role assumptions, such that female subordinates were rated higher in family-to-work conflict than male subordinates, despite the fact that the subordinates' self-reported conflict did not differ by gender. This study also underscores the importance of gender for understanding work–life constructs as they are perceived by other people. Building on this idea, the second goal of the present study was to investigate whether supervisors' appraisals of subordinates' work–life balance differed depending on the subordinate's gender, and to determine whether cross-national variation in these gender patterns was moderated by the gender egalitarianism of the countries.
We tested these research questions with multi-source ratings of work–life balance for a sample of over 40,000 managerial ratees working in 36 countries representing diverse global regions (see the Appendix for a list of countries organised by regions). Following recommendations by Taras, Rowney, and Steel (2009), who pointed out that cross-cultural research has been overly reliant on measures of cultural values, we used four measures that captured different aspects of gender egalitarianism and allowed us to assess their relative utility. These country-level measures included Gender Inequality Index (GII) scores, measuring objective gender inequalities in important realms, such as workforce opportunities (United Nations Development Programme, 2010), Project GLOBE gender egalitarian value and practice scores (House & Javidan, 2004), and a composite of gender-related World Values Survey items (World Values Survey 1981–2008 Official Aggregate v.20090901, 2009) that have been used in prior research as measures of patriarchal norms and values (Seguino, 2007).
We believe that the present study makes both theoretical and practical contributions to the literature. Our research was designed to provide one of the first empirical tests of predictions based in part on recent cross-cultural extensions of social role theory by Eagly and her colleagues (Eagly, 1987; Eagly & Wood, 2012; Wood & Eagly, 2002, 2010) about the effects of national gender role beliefs on both women's and men's self-reported work–life balance. Second, we extended prior cross-national work–life literature by developing additional theory to explain how societal gender egalitarianism might influence other people's perceptions, and tested these predictions with supervisors' appraisals of their female and male subordinates' work–life balance across countries that differed in gender egalitarianism. In addition to the theoretical contributions, the empirical findings based on a diverse global sample should enhance our understanding of these issues and also have practical utility, particularly for multinational corporations (MNCs) that operate across cultures. For example, in light of the prior evidence that supervisors' views of subordinates' work–life balance relate to their performance and promotability evaluations, employers need to be aware of whether these balance perceptions vary depending on subordinates' gender or cultural context. In the following sections we will review the relevant literature and introduce our hypotheses.
Although the terms “work–life balance” and “work–family balance” are sometimes used interchangeably, work–life balance, the focus of the present study, is a more inclusive term. Whereas “family” focuses on caregiving and family-related activities and responsibilities, “life” broadens the non-work domain focus to include family as well as other types of personal activities and interests, such as education, community involvement, and religious activities (Greenhaus & Allen, 2011). In addition, the concept of “work–life balance” is favored by employers and policy makers as it is considered to be more gender-neutral than work–family balance, and is also more inclusive of employees regardless of their family circumstances or involvement (Lewis & Campbell, 2008).
Although work–life balance has emerged as an increasingly frequent topic of study and discussion, there is not yet a well-accepted definition of the construct (Guest, 2002; Lewis & Campbell, 2008; Wada et al., 2010). In a recent review, Greenhaus and Allen (2011) identified three commonly used conceptualisations of work–family balance (which could similarly apply to work–life balance): (1) “the absence of work–family conflict”, (2) “high involvement across multiple roles”, and (3) “high effectiveness and satisfaction across multiple roles” (p. 172). Greenhaus and Allen developed a more comprehensive conceptualisation of work–family balance that incorporated key elements from these prior perspectives, such as involvement in multiple roles, and also extended these ideas by acknowledging the importance of individual values and the dynamic nature of the construct; thus, they proposed that work–family balance should be defined “as an overall appraisal of the extent to which individuals' effectiveness and satisfaction in work and family roles are consistent with their life values at a given point in time” (p. 174).
However, particularly in work settings, other people, such as supervisors, may have varying degrees of personal contact and insights about their work colleagues' personal lives, and these limitations need to be taken into account in determining how to conceptualise work–life balance. Thus, consistent with the second of the three approaches identified by Greenhaus and Allen (2011), we conceptualised work–life balance as involvement in multiple roles. Specifically, we used a measure that defines work–life balance as “balances work priorities with personal life so that neither is neglected” (Center for Creative Leadership, 2004). Also, in line with Guest's (2002) distinction between subjective and objective approaches, our conceptualisation of work–life balance is subjective as it was based on multisource perceptions, i.e. of the ratee and his or her supervisor. This approach is similar to subjective conceptualisations used in some prior research in work settings (e.g. Hill, Hawkins, Ferris, & Weitzman, 2001).
It is also possible that societal gender roles and norms influence an employee's preferences or involvement with work and family or other non-work roles. In the following sections we will expand upon these ideas and review prior literature, such as Eagly and colleagues' (Eagly, 1987; Eagly & Wood, 2012) social role theory that offers an explanation of how societal gender roles develop, which we later use to explain why there might be gender differences in work–life balance experiences and perceptions. In addition, we will review the limited prior literature about gender egalitarianism, an important aspect of national context that is helpful for explaining cross-national variations in societal gender roles, gender stereotypes, and gender differences in work–life balance experiences.
Gender Roles, Gender-Related Cultural Values, and Work–Life Balance
The focus on gender in work–family theories dates back to seminal ideas from the 1940s that the workplace and family are “separate spheres”, and that “families, institutions and society all work best when men and women specialize their activities in separate spheres, women at home doing expressive work and men in the workplace performing instrumental tasks” (MacDermid, 2005, p. 21). Eagly and her colleagues expanded on these ideas in their social role theory which argues that as people observe men and women in certain roles, these collective observations result in shared gender stereotypes, such that women and men are also thought to possess corresponding attributes that make them well suited to perform their prescribed gender roles (e.g. Eagly & Wood, 2012). Moreover, according to this theory, various societal processes, such as socialisation and societal members' expectations, facilitate conformity to prescribed gender roles, and ultimately women and men are thought to internalise their society's normative gender role beliefs causing them to engage in self-regulation of their own behavior which further reinforces conformity to societal gender roles (Eagly & Wood, 2012).
There is also evidence from studies in the US that societal gender stereotypes affect workplace behavior. For example, both women and men tend to hold jobs requiring skills thought to be congruent with their stereotypic attributes, as illustrated by women holding jobs that require caretaking or domestic behaviors and men holding jobs that require assertive behaviors (Dipboye, 1985; Eagly & Wood, 2012; Heilman, 2001). In addition, co-workers and supervisors reward work colleagues who conform to gender norms, for example, by cooperating with them, but react negatively toward those who violate gender norms, with sanctions ranging from subtle social rejection to more extreme outcomes, such as job loss (Eagly & Wood, 2012).
Although social role theory (Eagly, 1987) was originally developed to explain gender role-related beliefs and behaviors in the United States, Eagly and her colleagues have extended it to address cross-national variations in gender norms and behaviors (Eagly & Wood, 2012; Wood & Eagly, 2002, 2010). Specifically, there are thought to be variations across countries in the extent to which women and men adhere to the traditional homemaker–breadwinner division of labor, which are in turn reflected in both normative societal beliefs about appropriate gender roles and corresponding beliefs about the stereotypic attributes of women and men (Eagly & Wood, 2012).
Although limited, there is some prior cross-national research supporting these ideas about the influence of societal gender role beliefs. For example, a 23-country study showed that aggregated measures of national attitudes about work-related gender norms predicted women's actual labor force participation rates under various conditions, such as motherhood (Treas & Widmer, 2000). In addition, a 31-country study found that a country-level measure reflecting the societal history of maternal employment predicted the extent to which fathers and mothers shared housework responsibilities (Treas & Tai, 2012). Thus, the ideas from the recent cross-national extensions of social role theory fit well with cross-cultural research findings showing that there are indeed variations in societal gender roles and norms, and that these societal beliefs predict actual behavior.
As we noted earlier, gender egalitarianism is the dimension of national culture that reflects the degree of differentiation in prescribed gender roles for men and women. This construct in combination with Eagly's social role theory suggests that cross-national variations in societal beliefs about gender roles, as captured by the extent of gender egalitarianism, shape not only women's and men's behaviors and ability to achieve work–life balance but also corresponding stereotypic beliefs about the attributes of women and men. Specifically, in traditional, low gender egalitarian cultures men are expected to prioritise work over family, and women are expected to prioritise family over work, whereas in more egalitarian cultures there is less differentiation of gender roles and expected priorities by gender (McDaniel, 2008). As we noted above, in low egalitarian cultures men's work involvement and long work hours are congruent with their traditional breadwinner responsibilities to provide for their families, which suggests higher levels of work–life balance. However, similar high work involvement by women is incongruent with societal expectations that they should focus on family responsibilities (Aryee, Srinivas, & Tan, 2005), thus making it much more difficult for women to achieve balance. In contrast, in high egalitarian cultures, there is more overlap in men's and women's gender roles, suggesting that both women's and men's involvement with paid work and caretaking responsibilities would be more accepted, thus leading to the prediction that women and men would have more similar levels of work–life balance.
There are several different approaches that have been used to measure the gender egalitarianism construct. First, the well-known Project GLOBE research, based on surveys of over 17,000 middle-level managers in 62 societies, asked respondents about both cultural practices, defined as “the way things are” in their societies, and cultural values, defined as “the way things should be” (Javidan, House, & Dorfman, 2004). The researchers aggregated these responses and created country-level scores for both gender egalitarian practices (GEP) and gender egalitarian values (GEV), which were generally positively related (House & Javidan, 2004; House, Hanges, Javidan, Dorfman, & Gupta, 2004). Second, the World Values Survey (WVS v.20090901, 2009) provides information about gender-related values that can also be aggregated to create country scores. The WVS measures of patriarchal (low egalitarian) values, which represent beliefs that men should be given preference over women in employment and other societal domains, have been shown to be related to objective measures of actual gender equality, such as women's representation in the labor force and parliament, and gender equality of wages, educational opportunities, and governmental support for parental leave policies and women's reproductive rights (Fortin, 2005; McDaniel, 2008; Seguino, 2007; Wernet, Elman, & Pendleton, 2005). Third, economists at the United Nations created a Gender Inequality Index (GII), which is a broad composite of objective measures of gender inequalities in each country, such as women's relative representation in the labor force and parliament (United Nations Development Programme, 2010). As mentioned earlier, consistent with Taras et al.'s (2009) recommendations, we used four measures in order to assess multiple aspects of gender egalitarianism including cultural values, practices, and objective measures of gender equality (i.e. observable cultural artifacts), in order to ensure that we had comprehensively captured the construct.
Gender Egalitarianism and Gender Differences in Self-Reported Work–Life Balance
Empirical studies of cultural values in the work–life literature are often limited to single countries, but there is some cross-national evidence to suggest that there are smaller gender differences in self-reported work–life balance and related constructs in countries with more egalitarian cultures. Of particular relevance, a 48-country IBM study of fathers and mothers examined balance-related constructs, such as work–family fit, measured as the ease or difficulty of managing work and personal/family life demands, in regions differing in egalitarianism (Hill, Hawkins, Martinson, & Ferris, 2003). Although they did not directly measure egalitarianism, their results showed that in the low egalitarian Asian region fathers reported higher work–family fit than mothers, whereas both fathers and mothers reported the highest levels of work–family fit as well as equal sharing of childcare responsibilities in the high egalitarian Scandinavian region. Similarly, Aryee et al.'s (2005) research in India, a low egalitarian country, found higher work–family balance for men than women, based on Frone's (2003) conceptualisation linking balance to both work–family enrichment and work–family conflict.
As these studies illustrate, there is some prior evidence that variations across societies in gender egalitarianism are related to the extent of differentiation in women's and men's gender roles. As we discussed earlier, according to social role theory, as societal members observe men and women in certain roles, their collective observations result in shared societal beliefs about gender roles and norms that specify appropriate behavior for women and men (Eagly & Wood, 2012). Moreover, according to this theory, several complementary processes all reinforce conformity to prescribed gender roles, including societal members' socialisation of their children and ongoing social rewards that shape behavior, resulting in internalisation of societal gender norms and self-regulation (Eagly & Wood, 2012). However, in contrast to these ideas, the relatively limited empirical research suggests that although societal values are influential, many individual factors, such as various work and family demands, and partners' division of domestic labor, are also important in determining work–life balance (Crompton & Lyonette, 2006; van der Lippe, Jager, & Kops, 2006).
Thus, based on our literature review, we thought that managers' self-reported work–life balance is influenced by salient cultural variables, such as gender egalitarianism, as well as their personal values and circumstances. Moreover, to the extent that men's and women's work–life balance experiences reflect personal values or circumstances that deviate from societal gender norms, there is greater within-country variation and correspondingly less systematic between-country variation in gender differences in work–life balance. Although it was impossible to predict the relative influence of country context as compared to individual factors in determining gender differences in work–life balance, because of the variation in gender egalitarianism across our diverse sample of countries, gender differences in work–life balance should also vary across countries. In addition, according to social role theory, men's and women's collective behaviors shape societal beliefs about gender roles, norms, and values, suggesting that there is some congruence among the various measures of gender egalitarian practices and values. Thus, we propose:
- Hypothesis 1: Societal gender egalitarian cultural practices and values moderate the relationship of managers' gender to their self-reported work–life balance, such that there are smaller gender differences in high gender egalitarian countries than in low gender egalitarian countries, with women reporting lower work–life balance than men in low gender egalitarian countries.
Supervisors' Perceptions of Subordinates' Work–Life Balance
As we explained in the introduction, there is recent research showing that supervisors' judgments about how well their subordinates are able to balance or benefit from their non-work involvement are related to appraisals of work performance and promotability (Carlson et al., 2008; Lyness & Judiesch, 2008). However, we know little about the determinants of supervisors' perceptions of their subordinates' work–life balance, and whether these perceptions vary depending on subordinates' gender in combination with national context. Although the empirical findings and related theories about the role of contextual characteristics as determinants of work–life balance may be relevant, they do not necessarily generalise to how context relates to perceptions of other people's work–life balance by work colleagues, such as supervisors.
As this aspect of the work–life literature is not well developed, we incorporated ideas from more general literature about social perceptions, and especially how shared societal stereotypes about demographic characteristics, such as gender, may influence perceptions of other people. In fact, there is a body of US literature showing that those in power, as is the case with supervisors, are more likely to base their perceptions of others on demographic stereotypes rather than individuating information, and that gender stereotypes often play a key role in these perceptions (see Fiske & Berdahl, 2007, for a review).
Of particular relevance to the present study, Hoobler et al.'s (2009) US finding that supervisors perceived female subordinates as experiencing more family-to-work conflict than male subordinates appears to fit with this explanation, suggesting that societal gender role beliefs influence supervisors' perceptions, and the researchers also attributed the gender differences to the supervisors' beliefs in traditional gender roles. Moreover, Hoobler et al. found no gender differences in the subordinates' self-reported family-to-work conflict, suggesting that the supervisors' appraisals were somewhat influenced by assumptions based on traditional gender stereotypes rather than individuating information about their subordinates' actual behavior.
Thus, there is evidence that those in power, such as supervisors, perceive subordinates based on societal beliefs about gender stereotypes (Fiske & Berdahl, 2007). Also, there are cross-national variations in gender egalitarian values and practices, reflecting the extent of differentiation in women's and men's gender roles and norms. Moreover, based on social role theory, we reasoned that through their influence on women's and men's behaviors, gender egalitarian values and practices influence corresponding societal stereotypes about women's and men's attributes (Eagly, 1987; Eagly & Wood, 2012). Thus, supervisors' appraisals should show more systematic variation in gender differences across countries than self-appraisals due to supervisors' greater reliance on societal gender stereotypes as compared to individuating information.
Therefore, taken together, the prior literature and theoretical ideas suggested that supervisors' appraisals of their subordinates' work–life balance differ depending on the subordinate's gender, and that the extent of these gender differences is moderated by gender egalitarian values and practices. Accordingly, we propose:
- Hypothesis 2: Societal gender egalitarian cultural values and practices moderate the relationship of ratee gender to supervisors' appraisals of their subordinates' work–life balance, such that there are greater perceived differences in women's and men's work–life balance in low gender egalitarian countries than in high gender egalitarian countries, with women rated lower in work–life balance than men in low gender egalitarian countries.
Participants and Procedures
Our multisource data were collected by the Center for Creative Leadership from 2000 to 2007 in connection with management development programs, using version 2 of the Benchmarks® multisource feedback instrument, available in eight languages (Center for Creative Leadership, 2004). Our sample was limited to cases with both the manager's self-rating and his or her supervisor's rating, and complete data (N = 40,921). The manager ratees worked in many different types of organizations, and were located in 36 countries (see Appendix), with 81 per cent of the sample from the US. One-third of the ratees were women, but only 21 per cent of the supervisors were women. The average age of the ratees was 42 years, their average organisational tenure was ten years, the majority (89%) had a Bachelor's or graduate degree, and 13 per cent were expatriates.
Each ratee's work–life balance was measured based on his/her self-report and his/her supervisor's appraisal, using three items from the “Balance between Personal Life and Work” scale in Section 1 of the Benchmarks® instrument; work–life balance was defined as “balances work priorities with personal life so that neither is neglected” (Center for Creative Leadership, 2004). Many studies have been conducted with the original and revised Benchmarks® instruments to examine their psychometric properties, including measurement equivalence across different translations of the instrument; the three items that we used have generally been found to be a sound measure (Center for Creative Leadership, 2004, 2010; Spangler, 2003; Zedeck, 1995). The items all refer to balancing career and non-work involvement, e.g. “Acts as if there is more to life than just having a career”. Similarly, the other items concern the extent of involvement in non-work activities and not allowing career commitments to interfere with personal life, and are similar to items used in prior organisational research (e.g. Hill et al., 2001). In the Benchmarks® instrument raters and ratees are instructed: “Please rate the extent to which this person displays each of the following characteristics” using a 5-point response scale (1 = not at all to 5 = to a very great extent). The three items were averaged, and the alpha coefficients for the scale were .82 and .78 for supervisor and ratee self-ratings, respectively. High scores reflected greater perceived balance of work and non-work priorities.
Ratee gender was coded 0 = female and 1 = male.
Gender egalitarianism was measured with country scores from multiple sources, based on data from the 1990s, prior to collection of the multisource ratings. There were two measures of gender egalitarian practices: (1) the 1995 Gender Inequality Index (GII), a broad, composite measure of objective indicators reflecting “gender disparities in reproductive health, empowerment and labour force participation” with high scores reflecting lack of gender equality (United Nations Development Programme, 2010, p. 26), and (2) the Gender Egalitarian Practices (GEP) measure from the Project GLOBE research (House et al., 2004), based on surveys of over 17,000 middle-level managers in 62 societies, which measured cultural practices, defined as “the way things are” and cultural values, defined as “the way things should be” using 7-point scales (Javidan et al., 2004). The Gender Egalitarianism dimension was defined as “the degree to which … a society minimises gender role differences while promoting gender equality” with low scores reflecting greater male domination and high scores reflecting greater female domination (House & Javidan, 2004; House et al., 2004). Information about scale validation, cross-cultural comparability, and correction for response bias is available in House et al. (2004).
There were also two measures of gender egalitarian values: (1) the Gender Egalitarian values (GEV) measure from the Project GLOBE research (House et al., 2004), where low scores reflect greater male domination and high scores reflect greater female domination (House & Javidan, 2004; House et al., 2004), and (2) a composite measure developed for the present study from three World Values Survey items (World Values Survey 1981–2008 Official Aggregate v.20090901, 2009) that have been used in prior research as measures of patriarchal norms and values (Seguino, 2007). The WVS items were: “When jobs are scarce, men should have more right to a job than women”, “On the whole, men make better political leaders than women do”, and “A university education is more important for a boy than a girl”; we standardised the three items (due to differences in response scales) and averaged the items to create country scores, alpha = .90; high scores represented stronger paternalistic attitudes and less gender equality (n = 32 countries; unavailable for Denmark, Greece, Ireland and Portugal).
Because we wanted to isolate the effects of country egalitarianism on ratee gender differences in self-reported work–life balance, we controlled for six demographic characteristics. These included four ratee characteristics that were shown in prior research to be related to self-reported work–life balance (Lyness & Kropf, 2005; Tausig & Fenwick, 2001): age and organisational tenure were measured in years, and education was measured as highest degree attained (coded 1 = less than Bachelor's degree, 2 = Bachelor's degree, 3 = Master's degree, and 4 = PhD or professional degree). We controlled for ratee expatriate status (coded 1 = yes and 0 = no) because expatriates' work–life balance might differ from that of local employees, and expatriates might not share local cultural beliefs. Also, we controlled for employment sector (coded public = 1, private and other = 0) as work–life policies and practices are likely to differ across employment sectors, and year of assessment as gender-related attitudes have changed over time (Seguino, 2007).
In analyses of supervisors' appraisals we used two additional control variables: supervisor's familiarity with the ratee (1 = I hardly know this person to 4 = I know this person extremely well) as this could affect supervisors' ability to assess subordinates' work–life balance, and the supervisor's gender (coded 0 = female, 1 = male), to control for gender differences in gender-related attitudes as reported in prior research (e.g. McDaniel, 2008).
We conducted multilevel analyses using hierarchical linear modeling (HLM) software version 6 (Raudenbush, Bryk, Cheong, Congdon, & du Toit, 2004) to test our hypotheses, with individual-level data about ratees and raters at level 1, and country characteristics at level 2. We used random coefficient regression analyses and allowed for random variation of the intercepts and the focal variable (ratee gender), as well as the other contextual variables, i.e. employment sector and year of assessment, as their effects differed across countries. We treated the other ratee and rater control variables as fixed effects to avoid correlated random effects that could confound the random effects of the focal predictor, i.e. ratee gender. Following procedures recommended for testing cross-level moderation effects (Hofmann & Gavin, 1998), we used group-centering for the level-1 variables, included country-level means for the level-1 variables in the level-2 intercept models, and used grand-mean centering for the level-2 variables.
We first analyzed unconditional multilevel models with no predictors to assess the relative variation between and within countries in work–life balance ratings, and computed the overall intraclass correlation coefficients for each rating source, as well as separate coefficients by ratee gender. We found that from 5 per cent to 7 per cent of the variance in the supervisors' appraisals was due to differences between countries, the overall ICC(1) = .051, and the gender-specific ICC(1) = .073, .052 for ratings of women and men, respectively. These coefficients can be interpreted as small to medium effects for the influence of country context on supervisors' ratings, which in combination with the 36 country rwg values that ranged from .67 to .88, generally meet LeBreton and Senter's (2008) criteria to justify aggregation by country to conduct multilevel analyses.
However, less than 3 per cent of the variance in self-ratings was due to differences between countries, the overall ICC(1) = .027, and the gender-specific ICC(1) = .025, .027 for ratings by women and men, respectively. All of these are considered small effects. Also, country rwg values ranged from .63 to .82 (with six values below .70). Thus, although we conducted our planned multilevel analyses of the self-ratings, it is important to keep in mind that there was less systematic between-country variation in self-ratings than for supervisors' ratings.
The means, standard deviations, and correlations of the level-1 data are shown in Table 1. The relationship between supervisors' and ratees' own work–life balance appraisals was positive but modest in size, r = .33, p < .001, and the balance appraisals were not significantly related to ratee gender. The means, standard deviations, and correlations of the level-2 country characteristics are shown in Table 2. The gender egalitarian practice measures, GII and GLOBE GEP scores, had a non-significant negative relationship, r = −.31, ns, and the gender egalitarian value measures, GEV and WVS patriarchal values, had a negative relationship, r = −.77, p < .01.
|1. Supervisor's rating||4.06||0.67|
|3. Ratee gender||0.66||0.47||0.00||−0.01|
|4. Ratee age||42.39||7.35||0.07**||0.03**||0.03**|
|5. Ratee education||3.38||0.97||0.02**||0.02**||0.00||0.02**|
|6. Ratee tenure||10.44||7.95||0.07**||0.03**||0.04**||0.40**||−0.12**|
|7. Ratee expatriate status||0.13||0.33||−0.09**||−0.07**||0.05**||−0.07**||0.08**||−0.07**|
|8. Supervisor's gender||0.79||0.40||−0.08**||–||0.21**||0.01||0.00||0.03**||0.05**|
|9. Supervisor's familiarity with ratee||3.38||0.56||0.07**||–||−0.01||0.01||0.00||0.05**||−0.01||0.02**|
|10. Employment sector||0.22||0.41||0.15**||0.09**||−0.11**||0.18**||0.09**||0.20**||−0.09**||−0.11**||−0.06**|
|11. Year of assessment||2004.19||1.65||0.06**||0.04**||0.01||0.02**||−0.01**||−0.02**||0.03**||−0.01*||0.03**||0.03**|
|1. Gender Inequality Index (GII) 1995||.31||.18|
|2. Project GLOBE Gender Egalitarian Practices (GEP)||3.38||.34||−.31|
|3. Project GLOBE Gender Egalitarian Values (GEV)||4.60||.47||−.54**||.47**|
|4. World Values Survey (WVS) Patriarchal Values||.00||.91||.69**||−.37*||−.77**|
Gender Egalitarianism and Gender Differences in Self-Reported Work–Life Balance
Our analyses of managers' self-rated work–life balance are shown in Table 3. Self-ratings of work–life balance did not differ by gender, γ10 = .02, ns. (Model 1). Also, the random effect for ratee gender was not statistically significant, μ1 = .00091, ns, indicating that there was very little systematic cross-national variation in gender slopes, which should be kept in mind as we report our tests of Hypothesis 1. Education had a positive relationship and expatriate status had a negative relationship to self-reported work–life balance. Public sector managers rated their work–life balance higher than managers in other employment sectors, and assessment year was positively related to balance.
|Variables||Model 1||Model 2||Model 3||Model 4||Model 5|
|Intercept (γ00)||3.63*** (.02)||3.62*** (.02)||3.63*** (.02)||3.62*** (.02)||3.62*** (.02)|
|Level 1: Individual-level Fixed Effects|
|Ratee gender (γ10)||.02 (.01)||.04** (.01)||.03 (.01)||.04 (.02)||.05* (.02)|
|Ratee age (γ20)||.00 (.00)||.00 (.00)||.00 (.00)||.00 (.00)||.00 (.00)|
|Ratee education (γ30)||.02*** (.00)||.02*** (.00)||.02*** (.00)||.02*** (.00)||.02*** (.00)|
|Ratee tenure (γ40)||.00 (.00)||.00 (.00)||.00 (.00)||.00 (.00)||.00 (.00)|
|Ratee expatriate (γ50)||−.11*** (.01)||−.11*** (.01)||−.11*** (.01)||−.11*** (.01)||−.11*** (.01)|
|Employment sector (γ60)||.14*** (.03)||.13*** (.02)||.13*** (.02)||.13*** (.02)||.13*** (.03)|
|Year of assessment (γ70)||.02* (.01)||.02† (.01)||.02* (.01)||.02* (.01)||.02 (.01)|
|Level 2: Country-level Fixed Effects|
|Int. × Gender Inequality Index (GII) (γ01)||−.05* (.02)|
|Int. × Gender egal. practices (GEP) (γ01)||.05 (.02)|
|Int. × Gender egal. values (GEV) (γ01)||.05 (.03)|
|Int. × WVS patriarchal values (WVS) (γ01)||−.08** (.02)|
|Ratee gender slope × GII (γ11)||.08* (.03)|
|Ratee gender slope × GEP (γ11)||−.04 (.02)|
|Ratee gender slope × GEV (γ11)||−.04 (.02)|
|Ratee gender slope × WVS (γ11)||.05* (.02)|
|Intercept random effect (μ0)||.01217***||.01141***||.01013***||.01035***||.00876***|
|Ratee gender slope random effect (μ1)||.00091||.00069||.00103||.00053||.00027|
The tests of Hypothesis 1 predicting that gender egalitarian practices and values would moderate the relationship of ratees' gender to their self-reported work–life balance are shown in Table 3. We found partial support based on gender egalitarian practice measures, as GII was a significant cross-level moderator of the relationship of ratee gender to work–life balance, γ11 = .08, p < .05 (Model 2), but GLOBE GEP did not have a significant relationship (Model 3). We computed the effect size for the ratee gender by GII interaction using Raudenbush and Bryk's (2002) formula, and found that GII explained 24 per cent of the small cross-country variation in gender slopes (i.e. the random variation in country gender slopes).
In order to facilitate interpretation, we graphed the GII by ratee gender interaction, for low (1 SD below the mean) and high (1 SD above the mean) GII cultures, for female and male ratees, using procedures for graphing cross-level interactions (Preacher, Curran, & Bauer, 2006). The interaction followed predictions (Figure 1), with non-significant gender differences in self-rated balance in low GII (i.e. more egalitarian) cultures, t = −1.39, ns, and women reporting significantly lower work–life balance than men in high GII (i.e. less egalitarian) cultures, t = 3.17, p < .01. Also, women's self-reported work–life balance was negatively related to GII, simple slope t = −2.15, p < .05, but men's ratings were not significantly related to GII.
We also found partial support for Hypothesis 1 based on gender egalitarian values measures, as GLOBE GEV was not a significant moderator of the gender relationship to work–life balance (Model 4), but WVS patriarchal values did significantly moderate the relationship, γ11 = .05, p < .05, based on n = 32 countries (Model 5). The graph of the significant patriarchal values interaction (Figure 2) followed the predicted pattern, with women reporting significantly lower work–life balance than men in more patriarchal (less egalitarian) societies, t = 2.51, p < .05, and non-significant gender differences in less patriarchal societies, t = .26, ns. Also, women's work–life balance ratings were negatively related to societal patriarchal values, simple slope t = −3.24, p < .01, but men's balance ratings were not significantly related to patriarchal values. Thus, we found partial support for Hypothesis 1, based on GII and WVS patriarchal values as moderators of the gender to balance relationships.
Gender Egalitarianism and Ratee Gender Differences in Supervisors' Appraisals of Work–Life Balance
Our analyses of supervisors' ratings of their subordinates' work–life balance are shown in Table 4. The tests of the individual level-1 characteristics (Model 1) showed that the overall fixed effect for ratee gender was not significantly related to supervisors' work–life balance ratings, γ10 = .02, ns. However, consistent with our expectation, the random effect for ratee gender was significant, μ1 = .01127, p < .001, indicating that there was significant cross-national variation in gender slopes that could potentially be explained by the level-2 (country) measures.
|Variables||Model 1||Model 2||Model 3||Model 4||Model 5d|
|Intercept (γ00)||3.80*** (.02)||3.80*** (.02)||3.80*** (.02)||3.80*** (.02)||3.79*** (.02)|
|Level 1: Individual-level Fixed Effects|
|Ratee gender (γ10)||.02 (.03)||.05 (.02)||.02 (.03)||.04 (.03)||.04 (.03)|
|Ratee age (γ20)||.00** (.00)||.00** (.00)||.00** (.00)||.00** (.00)||.00** (.00)|
|Ratee education (γ30)||.02*** (.00)||.02*** (.00)||.02*** (.00)||.02*** (.00)||.02*** (.00)|
|Ratee tenure (γ40)||.00*** (.00)||.00*** (.00)||.00*** (.00)||.00*** (.00)||.00*** (.00)|
|Ratee expatriate (γ50)||−.11*** (.01)||−.11*** (.01)||−.11*** (.01)||−.11*** (.01)||−.11*** (.01)|
|Rater gender (γ60)||−.10*** (.01)||−.10*** (.01)||−.10*** (.01)||−.10*** (.01)||−.10*** (.01)|
|Rater familiarity with ratee (γ70)||.08*** (.01)||.08*** (.01)||.08*** (.01)||.08*** (.01)||.08*** (.01)|
|Employment sector (γ80)||.27*** (.05)||.28*** (.05)||.26*** (.05)||.26*** (.05)||.27*** (.05)|
|Year of assessment (γ90)||.02*** (.00)||.02*** (.00)||.02*** (.00)||.02*** (.00)||.02** (.01)|
|Level 2: Country-level Fixed Effects|
|Int. × Gender Inequality Index (GII) (γ01)||−.03 (.03)|
|Int. × Gender egal. practices (GEP) (γ01)||.07** (.02)|
|Int. × Gender egal. values (GEV) (γ01)||.07* (.03)|
|Int. × WVS patriarchal values (WVS) (γ01)||−.10* (.03)|
|Ratee gender slope × GII (γ11)||.10** (.03)|
|Ratee gender slope × GEP (γ11)||−.05 (.03)|
|Ratee gender slope × GEV (γ11)||−.08** (.03)|
|Ratee gender slope × WVS (γ11)||.06* (.03)|
|Intercept random effect (μ0)||.01733***||.01757***||.01230***||.01311***||.01340***|
|Ratee gender slope random effect (μ1)||.01127***||.00743***||.01129***||.00837***||.01090***|
All of the level-1 control variables had significant relationships to supervisors' work–life balance ratings; supervisors gave higher balance ratings to subordinates who were older, had more education or longer tenure, and gave lower ratings to expatriates than non-expatriates. The negative relationship for rater gender indicates that female supervisors gave higher balance ratings than male supervisors; also, rater familiarity was positively associated with balance ratings. However, the interaction of rater gender by ratee gender was not statistically significant (not shown), so we did not use this interaction as an additional control variable. Public sector employment and assessment year were positively related to supervisors' balance ratings.
The tests of Hypothesis 2 predicting that gender egalitarian practices and values would moderate the relationship of ratee gender to supervisors' work–life balance appraisals are shown in Table 4. Of the two gender egalitarian practice measures, only GII significantly moderated the relationship, γ11 = .10, p < .01, and explained 34 per cent of the cross-national variation (random effect) for ratee gender (Model 2), but GLOBE GEP was not significant as a moderator (Model 3).
Our graph of the GII by ratee gender interaction showed that the direction was consistent with our prediction, as supervisors gave similar work–life balance ratings to women and men in low GII (i.e. more egalitarian) countries, t = −1.65, ns, and significantly lower work–life balance ratings to female than male subordinates in high GII (i.e. less egalitarian) countries, t = 3.97, p < .001.
Our tests of Hypothesis 2 with gender egalitarian value measures indicated that both GLOBE GEV, γ11 = −.08, p < .01 (Model 4) and WVS patriarchal values, γ11 = .06, p < .05, based on n = 32 countries (Model 5), significantly moderated relationships of ratee gender to supervisors' ratings of work–life balance. Based on reduction in the random effects, the GLOBE GEV measure explained 26 per cent of the cross-national variation in gender slopes, and the WVS patriarchal values measure explained 5 per cent of the cross-national variation in gender slopes among the 32 countries used in that analysis.
The graph of the GEV by ratee gender interaction (Figure 3) followed the predicted direction as supervisors rated women significantly lower than men in low egalitarian societies, t = 2.82, p < .01, with no gender difference in high egalitarian societies, t = 1.36, ns. Also, supervisors' ratings of women were positively related to GEV, simple slope t = 2.55, p < .05, but ratings of men did not differ depending on GEV. The graph of the WVS patriarchal values by ratee gender interaction (Figure 2) was also in the predicted direction, as women were rated significantly lower than men in more patriarchal societies, t = 2.17, p < .05, but similarly to men in less patriarchal societies, t = −.51, ns. Also, ratings of women were negatively related to societal patriarchal values, simple slope t = −2.78, p < .01, but ratings of men were unrelated to societal patriarchal values. Thus, we found strong support for Hypothesis 2 based on both measures of egalitarian values as moderators.
As discussed earlier, there was a small to moderate amount of cross-country variation in supervisors' ratings of women, ICC(1) = . 073. Because all of the significant relationships were based on ratings of women (but not men), we conducted separate analyses with supervisors' appraisals of women. We found that the two measures of gender egalitarian values explained the most cross-national variation in supervisors' ratings of women, with 65 per cent and 49 per cent explained by GLOBE GEV and WVS patriarchal values, respectively, as compared to 29 per cent of the cross-national variation that was explained by the GII egalitarian practice measure.
In addition to the importance of work–life balance for personal well-being (MacDermid, 2005), extant research has shown that supervisors' perceptions of their subordinates' work–life balance were related to appraisals of performance and promotability (Carlson et al., 2008; Lyness & Judiesch, 2008). Yet, prior to the present study we knew little about how these perceptions are formed and whether they differ based on subordinates' gender in combination with country context. Accordingly, in the present study we examined these questions with both supervisors' appraisals and managerial ratees' self-reported work–life balance, using a sample of over 40,000 managers working in 36 culturally diverse countries.
Our theoretical framework included ideas from cross-cultural extensions of social role theory (Eagly & Wood, 2012; Wood & Eagly, 2002, 2010) about how societal normative gender roles influence members' behavior (as reflected in self-perceptions) and also shape corresponding societal gender stereotypes—which we suggested would influence perceptions by others (as reflected in supervisors' appraisals). We combined these ideas about the influence of societal gender role beliefs with literature related to gender egalitarianism, a key dimension of societal culture that helps to clarify the nature (i.e. content) of societal gender role beliefs and how they differ across cultures. Based on these theoretical ideas combined with examples of the limited relevant work–life research, we predicted that country gender egalitarianism would moderate relationships of ratee gender to work–life balance based on both supervisors' perceptions of their subordinates' work–life balance and the subordinates' self-reported work–life balance. Specifically, we predicted that gender differences in balance would be larger in low egalitarian cultures than in high egalitarian cultures.
Gender Egalitarianism and Gender Differences in Self-Reported Work–Life Balance
We found little evidence of gender differences in managers' self-reported work–life balance, as there was not an overall gender difference, and the random effect representing cross-national variation in gender differences was small and non-significant. Nevertheless two country-level measures of gender egalitarianism, i.e. GII, an index of objectively measured gender inequalities (United Nations Development Programme, 2010), and a WVS-based measure of patriarchal values (World Values Survey 1981–2008 Official Aggregate v.20090901, 2009), significantly moderated the small cross-national variation in gender differences in balance. Also, for both measures the moderation followed the predicted direction, with women reporting lower work–life balance than men in low egalitarian countries, and a non-significant gender difference in high egalitarian countries. Another important finding was that women's reported work–life balance varied depending on these two measures of country context, but men's reported work–life balance did not vary by context. However, as will be discussed later, the Project GLOBE GEV and GEP measures were not significant moderators of the relationship between gender and self-reported work-life balance.
Gender Egalitarianism and Ratee Gender Differences in Supervisors' Appraisals of Work–Life Balance
In contrast to the self-reported balance results, we found significant cross-national variation by ratee gender in supervisors' appraisals of their subordinates' work–life balance, with a small to medium effect size. The larger systematic between-country variation in ratee gender differences for supervisors' appraisals as compared to self-appraisals is consistent with prior US literature suggesting that those in power rely more on demographic stereotypes, such as gender stereotypes, than individuating information (Fiske & Berdahl, 2007), whereas the small, non-significant between-country variation in gender differences for self-ratings suggests greater influence of individual circumstances or values, as reflected in more within-country variation.
As we noted earlier, Hoobler et al.'s (2009) US findings showed that supervisors rated their female subordinates as having more family–work conflict (a component of work–life balance) than their male subordinates, which is consistent with our results showing systematic between-country variation by ratee gender in supervisors' balance appraisals. Also, our findings that gender differences in self-ratings showed more within-country variation (and less between-country variation), suggesting more reliance on individuating information, might help to explain Hoobler et al.'s non-significant gender differences in self-ratings of family–work conflict. However, by examining supervisor and self-ratings from 36 culturally diverse countries, the present study extends our understanding of these issues beyond the US, and indicates that gender egalitarianism, an important aspect of country context, moderates the relationship of ratee gender to supervisors' work–life balance ratings. Thus, our findings suggest that whether or not supervisors' work–life balance ratings would be expected to differ by ratee gender in a particular country depends to some extent on the level of country gender egalitarianism.
An additional factor that may be relevant to supervisors' appraisals is that women may hold a smaller proportion of managerial positions in low egalitarian countries than in high egalitarian countries. According to tokenism theory, when there are relatively few women in a work group, the token women are more likely to be viewed stereotypically as women rather than as managers (Kanter, 1977; Yoder, 1991, 1994). Moreover, related research has shown that women received lower performance evaluations when they worked in male-dominated work groups or held jobs that were typically held by men, as compared to evaluations of women in more gender-balanced groups or jobs (Lyness & Heilman, 2006; Pazy & Oron, 2001).
Our use of multiple measures of gender egalitarianism helps to clarify and strengthen the findings. Notably, three out of four measures of gender egalitarianism significantly moderated the cross-national variation in ratee gender differences found in supervisors' appraisals. We found significant moderation with both measures of country egalitarian values, i.e. the Project GLOBE GEV scores and a WVS-based measure of patriarchal values, as well as the GII measure of objective gender inequalities. The egalitarian measures had meaningful effect sizes, as 34 per cent, 26 per cent, and 5 per cent of the cross-national variation in gender differences were explained by GII, GEV, and WVS patriarchal values, respectively. This pattern of results suggested that supervisors' appraisals may have been influenced by gender stereotypes that varied depending on country context, as reflected in the country level of gender egalitarianism.
As we were unable to find prior cross-national work–life studies testing direct measures of gender differences in work–life balance in relation to country gender egalitarianism, we had to rely on research that examined gender differences in other work–life constructs, such as work–family conflict, considered to be components of work–family balance (e.g. Frone, 2003), in countries thought to differ in gender egalitarianism. Thus, a key empirical contribution of the present study is its direct tests of cross-national relationships of ratee gender differences in multisource ratings of work–life balance and country gender egalitarianism.
It is also noteworthy that supervisors' appraisals of women were significantly related to country context, based on three measures of gender egalitarianism, but their appraisals of men did not vary based on egalitarian context. Our findings that the Project GLOBE GEP measure was not a significant moderator of either self-rated or supervisors' appraisals may have been due to its low variance, which was the explanation Project GLOBE researchers offered for non-significant relationships of GEP with measures of similar constructs (Emrich et al., 2004).
An important goal of the present study was to provide one of the first empirical tests of predictions based in part on cross-cultural extensions of social role theory by Eagly and her colleagues (Eagly & Wood, 2012; Wood & Eagly, 2002, 2010). As we mentioned earlier, social role theory offers explanations about the processes by which collective observations of women's and men's behavior shape societal beliefs about both normative gender roles and corresponding gender stereotypes. Thus, we predicted that both self-ratings and supervisors' perceptions of female and male subordinates' work–life balance would reflect these shared societal beliefs about gender roles. We combined these ideas with prior cross-national literature to formulate predictions that country gender egalitarianism would moderate and help to explain cross-national gender patterns in multisource appraisals of work–life balance.
An important finding is that two measures of gender egalitarianism, i.e. GII and WVS patriarchal values, moderated gender differences in both self-reported and supervisors' ratings of work–life balance across 36 countries that varied in levels of gender egalitarianism. Moreover, with both measures we found very similar patterns of moderation that corresponded to the predicted directions. There were significant gender differences in low egalitarian countries where gender roles are more distinct, based on the traditional male breadwinner–female caregiver division of labor, and non-significant gender differences in high egalitarian countries, where gender roles are more similar. Because our findings were based on two independent rating sources and perspectives (i.e. self versus other), they provide empirical support for our extensions of social role theory and the work–life literature to predict that not only women's and men's self-reported actual work–life balance but also supervisors' perceptions of their female and male subordinates' work–life balance would reflect common grounding in societal gender role beliefs and corresponding gender stereotypes.
Moreover, our empirical findings also extended the social role theory-based predictions by clarifying the relative influence of societal gender role beliefs and individuating information across the two perspectives. Specifically, a key finding was that appraisals by supervisors showed more systematic cross-national variation by ratee gender, suggesting that supervisors' appraisals might have reflected country gender stereotypes. Also, our control for supervisors' familiarity with ratees helped to rule out lack of familiarity as an alternative explanation for these findings. In contrast, the managers' self-appraisals showed less cross-national variation by gender, suggesting less conformity or influence of societal gender roles, and more within-country variation reflecting individuating factors.
These results suggest that social role theory might be refined to more explicitly acknowledge important individual differences, such as personal values (Greenhaus & Allen, 2011) and circumstances, that may weaken women's and men's conformity to their respective societal gender roles and norms, as reflected in our self-reported balance appraisals. These findings are also relevant to the nascent cross-national work–life literature as they underscore the need to investigate not only gender differences, but also the effects of both societal context and individual circumstances in order to understand work–life balance and related constructs.
In addition, the findings about supervisors' appraisals could be used to extend social role theory which focuses primarily on the influence of societal gender role beliefs on individual behavior. Our results show, however, that through their influence on societal gender stereotypes, societal gender roles and norms may exert even more influence on perceptions of women and men by others, i.e. supervisors in our research, suggesting that this aspect of social role theory warrants further development and testing.
Our comparison of four country-level measures assessing different aspects of gender egalitarianism also contributes to cross-cultural work–life theory by suggesting that different aspects of gender egalitarianism are more relevant for understanding others' (i.e. supervisors') perceptions as compared to an individual's actual experiences, reflected in self-rated balance. Cross-national gender variations in supervisors' balance appraisals, especially their appraisals of women, were mostly explained by country gender egalitarian values, which were probably also reflected in societal gender stereotypes. In contrast, self-reported work–life balance was significantly related to actual gender inequalities in country conditions, such as women's access to education and work opportunities, as measured by GII, and was also related to WVS patriarchal values, shown in prior research to predict similar types of gender inequalities (Fortin, 2005; McDaniel, 2008; Seguino, 2007; Wernet et al., 2005). It is not surprising that women's self-reported work–life balance, and perhaps their ability to achieve balance, was linked to societal gender inequalities. For instance, consistent with our data showing that India ranked highest in GII, a 21-country Nielsen study found that the highest levels of stress were reported by women in India; Indian women's high stress was said to reflect expectations that they cope with demanding jobs, due to India's rapid economic development, combined with unchanged traditional familial obligations that exceed those for Western women (Goyal, 2011).
As explained above, our predictions about greater gender differences in low as compared to high gender egalitarian countries were based in part on the limited prior cross-national work–life research measuring gender differences in balance-related work–life constructs in countries thought to differ in gender egalitarianism. This literature suggested that in low egalitarian cultures men had higher levels of work–life balance due to the congruence between their traditional work and family roles as breadwinners than women whose traditional caregiver role is less congruent with involvement in paid work, but in high egalitarian cultures there was less differentiation in gender roles and similar levels of work–life balance for both women and men due to more equitable division of labor (Aryee et al., 2005; Hill et al., 2003). The present results are consistent with this prior literature but extend our knowledge by providing direct tests of these ideas. Our findings show that country gender egalitarianism significantly moderated cross-national variation by ratee gender in multisource work–life balance appraisals, based on three egalitarianism measures for supervisor appraisals and two measures for self-appraisals of balance. Notably, all five of these significant interactions followed the predicted pattern.
Moreover, our findings also offer possible insights into how gender egalitarianism moderates gender differences in work–life balance. Specifically, through fine-grained analyses, we found that all of the significant interactions were driven by variations in appraisals of women's work–life balance depending on cultural context, as measured by gender egalitarianism, but appraisals of men did not vary according to cultural context. These findings should be kept in mind when future cross-national research is conducted to see if similar patterns are present.
Prior research has suggested that supervisors form impressions about how well their subordinates balance work with non-work, and that supervisors' appraisals of work–life balance and similar constructs are related to appraisals of the subordinates' performance and promotability (Carlson et al., 2008; Hoobler et al., 2009; Lyness & Judiesch, 2008). Thus, as the first cross-national examination of how subordinate gender and country context relate to both supervisors' appraisals and managerial subordinates' self-reported work–life balance, our findings have important practical implications for managers and their organisations.
In particular, our results showing that supervisors' appraisals differed, depending on subordinates' gender in combination with country context, and were also linked to country gender egalitarian values, suggest that supervisors' appraisals were influenced by local gender norms and stereotypes. It is also notable that in low egalitarian countries supervisors had more negative perceptions of female subordinates' work–life balance in comparison to male subordinates, and analyses limited to appraisals of female subordinates showed that supervisors also had more negative perceptions of women's work–life balance in low egalitarian cultures relative to their female counterparts in high egalitarian cultures.
A possible interpretation, with important implications for multinational employers, is that supervisors' judgments may have varied depending on subordinates' gender and country context. According to scarcity and depletion theories, an individual is thought to have a limited amount of resources, such as time and energy, such that involvement in one role depletes these resources, resulting in lower performance in other roles (Goode, 1960; Rothbard, 2001). Thus, consistent with this theory and the distinct gender roles in low egalitarian cultures, supervisors might have assessed working women in low egalitarian countries as less balanced due to their perceived inability to fulfill the greater traditional expectations about women's family involvement—such that these women were judged against stricter standards than either their male counterparts or women in more egalitarian countries.
Also, our findings revealed that supervisors' ratings of women but not of men varied depending on country context, and this raises important questions about whether supervisors' appraisals of women were reflecting the influence of local gender stereotypes or whether male gender stereotypes are generally more positive because of the greater perceived congruence between work involvement and the male breadwinner role, as was discussed earlier. Nevertheless our findings that supervisors' appraisals of women but not men varied depending on national context suggest that multinational employers need to carefully analyze such appraisals before assuming that they are comparable across genders and culturally diverse countries.
Limitations and Suggestions for Future Research
Our use of cross-sectional, archival data allowed us to directly test our hypotheses with multisource work–life balance appraisals for managers from 36 culturally diverse countries. However, an important limitation is the lack of information about ratees' non-work lives, including marital status, children, elder care, and other non-work responsibilities, that might help to explain the within-country variation that we found in self-ratings. Thus, it is important to conduct future research incorporating these types of measures to examine their relationships to work–life balance measured with self-ratings and supervisors' perceptions. In addition, new insights might be gained from adding other perspectives, such as spouses and significant others.
Our specific results also raise some questions for future research. For example, further exploration is warranted to clarify why we found that appraisals of women but not men vary in relation to contextual factors. Also, it would be helpful to learn more about how supervisors judge their subordinates' work–life balance, and whether criteria differ depending on subordinates' gender or country. In addition, future research should examine other aspects of national context to test for additional moderators that might help to explain cross-national variations in work–life balance and related constructs. For example, work–life balance has been linked to a focus on quality of life (Guest, 2002), which is associated with GLOBE cultural dimensions, such as humane orientation and low assertiveness (House et al., 2004).
Work–life balance has important implications for both personal well-being and work-related outcomes. However, up until now, little has been known about supervisors' perceptions of their subordinates' work–life balance, or whether supervisors' perceptions varied by subordinate gender in combination with country context. The present study addressed these questions by combining ideas from Eagly and Wood's (2012) cross-cultural extension of social role theory, and the limited prior cross-national literature about gender differences in work–life constructs. We extended the prior theoretical ideas to develop and test predictions about country gender egalitarianism as a moderator of cross-national variations in ratee gender differences in managers' self-reported work–life balance as well as supervisors' appraisals of these managers' work–life balance. Based on multilevel analyses of appraisals of over 40,000 managers working in 36 countries with varying degrees of gender egalitarianism, we found more cross-national variation by ratee gender in supervisors' appraisals than self-reported work–life balance. Women were rated lower in work–life balance than men in low egalitarian cultures, but gender differences were non-significant in high egalitarian cultures. Also, appraisals of women's but not men's work–life balance varied depending on egalitarian context.
Our findings provide important clarification about the relative importance of societal gender roles depending on the perspective, as self-ratings appeared to reflect more individuating information and less conformity to societal gender roles, relative to supervisors' appraisals, which appeared to reflect more influence of societal gender stereotypes, as reflected in cross-national variation by subordinate gender and country gender egalitarian values. The findings also underscore the value of using multisource data and cross-national research to better understand the effects of contextual characteristics.
- 2005). Rhythms of life: Antecedents and outcomes of work–family balance in employed parents. Journal of Applied Psychology, 90, 132–146. , , & (
- 2008). Cross-cultural approaches to work–farnily conflict. In K. Korabik , D.S. Lero , & D.L. Whitehead (Eds.), Handbook of work–family integration: Research, theory, and best practices (pp. 353–370). San Diego, CA: Academic Press. (
- 2008). Supervisor appraisal as the link between family–work balance and contextual performance. Journal of Business & Psychology, 23, 37–49. , , , , & (
- Center for Creative Leadership (2004). Benchmarks Facilitators Manual, Version 3.0. Greensboro, NC: Center for Creative Leadership.
- Center for Creative Leadership (2010). Benchmarks Technical Manual, Version 1.0. Greensboro, NC: Center for Creative Leadership.
- 2006). Work–life “balance” in Europe. Acta Sociologica, 49, 379–393. , & (
- 1985). Some neglected variables in research on discrimination in appraisals. Academy of Management Review, 10, 116–127. (
- 1987). Sex differences in social behavior: A social-role interpretation. Hillsdale, NJ: Erlbaum. (
- 2012). Social role theory. In P.A.M. Van Lange , A.W. Kruglanski , & E.T. Higgins (Eds.), Handbook of theories of social psychology (Vol. 2, pp. 458–476). Thousand Oaks, CA: Sage. , & (
- 2005). Work and family research in IO/OB: Content analysis and review of the literature (1980–2002). Journal of Vocational Behavior, 66, 124–197. , , , , & (
- 2004). Cross-cultural differences in gender egalitarianism: Implications for societies, organizations, and leaders. In R.J. House , P.J. Hanges , M. Javidan , P.W. Dorfman , & V. Gupta (Eds.), Culture, leadership, and organizations: The GLOBE study of 62 societies (pp. 343–394). Thousand Oaks, CA: Sage. , , & (
- 2007). Social power. In E.T. Higgins & A.W. Kruglanski (Eds.), Social psychology: Handbook of basic principles (pp. 678–692). Oxford: Oxford University Press. , & (
- 2005). Gender role attitudes and the labour-market outcomes of women across OECD countries. Oxford Review of Economic Policy, 21, 416–438. (
- 2003). Work–family balance. In J.C. Quick & L.E. Tetrick (Eds.), Handbook of occupational health psychology (pp. 143–162). Washington, DC: American Psychological Association. (
- 1960). A theory of role strain. American Sociological Review, 25, 483–496. (
- 2011). Indian women most stressed in the world: Nielsen survey [ET Cetera], from http://search.proquest.com/docview/873955235?accountid=8500 (
- 2011). Work–family balance: A review and extension of the literature. In J.C. Quick & L.E. Tetrick (Eds.), Handbook of occupational health psychology (2nd edn., pp. 165–183). Washington, DC: American Psychological Association. , & (
- 2002). Perspectives on the study of work–life balance. Social Science Information, 41, 255–279. (
- 2004). Regional and climate clustering of societal cultures. In R.J. House , P.J. Hanges , M. Javidan , P.W. Dorfman , & V. Gupta (Eds.), Culture, leadership, and organizations: The GLOBE study of 62 societies (pp. 178–218). Thousand Oaks, CA: Sage. , & (
- 2001). Description and prescription: How gender stereotypes prevent women's ascent up the organizational ladder. Journal of Social Issues, 57, 657–674. (
- 2001). Finding an extra day a week: The positive influence of perceived job flexibility on work and family life balance. Family Relations: Interdisciplinary Journal of Applied Family Studies, 50, 49–58. , , , & (
- 2003). Studying “working fathers”: Comparing fathers' and mothers' work–family conflict, fit, and adaptive strategies in a global high-tech company. Fathering, 1, 239–261. , , , & (
- 1998). Centering decisions in hierarchical linear models: Implications for research in organizations. Journal of Management, 24, 623–641. , & (
- 1980). Culture's consequences: International differences in work-related values (Vol. 5). Newbury Park, CA: Sage. (
- 2009). Bosses' perceptions of family–work conflict and women's promotability: Glass ceiling effects. Academy of Management Journal, 52, 939–957. , , & (
- House, R.J. , Hanges, P.J. , Javidan, M. , Dorfman, P.W. , & Gupta, V. (Eds.) (2004). Culture, leadership, and organizations: The GLOBE study of 62 societies. Thousand Oaks, CA: Sage.
- 2004). Overview of GLOBE. In R.J. House , P.J. Hanges , M. Javidan , P.W. Dorfman , & V. Gupta (Eds.), Culture, leadership, and organizations: The GLOBE study of 62 societies. Thousand Oaks, CA: Sage. , & (
- 2004). A nontechnical summary of GLOBE findings. In R.J. House , P.J. Hanges , M. Javidan , P.W. Dorfman , & V. Gupta (Eds.), Culture, leadership, and organizations: The GLOBE study of 62 societies (pp. 29–50). Thousand Oaks, CA: Sage. , , & (
- 1977). Some effects of proportions on group life: Skewed sex ratios and responses to token women. American Journal of Sociology, 82, 965–990. (
- 2008). Answers to 20 questions about interrater reliability and interrater agreement. Organizational Research Methods, 11(4), 815–852. , & (
- 2008). Assumptions, research gaps and emerging issues: Implications for research, policy and practice. In K. Korabik , D.S. Lero , & D.L. Whitehead (Eds.), Handbook of work–family integration: Research, theory, and best practices (pp. 371–397). San Diego, CA: Academic Press. , & (
- 2008). What's in a name? “Work and family” or “work and life” balance policies in the UK since 1997 and the implications for the pursuit of gender equality. Social Policy & Administration, 42, 524–541. , & (
- 2006). When fit is fundamental: Performance evaluations and promotions of upper-level female and male managers. Journal of Applied Psychology, 91, 777–785. , & (
- 2008). Can a manager have a life and a career? International and multisource perspectives on work–life balance and career advancement potential. Journal of Applied Psychology, 93, 789–805. , & (
- 2005). The relationships of national gender equality and organizational support with work–family balance: A study of European managers. Human Relations, 58, 33–60. , & (
- 2008). Measuring gender egalitarianism: The attitudinal difference between men and women. International Journal of Sociology, 38, 58–80. (
- 2005). (Re)Considering conflict between work and family. In E.E. Kossek & S.J. Lambert (Eds.), Work and life integration: Organizational, cultural, and individual perspectives (pp. 19–40). Mahwah, NJ: Lawrence Erlbaum. (
- 1996). Multiple roles and the self: A theory of role balance. Journal of Marriage and the Family, 58, 417–432. , & (
- Mayo Foundation for Medical Education and Research (2012, 12 July). Work–life balance: Tips to reclaim control. Retrieved 21 September 2012 from: http://www.mayoclinic.com/health/work-life-balance/WL00056
- 2001). Sex proportion and performance evaluation among high-ranking military officers. Journal of Organizational Behavior, 22, 689–702. , & (
- 2006). Computational tools for probing interaction effects in multiple linear regression, multilevel modeling, and latent curve analysis. Journal of Educational and Behavioral Statistics, 31, 437–448. , , & (
- 2002). Hierarchical linear models: Applications and data analysis methods (2nd edn.). Thousand Oaks, CA: Sage. , & (
- 2004). HLM 6: Hierarchical linear and nonlinear modeling. Lincolnwood, IL: Scientific Software International, Inc. , , , , & (
- 1985). Clustering countries on attitudinal dimensions: A review and synthesis. Academy of Management Review, 10, 435–454. , & (
- 2001). Enriching or depleting? The dynamics of engagement in work and family roles. Administrative Science Quarterly, 46, 655–684. (
- 2007). Plus ça change? Evidence on global trends in gender norms and stereotypes. Feminist Economics, 13, 1–28. (
- 2003). Review of the Benchmarks (Revised). In B.S. Plake , J.C. Impara , & R.A. Spies (Eds.), The fifteenth mental measurements yearbook (pp. 124–126). Lincoln, NE: Buros Institute of Mental Measurements. (
- 2009). Half a century of measuring culture: Review of approaches, challenges, and limitations based on the analysis of 121 instruments for quantifying culture. Journal of International Management, 15, 357–373. , , & (
- 2001). Unbinding time: Alternate work schedules and work–life balance. Journal of Family and Economic Issues, 22, 101–119. , & (
- 2012). Apron strings of working mothers: Maternal employment and housework in cross-national perspective. Social Science Research, 41, 833–842. , & (
- 2000). Marrried women's employment over the life course: Attitudes in cross-national perspective. Social Forces, 78, 1409–1436. , & (
- United Nations Development Programme (2010). Human Development Report 2010: The Real Wealth of Nations: Pathways to Human Development. New York: United Nations Development Programme.
- 2006). Combination pressure: The paid work–family balance of women and men in European countries. Acta Sociologica, 49, 303–319. , , & (
- 2010). Theoretical perspectives of balance and the influence of gender ideologies. Journal of Occupational Science, 17, 92–103. , , & (
- 2005). The postmodern individual: Structural determinants of attitudes. Comparative Sociology, 4, 339–364. , , & (
- 2002). A cross-cultural analysis of the behavior of women and men: Implications for the origins of sex differences. Psychological Bulletin, 128, 699–727. , & (
- 2010). Gender. In S.T. Fiske , D.T. Gilbert , & G. Lindzey (Eds.), Handbook of social psychology (Vol. 1, 5th edn., pp. 629–667). Hoboken, NJ: John Wiley & Sons. , & (
- World Values Survey 1981–2008 Official Aggregate v.20090901 (2009). World Values Survey Association (www.worldvaluessurvey.org) Aggregate File Producer: ASEP/JDS. Retrieved 1 September 2012 from: http://www.wvsevsdb.com/wvs/WVSData.jsp
- 1991). Rethinking tokenism: Looking beyond numbers. Gender and Society, 5, 178–192. (
- 1994). Looking beyond numbers: The effects of gender status, job prestige, and occupational gender-typing on tokenism processes. Social Psychology Quarterly, 57, 150–159. (
- 1995). Benchmarks. In J.C. Conoley , J.C. Impara , & L.L. Murphey (Eds.), The twelfth mental measurements yearbook. Lincoln, NE: Buros Institute of Mental Measurements. (
Country and Gender Distribution of Manager Rateesa
|Countries within Cultural Regions||Women||Men||Total|
|People's Republic of China||21||71||92|
|Total Sample of Ratees||13,988||26,933||40,921|