Abstract
This paper integrates the rapidly growing literatures on the individual and organizational factors that contribute to women’s career equality. We organize studies into three research perspectives: career preference, gender bias, and work-family explanations. These literatures diverge on whether women “opt out” or are “pushed out” of leadership positions in organizations. Further, the interconnectedness of these “pushes” and “pulls” and micro-macro linkages are not well-integrated. This creates a lack of clarity about what scholars should study and what practices organizations should implement. We define women’s career equality as an individual and organizational phenomenon involving the degree to which women (a) have equal access to and participation in career opportunities, and (b) experience equal intrinsic and extrinsic work and nonwork outcomes compared to men. We bridge the interdisciplinary divides by developing an integrative multi-level model of women’s career equality. We propose that individuals’ career perceptions and experiences are embedded in social contexts reflecting the climate for gender inclusion and interact with these contexts to shape women’s career equality outcomes. The climate for gender inclusion has three dimensions: fairness, leveraging talent, and workplace support. We identify coalescing themes to stimulate future research, including attention to national socio-economic influences, improving metrics and measurement of gender inclusion climate, multi-level career equality outcomes, a joint focus on implicit and explicit bias, and designing cross-disciplinary interventions for experiments. In order to foster theory-based research that is linked to practice, we suggest implementing and scientifically evaluating comprehensive workplace interventions that integrate perspectives and levels.
Keywords
Women rarely make one big decision to leave the workforce. Instead, they make a lot of small decisions along the way. (Sandberg, 2013: 93)
Evidence is accumulating that women remain vastly underrepresented in leadership and major institutions from business to politics in every country (World Economic Forum, 2015). Women compose half the population, yet they constitute less than 5% of CEOs and 19% of corporate board members (Catalyst, 2015). Women with similar backgrounds compared to men are more likely to turn over and less likely to advance in the prestigious, fast-growing science, technology, engineering, and mathematics (STEM) fields (Lubinski, Benbow, & Kell, 2014).
It is unclear whether the lack of progress in career equality is due to women not “leaning in” or enduring obstacles holding back advancement. Our goal is to review three under-integrated perspectives on women’s underrepresentation as leaders. They are the (a) career preference, (b) gender bias, and (c) work–family views, which we identified based on our knowledge of and reflection on decades of study on women’s careers. Some studies emphasize “pushed-out” factors, while others highlight “opting-out” pressures. A challenge in examining these “pushes” and “pulls” is their interconnectedness. What appear to be women’s individual “choices” are shaped by social context factors in which they are embedded. Instead of pitting views against each other, we take a cross-disciplinary approach to define women’s career equality, develop a multilevel framework, and offer an agenda for research and interventions. We ground this review by searching in PsycINFO using keywords such as gender bias and discrimination, career preference, work–family, gender inclusion, and career equality. Our search focused (over 1.000 appeared) on articles representing a view of fostering integration.
Women’s Career Equality: A Multilevel Phenomenon
Women’s career equality is a multilevel, multidisciplinary dynamic phenomenon that reflects the degree to which women, compared to men, (a) have equal access to and participation in career opportunities and (b) experience equal work and nonwork outcomes: intrinsic (job, life, family satisfaction) and extrinsic (pay, promotions). It can be measured at the individual and collective (societal, occupational, organizational, group) levels. Outcomes at individual and collective levels are interrelated, reflecting social contextual effects. When women are better represented across the firm, women benefit individually (Joshi, Son, & Roh, 2015), as do organizations (Krishnan & Park, 2005). Career equality entails linking concepts involving career, family, and gender bias, as these are often related (Martins, Eddleston, & Veiga, 2002).
Organizational indicators of women’s career equality involve organizational structure and culture. Two key structural indicators are (a) vertical gender integration, the equal representation of women and men in leadership across hierarchical levels, and (b) horizontal gender integration, the equal representation of women and men in occupations that are pathways to the top, such as STEM or finance. When women are underrepresented hierarchically or functionally, it is respectively known as vertical or horizontal segregation. Women are overrepresented in human resources (HR) and administrative services (Catalyst, 2015), where they are less likely to advance. This structure reproduces career inequality culturally by giving women less access to power, decision making, opportunities, and networks, relative to men (Kanter, 1977). Cultural norms supporting gender equality, or the lack thereof, shape implementation of HR practices for equality in use of work–life policies and performance, turnover, and pay metrics.
Turning to individual indicators, one type is the degree to which individual women report equal access to and participation in the same types of career opportunities as similar men. Women face a “glass cliff” in access to top leadership jobs (Ryan & Haslam, 2007) as opportunities offered are less desirable and more precarious. For instance, Mary Barra became the first woman to lead General Motors only after it emerged from bankruptcy. Equal workforce participation in high-potential jobs that shape promotions matters. Take expatriate jobs, which are key for advancement. Women reflect 40% of the global workforce but only 22% of expatriates (Welsh & Kersten, 2014). A contributor to this gap is that 90% of women professionals compared to 50% of men are in dual-career marriages. Yet few firms offer expatriate spousal job assistance, which indirectly hinders women’s ability to work abroad (Catalyst, 2000).
The second set of individual indicators relates to equality in intrinsic and extrinsic work and nonwork outcomes. A study using three decades of data showed that women with the same level of performance as men are less likely to receive similar pay and promotions, especially in prestigious, highly educated occupations (doctors and lawyers; Joshi et al., 2015). Women are more likely to view the path to advancement as more stressful (McKinsey and Company & Lean In, 2015), have lower career longevity and satisfaction (Metz, 2011), and receive less recognition (Treviño, Gomez-Mejia, Balkin, & Mixon, 2015), compared to men. Women’s career equality occurs over a career life course with interconnected work and nonwork interactions. Women’s turnover rates are higher than men’s only when reasons for leaving measure family concerns (Lee, 2012). We suggest that career studies include nonwork outcomes, such as equal work–family balance, well-being, and equality in family and career ambition (marital status, number of children) for men and women at similar levels, especially midcareer, when women lose ambition and confidence (Bain and Co., 2014).
Three Perspectives on Women’s Career Equality
Next, we review the perspectives grounding our framework. For each, we give an overview of theories, individual and organizational antecedents, cross-level dynamics, and interventions. By examining gender differences in career, bias, and work–family factors, we add to the understanding of workplace characteristics that are, on average, more likely to attract and retain women. Yet we caution readers to not use group differences to make attributions about all women within and across all societies. As an example of nested effects, we briefly comment on national socioeconomic context influences after the disciplinary reviews. Figure 1 shows linkages between context, antecedents, gender inclusion processes, and career equality outcomes.
An Integrated Multilevel Model of Women’s Career Equality: Antecedents, Gender Inclusion Processes, and Outcomes
Career Preference Perspective
The career preference view attributes gendered career paths and gender inequality largely to the interaction between women’s interests, values, and goals and characteristics of the work environment (Diekman, Brown, Johnston, & Clark, 2010). Drawing on person–environment (P-E) fit theories, this view holds that individuals choose work environments and jobs that are congruent with their interests, values, and goals in order to achieve better P-E fit, which, in turn, leads to more positive career outcomes, such as superior job performance, career advancement, and satisfaction (Nye, Su, Rounds, & Drasgow, 2012). The pursuit of congruence with work environments provides the basis for individuals’ motivations and behaviors across all career stages, such as career selection, performance and extrarole behaviors, accepting promotions, and ultimately, turnover. This perspective holds that women tend to opt out of certain jobs or occupational fields because of the perceived misfit between the work environment and their own career preferences. Since both the person and the organization matter for fit, for the career preference view, we examine (a) individual gender differences that contribute to women’s career choices and (b) characteristics of work environments that constrain their choices.
Individual level
The first individual factor relates to gender differences in career interests and goals. It has been consistently shown that, compared to men, women have stronger preference for work environments that provide more opportunities and activities to work with people and help others (Diekman et al., 2010; McCarty, Monteith, & Kaiser, 2014). For example, Su, Rounds, and Armstrong (2009) meta-analyzed data from 47 interest inventories with 503,188 respondents and reported substantial gender differences in interests. Men scored higher on realistic scales that measured interest in working with things, gadgets, or working outdoors (d = .84); women scored higher on social scales that measured interest in working with and helping people (d = –.68). Women also tend to endorse communal goals—an orientation to care about other people, favoring work environments compatible with these goals (Diekman et al., 2010). Gender differences in people-oriented interests and communal goals provide an important explanation for women’s underrepresentation in occupational fields such as STEM, which are perceived as things oriented and often incompatible with communal goals (Diekman et al., 2010). Women are more likely than men to forgo job opportunities that are seen as low in communion, even in non-STEM fields, a finding that helps explain why some women may be “leaning out” in the workplace (McCarty et al., 2014).
However, gender differences in leadership interests have been shown to be decreasing over time and were reported to be small to negligible in recent decades (Su et al., 2009). Younger cohorts of women were reported to have become more agentic (Twenge, 2001). Contrary to the belief that women have lower career ambition than men, studies have shown that women aspire to leadership roles as much as men do (Eagly, 2013), although their ambition is constrained by the lack of people-oriented opportunities and low communal affordance in such positions.
The second individual-level factor relates to gender differences in values and needs. According to preference theory, Hakim (2000) proposed that women are more likely to prioritize work–family balance, whereas men are more likely to prioritize their careers, which may lead to different labor market outcomes. Such general gender differences in career values have been also been replicated in groups with outstanding intellectual abilities. Ferriman, Lubinski, and Benbow (2009) followed a gifted sample of math/science graduate students for over 20 years since age 13 and found that women placed more importance than men on work–life balance and time with family. Major, Morganson, and Bolen (2013) showed that work–family considerations were weighted more heavily by women in their occupational and organizational commitment in the information technology industry. These gender differences intensified during parenthood and predicted differential male–female representation in prestigious time-intensive careers. Barbulescu and Bidwell (2013) also found that women’s occupational selection was influenced by preferences for better work–life balance, which had dramatic future (lower) earnings impact.
Organizational level
The career preference view suggests that job characteristics and work environments influence individuals’ perceptions of P-E fit and, in turn, their job attitudes and behaviors. Organizations with supportive cultures for women’s values, needs, and goal accomplishments place fewer constraints on women’s career choices and are more likely to be successful at attracting, retaining, and advancing women. Using an international sample, Bajdo and Dickson (2001) found that employers whose members reported stronger cultural valuing of a humane orientation reported higher percentages of women in management. In contrast, another study found that women in firms with cultures emphasizing masculine attributes have difficulty advancing (Jandeska & Kraimer, 2005). Thus, factors related to the creation of a gender-supportive organizational culture, work environment, and job design are key to enhance women’s perception of P-E fit and career equality outcomes. Yet organizations place more responsibility on women to “fit in” rather than creating “fit” for women. Even when firms adopt policies to support women, they are positioned as an “accommodation” framed within existing cultural assumptions rather than fundamental work redesign for gender equity (Bailyn, 2011).
Relatedly, studies have shown a need to more proactively redesign jobs to better fit women’s interests, goals, values, and needs. For example, Diekman, Clark, Johnston, Brown, and Steinberg (2011) report that demonstrating a STEM job’s affordance for communal goals elicits more positive attitudes toward the job, particularly for women. Another study shows that highlighting the compatibility of a promotion opportunity with communal goals increases women’s tendency to accept promotion (McCarty et al., 2014). Researchers also find that working in jobs offering more connection with people relates to increased job satisfaction for women who see themselves as people oriented (Carlson & Mellor, 2004).
A second stream of studies suggests the need to actively increase support from powerful mentors, sponsors, and role models to create a relationship-oriented organizational culture and environment, which also provides critical resources for women’s career development and advancement. Research has shown the positive effects of mentoring and supervisor support for women on many career outcomes: earnings, advancement, retention, job and career satisfaction, and job involvement (Allen, Eby, Poteet, Lentz, & Lima, 2004). Mentoring was found to be particularly effective in firms that culturally endorsed a high focus on relationships (Apospori, Nikandrou, & Panayotopoulou, 2006).
Interplay between individual and organizational levels
Because P-E fit theories by definition focus on the interaction between person and environment, cross-level dynamics are central to the career preference view. Specifically, women’s voluntary career choices are made within the context of their work environment and are constrained by the congruence between their career preferences and the characteristics of various levels of work environments (i.e., job, supervisor, team, organization; McCarty et al., 2014). These dynamics explain why women disproportionally opt out of technological work environments that are less compatible with social interests and why women forgo some career advancement opportunities that are perceived as competitive rather than collaborative.
Interventions
The career preference perspective suggests that organizational interventions should focus on changing job design, physical and social aspects of the work environment, and organizational culture to include more relational elements. Examples of these changes include creating more collaborative than competitive cultures, rewarding teamwork and helping behaviors, and promoting mentorship and sponsorship for women’s career advancement.
Gender Bias and Stereotype Perspective
The gender bias perspective (Eagly & Karau, 2002; Heilman, 2012) attributes women’s career inequality to explicit and implicit gender biases that exclude women from career opportunities. Drawing on social role theory (Eagly & Steffen, 1986), the gender bias view holds that differences in role occupancy in society, family, and occupations generate role expectations for each gender. As men traditionally occupy paid work and higher-level positions, they are expected to have agentic traits, such as being assertive, dominant, competitive, and achievement oriented. In contrast, women are expected to show communal traits, such as being helpful, kind, sympathetic, understanding, and compassionate (Diekman & Eagly, 2008). The enduring “beliefs about the characteristics, attributes, and behaviors of members” of gender groups (Hilton & von Hippel, 1996: 240), or gender stereotypes, leads to expectations about how women and men should behave (Heilman, 2012). Women, as a result, face two forms of prejudice in the workplace (Eagly & Karau, 2002). The first, descriptive prejudice, has its roots in the activation of descriptive beliefs about women’s stereotypical qualities, which is often dissimilar to the qualities expected and desired in leaders (Diekman & Eagly, 2008). This leads to less favorable evaluation of women’s potential for leadership roles. The second, prescriptive prejudice, stems from prescriptions about desirable female behaviors (e.g., be warm, nurturing; Eagly & Karau, 2002). For the bias view, studies on individual factors focus on biased self-assessments and stereotyping. Those on organizational factors study systemic bias, power dynamics, and hostility.
Individual level
The first individual factor involves gendered expectations and self-assessments. Gendered social role occupancy norms not only influence others’ expectations of women’s and men’s behavior but shape the development of gender-normative traits (Brown & Diekman, 2010). Gender socialization processes instill gender roles during childhood (Lippa, 2005), which are reinforced during adulthood by expectancy confirmation processes (Powell & Greenhaus, 2010). Individuals internalize expected social roles and regulate their behaviors based on gender-stereotypic expectations (Wood & Eagly, 2009). Diekman and Steinberg (2013) found that compared with men, women were less likely to consider their possible selves as high in status or in powerful positions with masculine traits. The internalized gender stereotypes lead to self-directed bias in women’s self-evaluation of their own fit with male gender–typed jobs (Heilman, 2012). In other research, women did not consider themselves equipped with the abilities and characteristics needed to satisfactorily perform these jobs (Heilman, 2012). Women choose not to enter male-dominated jobs even when qualified (Ceci, Ginther, Kahn, & Williams, 2014).
The second individual factor relates to individual stereotyping. Women who work in male gender–typed industries or leadership roles face stereotype threat, which is the likelihood of being judged or treated in ways that confirm negative stereotypes (Steele, 1997). The concern that one’s behavior can be explained by negative gender stereotypes leads to physiological stress response, increased monitoring of performance and the situation, and regulation of negative thoughts and emotions. These consume cognitive resources that can be used for work tasks (Johns, Inzlicht, & Schmader, 2008). Stereotype threat can harm women’s performance by creating anxiety (L. O’Brien & Crandall, 2003), harming expectancies (Smith, Brown, Thoman, & Deemer, 2015), and impeding STEM learning and knowledge acquisition (Appel, Kronberger, & Aronson, 2011). It also undermines women’s aspirations to seek leadership or work in male-dominated industries. As a coping strategy, individuals may avoid or withdraw from roles (Steele, 1997). When stereotype thereat was present, Smith and her colleagues (2015) reported that women’s motivation to pursue a career in science decreased, and they were less likely to recommend the field (von Hippel, Sekaquaptewa, & McFarlane, 2015). Davies, Spencer, and Steele (2005) found that exposure to gender-stereotypic commercials undermines women’s aspirations to take on a challenging leadership task. Su and Rounds’ (2015) meta-analysis showed that even after controlling for gender differences in interests, women were underrepresented in male-dominated STEM fields, such as engineering and computer science, and were overrepresented in female-dominated STEM fields, such as medical services. Women’s (and men’s) career choices were constrained by gender norms and stereotypes even beyond observed gender differences in interests.
Organizational level
With the passage of formal antidiscrimination laws (e.g., U.S. Civil Rights Act of 1964), overt forms of biases have been reduced. However, implicit bias persists in workplace systems (Cortina, Kabat-Farr, Leskinen, Huerta, & Magley, 2013). Implicit bias is when negative valence is unconsciously associated with a social object (e.g., women) and the biased behavior is not that obvious (Rudman, 2004). It sends powerful social cues creating differentiated workplace experiences for stigmatized and nonstigmatized groups (Emerson & Murphy, 2015). Implicit gender bias can be as damaging as explicit bias to work, career, physical, and psychological outcomes (Jones, Peddie, Gilrane, King, & Gray, 2013). Because women are not implicitly associated with abilities and traits required for leader roles (Eagly & Karau, 2002), they are less likely to be promoted for or seen as a good fit for management positions and masculine industries (Heilman, 2012). Since employers expect poor performance of women in these positions, hiring and promotion policies favor men, resulting in biased selection and advancement (Heilman, 2012). Roth, Purvis, and Bobko’s (2012) meta-analysis showed that although women received higher performance ratings, ratings of promotion potential favored men. Across numerous occupational samples, another meta-analysis (Joshi et al., 2015) found that gender differences favoring men in pay were almost 14 times larger than differences in performance evaluations (d = .56). Another study shows that female researchers were less likely to obtain named professorships, after controlling for performance (Treviño et al., 2015).
A second organizational factor is imbalanced career resources and power from social networks, role models, and mentoring/sponsorships (Michailidis, Morphitou, & Theophylatou, 2012). Women have to cope with hindered access to social networks (Casciaro & Lobo, 2005). Even when they occupy advantageous social network brokerage positions, their role is undervalued and occupancy benefits are reduced (Brands & Kilduff, 2014). Women also benefit less from supervisor and coworker support (Geller & Hobfoll, 1994). Having few women at the top creates barriers and limits access to role models and female mentors for young female employees (Hoobler, Lemmon, & Wayne, 2011). Regarding mentoring, the current research is mixed and barriers are subtle. While some studies report no gender differences in the amount of mentoring received (K. O’Brien, Biga, Kessler, & Allen, 2010), others report that male protégés receive more career development while female protégés receive more psychosocial support (Wanberg, Welsh, & Hezlett, 2003). Thus, career equality is also linked to the differential type of mentoring received as career mentoring may be critical to advancement.
A third organizational factor relates to the persistence of backlash and a hostile climate for female leaders. Posing a threat to existing norms, women’s entry in masculine fields and adoption of male-dominated roles is not universally welcome. Women can be penalized for role-incongruent characteristics and face “backlash.” According to the role congruity theory, not only do people often “think manager, think male” (Schein, 1973), but women can be punished or disliked when they hold managerial roles, as these roles are incongruent with stereotypical female traits and characteristics. Because agentic qualities are less desirable in women than in men, female leaders are often evaluated less favorably and are less likely to receive social approval for the same leadership behaviors (Eagly, 2013; Heilman, 2012). Compared to men, women leaders are more often faced with denial of credit for their success, personal derogation, and social disapproval (Heilman, 2012). Women have greater need to watch their emotion display at work because women’s negative emotion reactions are more often attributed to internal characteristics, whereas men’s emotion display is attributed to external factors (Brescoll & Uhlmann, 2008).
Interplay between individual and organizational levels
Much of the bias literature emphasizes how women are forced to leave jobs because of organizational factors. What needs greater attention is how gender bias and stereotyping get internalized, changing women’s self-concepts and shaping women’s self-efficacy and ability assessments (Heilman, 2012). In masculine cultures, it is more difficult for women to feel comfortable to gain authority or resources (Lyness & Thompson, 2000). Women may more frequently have to deal with expectations and criticisms that they lack hardiness, along with greater difficulties in building helpful relationships and networks (Timberlake, 2005). The potential for devalued performance may make women doubt their abilities, competencies, and qualifications (Heilman, 2012) and less likely to take credit for successful outcomes when collaborating (Haynes & Heilman, 2013). Yet Diekman and Goodfriend (2006) suggest that while women still face challenges in male-dominated contexts, women are gaining power by gradually attaining greater job responsibility. This change may encourage women to pursue careers in male-dominated environments.
Interventions
The gender bias view suggests focusing on change strategies to reduce explicit and implicit bias against women, stereotyping, and the adverse impact of career systems. Examples include increasing the proportion of women in leadership roles, using cluster hiring to reduce tokenism and stigmatization, leadership development promoting equality in selection and appraisal, and reducing ambiguity in evaluation and rewards.
Work–Family Perspective
The work–family perspective attributes career inequality to gender differences in men’s and women’s work–family experiences. 1 It emanates from women’s rising labor market participation and the growth of dual-career and single-parent female employees in organizations that are traditionally designed to support work role primacy and work–family separation (Kossek, 2006). Historically, drawing on work–family conflict theory, this view holds that work and family roles are incompatible, given conflicting expectations for time, energy, and behaviors (Greenhaus & Beutell, 1985). Studies focused (and still do) on the negative career outcomes (overload, strain, exit) for employed women, their children, and families and less on those for similar men (Williams, Berdahl, & Vandello, 2016). Perhaps this gap can be partly attributed to the fact that in general, reports of work-to-family conflict are much higher than family-to-work conflict (Byron, 2005). This systemic imbalance impacts those workers who are involved in family care the most (typically women). Despite this dynamic, examination of gender differences using the popular resource perspective on whether work–home processes of enrichment and crossover differ for men and women are limited to only a few sentences for future research in reviews (e.g., ten Brummelhuis & Bakker, 2012). The work–family perspective has underexamined multilevel processes (except supervisor relational or attribution studies; cf. Hoobler, Wayne, & Lemmon, 2009) and moderators, such as work–family bias, job level, occupation, class, and family structure (Kossek, 2006). For the work–family view, studies on individual factors focus on gender differences in work–family pressures, identities, and family structures. Organizational studies examine overworking, flexibility policies, and cultural support.
Individual level
The first individual factor involves unpacking the sometimes-conflicting sets of studies highlighting differential objective (actual domestic division of labor) and subjective work–family pressures (work-family perceptions and dynamics) for men and women. Working women consistently spend more time on child and elder care than men do (Bianchi & Milkie, 2010). Even among professional couples where both partners work full-time, over 40% of women report doing more child care, and nearly a third report doing more household chores (McKinsey and Company & Lean In, 2015). Turning to perceptions, seminal work suggests that the relationship between work–family conflict and job and life satisfaction may be stronger for women than for men (Kossek & Ozeki, 1998). A later replicative meta-analysis found sex and marital status weakly related to work–family conflict, with very small effect sizes as work interference with family was slightly stronger for men, and family-to-work interference was slightly stronger for women (Byron, 2005). Yet such meta-analyses focus on employed workers and underestimate gendered outcomes, as they do not capture women who would like to be in the labor force but are not due to family reasons, or vice versa for men. Studies need to move from measuring role occupancy to the actual time spent in family roles for men and women in similar jobs and household structures across cultures. For example, a study in Sri Lanka, which has traditional gender roles, found that men reported more work–family conflict if they had a nontraditional working spouse (Kailasapathy, Kraimer, & Metz, 2014).
A second set of findings focuses on the adverse impact on advancement and pay from having high family identities and great interest in and use of work–family flexibility. Compared to men, women are more family or “dual centric” (equally high identity with work and family; Kossek & Lautsch, 2012), have greater interest in jobs that have family flexibility, and are heavier users of flexibility practices (Kossek & Michel, 2011). These proclivities hinder women’s advancement as employers are more likely to promote and value work-centric employees who are committed to be “ideal workers” and make work the main focus of their lives (Williams, 2000). A study of Dutch physicians found incompatible gendered role prescriptions, whereby the “ideal mother” role conflicted with the “ideal physician” role (Pas, Peters, Doorewaard, Eisinga, & Largo-Janssen, 2014). And paradoxically, employed women’s use of flexibility and virtual multitasking to blend work and nonwork roles can backfire due to lower boundary control from higher integration (Kossek & Lautsch, 2012). Such coping strategies can cause lower work–life fit and greater role overload and psychological distress (Kossek & Lautsch, 2012). Further, using other flexibility strategies such as reducing hours or having career breaks for family creates a motherhood wage penalty, where women rarely catch up in earnings or advancement—a gap affecting high-skilled women across nations (Mandel, 2012).
The third individual factor relates to the differential effects of family structures on women’s career resources, opportunities, and outcomes. Women are more likely to be in low-income single-parent families, where finding child care and family-supportive jobs are major challenges (Bianchi & Milkie, 2010). Recent lab experiments show that being married was linked to attributions of higher male performance and less likelihood of layoffs, which were the opposite attributions for married women (Jordan & Ziteck, 2012). Three classic (and underreplicated) field studies on executives echo these complex linkages between family structure and career equality. Stroh, Brett, and O’Reilly (1992) found that executive men were 3 times more likely to have children (62%) and nearly twice as likely to be married (86%), respectively, compared to women (20%, 45%). Brett and Stroh’s (1997) second executive study focused on career mobility and found that women making a career move (40%) were over 6 times more likely to be single, compared to men (6%). A third study (Brett & Stroh, 2003) analyzed gender effects of working extreme hours—61 or more a week (referred to below as “overwork”). Results showed that most extreme-working men had a nonworking spouse at home—what they coined as a “facilitating resource.” In contrast, similar women executives had husbands involved in child care and/or paid domestic help. The economic payoff of working long hours was over 13 times greater for men. “Extreme” men made an average of $55,000 more, compared to only $4,000 more for women. These types of studies linking pay, hours, family structure, care arrangements, job level, and career and nonwork outcomes need updated replication to merge the gender bias, family, and career perspectives.
Organizational level
A first organizational factor involves the growing employer expectations to “overwork” (Cha & Weeden, 2014) and blur boundaries to be available 24/7, making it difficult to separate from the “always-on” workplace (Kossek, 2016). While executives may overwork as a tradeoff in return for their high status and pay, this trend has now expanded to include nearly all employees and especially professionals (increasingly women) who face demands where long hours (with little or no overtime pay) are equated with commitment and performance (Kossek, Ollier-Malaterre, Lee, Pichler, & Hall, 2016).
A second factor is the lack of access to and poor implementation of work–family policies. Research shows that use of formal policies, such as flexibility, may not lead to positive outcomes (Allen, Johnson, Kiburz, & Shockle, 2013). A quasiexperimental flexibility policy study found that formal use of telework did not reduce work–family conflict or improve well-being (Kossek, Lautsch, & Eaton, 2006). And most employers’ do not directly invest in dependent care resources—most have not invested in actual onsite child care support for decades (though health care employers, with their mainly female workforce and nursing shortages, remains an exception; Kossek, 2006). Yet finding and managing quality affordable child care and after school supervision of school age children and those with special needs remains a major underresearched career impact, with elder care a looming challenge.
A third factor relates to unsupportive work–family culture and stigmatization. Research shows that supportive work–family organizational and occupational cultures and schedules can reduce conflict in the structure of work and foster resilience (Kossek & Perrigino, 2016). A randomized control trial intervention experiment found that employees whose supervisors were trained to exhibit higher levels of family supportive behaviors were less likely to turn over, reported lower depression, and had higher job satisfaction (Hammer, Kossek, Bodner, Anger, & Zimmerman, 2011).
More research is needed on the fact that many companies stigmatize users of family-friendly policies (Williams, Blair-Loy, & Berdahl, 2013). Compared to men, women are more likely to access and use family-friendly policies, including part-time work, flextime, maternity and family care leaves, or to take time off when a dependent is sick. As a result, they are more likely to be stigmatized and penalized, experiencing lower pay and being skipped for promotion, which result in higher turnover.
Supervisor attributions clearly matter. One study found that supervisors, regardless of gender, were more likely to perceive women as having greater family–work conflict and as less promotable (Hoobler et al., 2009). A meta-analysis shows that supervisor work–family social support perceptions predict lower work–family conflict more than does perceived organizational support (Kossek, Pichler, Bodner, & Hammer, 2011) and that effect sizes of work social support for family were higher than effects from access to flexibility policies (Allen et al., 2013). Perhaps this is because supervisor performance attributions, gender, work hours, and hierarchical level may subtly moderate flexibility access, use, and outcomes. Take a recent study’s finding that when employees were attributed as using flexibility to support work, they were rated more positively than when the attribution was for family reasons (Leslie, Manchester, Park, & Mehng, 2012). Although this study also showed women worked fewer hours and were lower level than men, the scholars did not report linkages between gender, hierarchy, flexibility, hours, and performance. Yet research shows that most (90%) midlevel professionals working reduced hours are women (Kossek et al., 2016). Since most studies are cross-sectional, we cannot clarify whether high performers are given flexibility as a reward or whether use is attributed to performance.
Interplay between individual and organizational-level processes
When applying the work–family perspective to career equality, it is critical to examine individual and organizational multilevel processes. If women are more likely to use work–life flexibility policies that countervail the organizational culture of how to show career devotion, women will be more likely to be stigmatized and seen as having work–family conflict attributed to lower performance compared to men (Hoobler et al., 2011). Studies are needed on how women may alter or lower career ambition by choosing jobs that appear to be family-friendly or will not cause overwork and how they may face a “flexibility stigma” (Williams et al., 2013). Although overwork negatively affects the health and well-being of both men and women, work–family issues are framed as a women’s issue (Padavic, Ely, & Reid, 2013), perpetuating implicit bias.
Interventions
The work–family view suggests interventions should dually focus on how to not only increase work–family resources and facilitate greater control over work hours and boundaries but also reduce flexibility and family structure bias. Initiatives should promote leaders modeling balanced work–life behaviors, not overworking, and managing people without using “face time” as a proxy for performance. Longer paid or partial leaves for those who share child or elder care (with partner) should be implemented to foster gender egalitarianism.
National Socioeconomic Gender Context Influences
Clearly, regardless of disciplinary stance, the individual and organizational antecedents and interrelated dynamics are shaped by the national socioeconomic contexts in which they are embedded. As Figure 1 shows, these include many factors, such as cultural values on masculinity/femininity, societal socialization and expectations for gender roles, patriarchy ideology, national economic policy, and legal institutions governing property ownership, work–family policy, and equal opportunity (Metcalfe, 2007). Such factors influence the rigidity of norms for gender role congruity and socialization (Eagly, 2013) that regulate gender differences in behavior, which are rooted in cultural beliefs about the abilities and expected behaviors of men and women in work and nonwork domains. They also shape the degree to which management practices are gendered in implementation. Hofstede (1998) argues gender roles are more rigid in masculine cultures, where men are expected to dominate society, while roles are more flexible in feminine cultures, which tend to value gender equality. Studies show the overlaps of cultural forces on gender role congruity and leadership emergence. Take the case of Turkey, with a feminine culture, where individual factors, such as dominance, sex, and gender role orientation, were not found to predict leader emergence, unlike masculine cultures (Türetgen, Unsal, & Erdem, 2008).
National norms supporting traditional patriarchal cultures harm access to work–family supports. In such countries, because women are not expected to work outside of the home or get help with child care, a lack of cultural valuing of work–life supports may essentially perpetuate gender bias and labor market segregation. In a study of Middle Eastern companies, Metcalfe (2007) found that access to work–family supports, such as flexible work arrangements, was extremely limited. Training and career development opportunities were provided to men before women, as women were expected to give priority to family. A cultural factor was that gender and social relations were governed by a traditional patriarchal structure, with men seen as the sole family breadwinner (Metcalfe, 2007). The study found that women were expected to give up jobs when they marry and limit outside male contact. Many women faced barriers to owning a business, inheriting property, or reentering the workforce without spouse permission.
Wage disparity levels and economic growth may also shape the ability to afford in-home domestic care and economic benefits from equal gender labor market participation. Our review found that studies regardless of subfield approached career equality from a particular cultural point of view. Most of the literature we found was written from a Western view. Indeed, the very notion of “career equality” is decidedly culturally Western. Laws governing men’s responsibilities to protect women and to create a moral work environment are likely to be seen as perpetuating inequality using Western norms. Yet from a Middle Eastern perspective, such laws may be framed as preserving equilibrium in social relations (Metcalfe, 2007). We found few studies examining macrosocietal influences, a theme elaborated on in the discussion.
Climate for Gender Inclusion: Linking Mechanism
In this section, we turn to linkages between the antecedents and career outcomes via a cross-level mechanism, the climate for gender inclusion, which is a multilevel construct that should be explored in future studies. Defined below, it builds on research holding that climates have specific referents (Schneider & Reichers, 1983). It is grounded in seminal diversity climate research (Kossek & Zonia, 1993), which largely focused on equal-opportunity perceptions, from which current inclusion research evolved (Dwertmann, Nishii, & van Knippenberg, 2016) to also include social processes supporting “fitting in.” Individual-level climate, or psychological climate, pertains to the meanings that individuals hold regarding their work experiences (James et al., 2008). Climate at the organizational level reflects “shared psychological meanings” (James et al., 2008: 15). Assuming that organizations vary in their climates for inclusion for specific identity groups, we argue that they vary in their climates for gender inclusion. In measuring inclusion, it is key to distinguish between individual and organizational assessments. This enables differentiation between individual climate experiences and the taken-for-granted ways organizations perpetuate exclusion through institutionalized practices (Nkomo, 2014).
Individual inclusion involves perceptions of belongingness and uniqueness, as individuals seek to belong to and be valued by a social group without having to give up important identities (Brewer, 1991). It reflects perceptions that one’s self-concept or identity is supported and affirmed by the organization and its members (Shore et al., 2010). Researchers suggest that organizational inclusion includes equal-opportunity and fairness policies, leader actions, and positive coworker interpersonal interactions (Dwertmann et al., 2016). Also relevant is Nishii’s (2013) study of work units in one organization. She identified three inclusion factors: fairly implemented employment practices, inclusion in decision making, and integration of differences. These influenced the relation between unit performance and gender demography.
We replicate and extend inclusion research by reporting grounded climate issues emerging from the reviews as critical for women’s career equality. We define the climate for gender inclusion as the degree to which individuals and organizational social groups perceive and experience the work environment as one involving social interactions, cultures, and structures that are supportive of and effectively use the varied identities and values women bring to work in ways that foster their belongingness and ability to leverage their talents to contribute to the organization. Our reviews suggest three coalescing multilevel dimensions, which we elaborate on below: fairness for women, leveraging women’s talent, and workplace support for women’s values, interests, and needs. The grounded dimensions certainly have alignment and construct congruence validity with the general inclusion measures noted above (Dwertmann et al., 2016; Nishii, 2013). Following Schneider and Reichers (1983), the dimensions we identify add to previous work by including specific referents to women; a focus on women’s career equality issues, such as bias involving women’s leadership, greater use of work–life policies, and valuing communal roles; and a suggested ordering of dimensions that fairness must be in place before talent leveraging and support can be enacted. Figure 1 also suggests multilevel measures.
Dimension 1: Fairness and Nondiscrimination for Women in Work Access, Process and Outcomes
The first climate dimension captures the degree to which leaders and members would agree that work practices are in place and leaders take action to ensure fairness and nondiscrimination for women (Kossek & Zonia, 1993; Nishii, 2013). Collective perceptions would concur that organizational practices are implemented in ways that eliminate implicit bias (e.g., gender stereotyping, backlash, gendered work–life processes) and explicit bias (e.g., ending harassment, pay discrimination, job loss after childbirth) toward women. At the individual level, individuals would report perceptions of distributive and process justice and nondiscrimination in women’s personal job experiences. By far, the most research has been done on the fairness-and-discrimination dimension (Ely & Thomas, 2001). It is a critical foundational dimension as ensuring fairness is a necessary but insufficient condition that must be in place to preempt discriminatory processes in order for the other proactive dimensions to occur. While implementing fairness practices enables pluralism and increases representation in women’s hiring, we have noted that many firms typically expect underrepresented groups to assimilate into the norms of the dominant culture as a one-way socialization process (Nkomo, 2014). Such approaches do not necessarily leverage the talents of women, promote learning (Ely & Thomas, 2001), or actively change norms or resource allocation to support and advance women. Once there is consensus that equal-opportunity practices are in place, it is possible to have synergies from women’s diversity as a source of learning and skill utilization (Ely & Thomas, 2013: 1754) and for accessing new markets and benefiting from women’s views (Dwertmann et al., 2016).
Dimension 2: Culture and Strategy Support P-E Fit to Leverage Women’s Talents
Previous research (Ely & Thomas, 2001) suggests that a key second dimension of the climate for gender inclusion pertains to degree to which the collective and the individual experience the organization culture and strategy as having high P-E fit to learn from and use women’s talents. The climate enables the organization to utilize the skills and knowledge from women who may have social identities differing from dominant hierarchical groups (e.g., men). At first, members value women to sell to similar identity groups (other women) to mirror the market, as Avon does in its international subsidiaries. Then a norm develops that regardless of the business issue, the organization is able to learn from and incorporate women’s views and use their skills broadly. At the individual level, a person would perceive that women are able to contribute skills and knowledge and have high P-E fit. P-E fit measures would capture individual perceptions that the job context generally fits women’s personal interests, values, goals, and needs (Kristof-Brown & Billsberry, 2013). Related measures might include person–organization fit, assessments that women generally are a good fit for the organizational values and cultures, which would in turn foster their career advancement. In gender-inclusive environments, men and women will experience equal levels of P-E fit for the job and for work–life (Barnett & Brennan, 1997). Individuals would agree that women do not need to have major sacrifice of their gender-related identities or need to hide their true selves in order to advance. Such environments also likely have communal and relational values represented in the organizational culture and norms and well integrated into HR practices.
Dimension 3: Workplace Support for Women’s Interests, Voices, and Needs
This dimension captures individual and collective agreement that the organization actively supports women’s values on and off the job, including their needs, views, and interests. Workplace support is defined as the degree to which individuals believe that their well-being is valued and supported by supervisors and the organization as a whole (Eisenberger, Singlhamber, Vandenberghe, Sucharski, & Rhoades, 2002). Support can be psychological, such as perceptions of the degree to which women individually and as a whole generally agree they are valued and appreciated by leaders and coworkers. Individual women would report feelings of belongingness and psychological safety to speak up and would feel included in decision making at different hierarchical levels (Nishii, 2013). Support can also be tangible. Individuals and the collective would agree that women have access to direct or indirect resources to advance in the job and for nonwork roles, such as communication of information, joint problem-solving to remove career barriers, or tangible assistance (Kossek et al., 2011). In sum, this dimension reflects perceptions that leaders and colleagues are encouraging of women to be able to share their personal values, roles, and needs related to their gender identity at work and not have to hide or give up these salient values in order to be successful (and not stigmatized) at work and in personal life.
Climate for Gender Inclusion Links to Career Equality Outcomes
Improving the climate for gender inclusion at the individual or collective level is the mechanism to enhance career equality (see Figure 1). Individuals and groups within organizations will experience the workplace through accumulated social interactions and adapt their attitudes, beliefs, and behaviors to the reality of their social environment. The climate for gender inclusion is the pathway to affect both individual-level career outcomes, such as equal career success and positive nonwork experiences, and organizational-level outcomes, such as equal representation of men and women, vertically and horizontally. As our preceding review has suggested, individual and organizational cross-level dynamics reinforce climates and career outcomes. The more that climates are seen by members as highly gender inclusive, the more there will be greater alignment between individual- and organizational-level perceptions. Consistent with attraction-selection-attrition theory (Schneider, 1990), women are attracted to, select into, and stay in or have attrition from contexts based on the degree to which they experience the climate as being gender inclusive. Although firms are hiring increased numbers of women, work climates are not yet aligned to fully support gender inclusion. Simply adopting mentoring, bias, or work–life policies, when not reinforced by a gender-inclusive climate, will likely result in adverse mechanisms, like stereotyping, and it is unlikely they will be implemented in ways that lead to career equality. Regarding outcomes, the model suggests that the degree to which individual women experience equal career success and positive work–nonwork experiences relative to men is linked to organizational career equality indicators (equal representation of men and women, implementation of gender-equality HR policies).
Future Directions on Women’s Career Equality
Our integrative review suggests overarching themes to advance research and practice.
Stop Splitting Disciplinary Narratives in Research and Interventions
First, rather than isolating views, integrative approaches are needed to provide a holistic picture of career equality that takes into account cross-disciplinary organizational- and individual-level factors for research and interventions. The career preferences view suggests that offering opportunities supporting relational and communal values is likely to advance women. The gender bias view highlights institutionalized barriers, such as stereotyping and unfair work practices. The work–family view underscores critical resources, such as flexible work forms, that can be used without stigma to support more varied ways of advancing. Each perspective must be simultaneously considered for disconnects, conflicts, and synergies.
Studies are needed that clarify the sequencing of conditions and initiatives by subfield. Gender bias interventions need to be successful (reduce bias first) in order for career preference or work–life interventions to be successful. We need to better understand the extent to which overcoming the issues associated with one perspective addresses necessary but insufficient conditions that must occur in order to begin to address the challenges associated with another. Notions of how to change organizations to foster women’s career equality may have been less effective and even in conflict because of disciplinary siloes. Each perspective alone has drawbacks yet may be powerful when connected. The work–family view underexamines removal of career stigmas for using flexibility. Interventions mandating women’s leadership quotas or implicit-bias training that do not address fundamental conflict in work design creating work–family conflicts will not be successful. Making changes to support women’s career preferences without addressing institutional biases preventing advancement cannot be effective.
Attend to “Opting-Out” and “Pushed-Out” Cross-Level Dynamics
Second, our review shows that the “opting-out” and “pushed-out” explanations for women’s career equality are not in conflict but coexist. We need to unpack the cross-level dynamics shaping labor market pulls: the processes underlying the selection of women into contexts that are supportive of individual career needs. Examples include identifying characteristics of organizational contexts that fulfill people-oriented career interests, reducing implicit and explicit bias, and signaling work–family balance. Integrative research is essential on these multifaceted pulls to understand pathways and relative impacts on outcomes for specific women’s labor force segments and occupations. The context factors preventing selection of working in a startup may be different than those preventing going to medical school.
Similar research is needed to better understand the converging chilling effects of multiple contextual constraints, or pushes. Women are less likely to be effective in their careers or remain in organizations that are experienced as less gender inclusive, which constrain career equality outcomes. Women are unlikely to advance in firms that do not see their strengths, values, and interests as aligned with leadership. Women are unlikely to be viewed as leaders in contexts where they culturally and structurally lack access to powerful networks and are often stereotyped. Women will not remain in firms that foster long hours and lack structural supports to enable them to manage work–life demands without cultural stigmatization. Few studies identify how individual and organizational career preference, bias, and work–family barriers to career equality dovetail and contribute unique effects. Yet all perspectives converge on the need to provide multifaceted workplace resources specifically targeting women’s career equality, as specific (over general) support will have the most robust impact (Kossek et al., 2011).
Seriously Consider National Socioeconomic Influences
Third, a persistent theme is the need to give far greater attention to national socioeconomic and cultural gender context influences. Future studies need to increasingly specify and examine the influence of the national context of the sample and its impact on individual or organizational antecedents and the relevance of different perspectives. Career preferences, gender role beliefs, and work–life norms are inherently grounded in gender role socialization and social values regarding household caregiving roles and division of labor (Eagly & Wood, 1999). Public policy and legal institutions (e.g., maternal/paternal leave policies are sanctioned, women can legally work outside the home during or after pregnancy) are at play. Clearly more research is needed on how such factors instill expectations for prototypical masculine and feminine roles and sanctions for role violations (Eagly & Wood, 1999) and equality in a cultural context.
Future research should examine whether certain perspectives are more explanatory in some national economic and cultural contexts than others using cross-national databases. Such studies might carefully compare career-equality antecedents and outcomes while controlling for job, level, or organization. In high-GDP, free-market, high-wage countries, the work–family perspective may be more influential in explaining career inequality, since it is generally more expensive to hire “home help,” compared to countries with lower wages (China, Indonesia, Hong Kong), where it is more affordable to have in-home nannies and other household help. Historical gender patterns are in flux, though playing out in culturally specific ways. Metcalfe and Afanassieva (2005) argue that gender equality may be eroding in former socialist countries due to the transition from a communist society to a market economy. They suggest that the historical higher representation of women in senior positions is at risk due to declines in state-provided child care and growing recruitment discrimination. Turning to Korea, which ranks low globally among industrialized nations on gender equality, attitudes on division of labor (taking care of the house is primarily the wife’s responsibility) are slowly shifting in working millennial couples compared to older couples. Facing a slowing economy and plummeting fertility rates has led to what is termed a “Sampo generation” that gives up dreams of marriage, buying a house, or having children. Studies are needed on dual-earner couples’ navigation of work–life and career dynamics and organizational supports in such countries that are shifting gender equality. For example, are wives generally socialized to interrupt work for personal life during the day, unlike husbands? And do their companies support boundary permeability? How does this link to career equality? Such studies might classify family earning and household domestic structures and dependent-care configurations. We need a return to basics of how couples in like professions manage the logistics of child/elder care and timing of childbearing and how they shape career equality in specific economies and cultures.
Improve Measures of Gender Inclusion Climate, Career Equality Outcomes, Implicit and Explicit Bias
Fourth, we challenge researchers to develop improved cross-level gender inclusion and career equality measures. Future research should develop comprehensive interdisciplinary measures of career gender equality and climate for gender inclusion, shown in Figure 1. We identified three multilevel dimensions: fairness and antidiscrimination for women in work access, process, and outcomes; leveraging women’s talents; and workplace support for women. Studies should confirm that high effectiveness on fairness and antidiscrimination is a necessary prerequisite that must be present in order to leverage women’s talents and ensure high levels of women’s workplace support. New measures are needed to better link implicit and explicit bias. We have seen abundant studies focusing on promotions, performance ratings, and salaries. We also need research on subtler outcomes to capture how gender bias is implicitly displayed in the workplace and its implications for career outcomes. For example, we need to build on studies on the different attributions for men and women of the same emotional expression (Brescoll & Uhlmann, 2008), their different rewards for the same organizational citizenship behaviors (Heilman, 2012), and different consequences from using social networks (Brands & Kilduff, 2014) and their links to climate and career equality. Work–family implicit bias is underresearched, such as different attributions and career outcomes for men and women based on face time, primacy of family roles, and electronic availability.
Regarding career equality outcomes, studies should simultaneously measure women’s organizational horizontal and vertical integration and equality in work and nonwork outcomes. Multilevel career equality measures need to be developed focusing not only on the individual and organizational levels but also on the team, societal, and family levels. Cross-disciplinary research teams may be more likely to design such comprehensive measures. Studies might also move toward nuanced analysis of how gender interacts with HR policy implementation. What is lacking is cross-level analysis of the differential consequences for men and women for using work–family policies, for selecting into jobs that are more family supportive (but less rewarded), or for acting in ways that countervail current organizational leader and gender role norms.
Address the Research-to-Practice Translation Gap
Fifth, we need to better address the research to practice translation gap for comprehensive interventions. Reports from a conference on gender equality stated that some participants argued that there has been more than enough research and not enough action illustrating the research-to-practice gap (Miller, 2015). Research on women’s careers has become a politicized cottage industry often sold off to the most connected consultant, who often lacks research-based solutions. We need to have researcher, organization, and policy teams to support evidence-based intervention development and evaluation with rigorous certification of best practices.
Table 1 provides examples of women’s career equality metrics, scholarly and lay concepts, and evidence-based individual and organizational interventions to integrate scholarship and practice for a common language and broaden the possibilities for change. For example, in order to monitor and improve the career equality metric of equal job access, one would need to address the scholarly concept of removing the cultural and structural barriers to career advancement, which would tackle the practical narrative that women face an “unequal playing field.” A possible organizational strategy would be to ensure an “open” system for candidate nominations and to train recruiters how to remove implicit bias in selection, such as removing gender bias in résumé review and ensuring all candidates are asked the same interview questions and provide similar background information. Bohnet, van Geen, and Bazerman (2012) refer to this as an evaluation “nudge.” They note that for entry-level positions, junior men and women are often evaluated as a group, which helps level the playing field as men and women enter firms with equal credentials. Yet at middle and senior levels and for more complex projects, women and men are often selected and judged separately. Going back to group-based credential sharing would help level the playing field. An individual-level strategy would be to communicate openings using gender-neutral language. Linguistics research shows that using words like “we’re looking for rock stars” is more likely to attract men than women. If the ad states an interest in recruiting “leaders,” this gender-neutral term appeals equally well to both sexes (Snyder, 2016).
Examples of Multilevel Forms of Women’s Career Equality Interventions: Integrating Scholarly and Practice Views
Note: STEM = science, technology, engineering, and mathematics.
New policies are emerging that need to be evaluated for whether they are adopted more for publicity or to actually foster change. Many startup and established firms have hired more women, but they have not necessarily changed the structure of work to be more family supportive, leading women to turn over or downshift careers after having children. Companies are offering mixed messages with high-profile policies, such as offering to pay for egg freezing for later in vitro fertilization, as Facebook recently did. The message implicitly is “we would rather encourage you to freeze your eggs to have a baby later in your career than to restructure work now.” Facebook also increased the length of paid paternity and maternity leaves, which CEO Mark Zuckerberg recently used as a role model. Amazon just adopted a 30-hour mommy track with full benefits for women. Studies need to examine whether women and men at all levels can actually use work–family policies without career derailment, as the jury is still out on their level of cultural integration.
Evidence-based training
Research-based training deserves focused attention as it is one of the most common interventions used and the least scientifically evaluated. Research shows that although some major technology companies are spending considerable money on training, such as implicit-bias training, such efforts have not been effective as they have not significantly resulted in improved gender representation and could in fact increase stereotyping and backfire if poorly implemented or delivered in a way that blames participants (Miller, 2016). Most training initiatives were simply not subject to research scrutiny. Future research should identify the key ingredients for effective development and delivery of evidence and theoretically based training interventions designed to increase career equality. Studies are needed that identify how to frame gender career equality as a shared communal challenge. Instructional design research is needed to develop learning, change, and communication strategies that are linked to evaluation of identified career outcomes. The change targets of career preference, gender bias, and work–life strategies could be integrated and Q-sorted by experts into cognitive, behavioral, and attitudinal outcomes.
Evaluate and Conduct Rigorous Comparative Effectiveness Intervention Experiments
Given the social challenges in studying gender inclusion and bias, much of the rigorous experimental research is conducted in laboratories. Due to frequent challenges in gaining organizational access, most field research is in a single firm. Studies tend to be either field-based or lab experiments but not well linked, fostering gaps—often one that has quietly hired a consultant to try and improve gender practice. Quasiexperimental and randomized-control studies are needed that build on each other, moving iteratively between lab and field experimentation and practice evaluation. We need comparative effectiveness studies varying across sociocultural contexts. Experiments are needed that compare the efficacy of different types of interventions within and across contexts and individual and organizational levels. Such research might compare the career efficacy of organizational-focused, individual-focused, or multilevel change approaches and the degree to which the workplace is experienced as supportive or controlling as well as inclusive or marginalizing.
Studies might identify whether single or multidisciplinary forms of leader and organizational support interventions are most effective and produce lasting change strategies. Is focusing on work–family but not gender bias or career issues less effective than when synergistically implemented? While we know that family-supportive supervisors are essential to reduce work–life conflict (Hammer et al., 2011), this support may be coming at a cost to women’s advancement. Studies are needed to explore whether high use of work–life support by women (and less so men) is linked to leader attitudes that women have higher work–family conflict than men and are less career motivated. Experiments might assess whether career resources are most effective when implemented as a focused strategy designed to increase access to leader role models and mentors, compared to a multifaceted strategy where mentoring resources are simultaneously implemented with sponsors who take action to countervail gender bias. Perhaps when women-focused mentoring is implemented as an isolated strategy, women are seen as needing more “career help” than men, which perpetuates stereotypes that do not support advancement.
Last, these studies need to attend to contextual moderators and understudied samples. The equality gap may vary depending on the job prestige, gender demography, and supportiveness of the context (Joshi et al., 2015). One strategy that merits future research is on how to reduce ambiguity over competency for selection decisions where jobs have historically been male dominated. A meta-analysis found that gender bias was more likely to occur in contexts when there was ambiguity over the competency of the applicants, but when higher competence was clear, bias was reduced (Koch, D’Mello, & Sackett, 2015). We also need more studies on overlooked and emerging occupational samples, such as entrepreneurship. We also need to study people not in the labor force, as there are unemployed women who want to have careers or different avocations who may have been pushed out due to the factors we identified, yet they are missing from the literature.
Conclusion
Women’s career inequality harms not only women but families and society. Often, workplace initiatives focus mainly on changing women instead of changing contexts that disadvantage women. Integrating bias, career, and work–family narratives in women’s lagging advancement bridges the “opting-out” and “pushed-out” views to advance research and policy.
Footnotes
Acknowledgements
We would like to thank associate editor Eden King and several anonymous reviewers for helpful feedback. We thank Dr. KungHee Lee for thoughtful suggestions on this manuscript. An earlier version of this paper was presented at the Work Family Researchers Network conference in Washington, DC, on June 23, 2016.