Breaking Through the Glass Ceiling: Social Capital Matters for Women’s Career Success?

ABSTRACT

This study examines the relationship between social networking as a potential upward mobility strategy for women and their career advancement. The federal data from the 2007 Career Advancement Survey (CAS) were analyzed using ordered logistic regressions and OLS multiple regressions. The results showed that having a supportive supervisor and/or mentor was positively related to the likelihood of being appointed to critical roles and being assigned important work, while having important contacts was positively associated with temporary promotions. In addition, engaging in professional and formal networks showed a marginally positive association with work assignment. Gender congruence in supervisory dyads was positively associated with role assignment. The findings regarding gender differences in relationships reported a stronger positive relationship between having a supportive supervisor and/or mentor and temporary promotions for women than for men. Women were more likely to be temporarily promoted than men when they have a supportive supervisor and/or mentor. Women who have a female supervisor were more likely to be temporarily promoted and to be assigned critical roles than men or women who have a male supervisor.

Notes

As more women obtain higher education, their proportion in the workforce has also increased significantly in recent years. Nevertheless, because they sought employment later than men, the more powerful positions that often confer the greater authority of decision making are likely to be dominated by men. The underrepresentation of women in higher positions in organizations offers evidence for the argument.

The concept of social capital is distinct from that of networking (Wolff and Moser 2009). Social networking is defined as “individuals’ attempts to develop and maintain relationships with others who have the potential to assist them in their work or career” (Forret and Dougherty 2004:420). Wolff and Moser (2009) noted that social networking, which is an individual-level construct, should be distinguished from social capital, which refers to a structural-level concept. Social networking focuses on individual behaviors or actions that build and foster relationships and contacts, while social capital is associated with “the quality of the position of an individual in a network” (e.g., holding a supervisory position or a highly critical position) and the characteristics of network structures (e.g., network size, density, or structural holes) (Wolff and Moser 2009:197). Individuals can acquire social capital by engaging in various social networking behaviors, such as joining professional associations, participating in informal groups within the organization, mentoring, and seeking highly visible assignments (Forret and Dougherty 2004).

Burt (1998) argued that “legitimate” members of the population in social structures, such as men in senior ranks, accomplish higher returns to social capital than “illegitimate” members, including women and young males. While young men eventually gain legitimacy as they enter the senior ranks, women are likely to remain illegitimate. Burt (1998:6) suggested that it will be more successful in competition for illegitimate members (e.g., women and young men) to borrow the social capital of legitimate members than to build their own.

Despite a debate over the use of survey weights, more researchers seem to favor analyzing the weighted data because of the increased accuracy of statistical inferences by correcting for a sample’s disproportionate representation of the target population or “unequal sample inclusion probabilities” (Pfeffermann 1993:317) and compensating for nonresponse errors.

The results from the models without agency controls were slightly different from those with agency fixed effects. For example, the positive moderation of gender on the relationship between having same-gender supervisors and role assignment proved to be insignificant after removing agency controls. On the other hand, the insignificant moderating of gender on the relationship between gender congruence and work assignment has become marginally significant. This suggests that agency effects in the models are marginal, but should not be ignored.

A survey question for permanent promotion was not available. Temporary promotions and details refer to informal assignment of employees to positions above their current grade level. In general, employees detailed to a higher-graded position for 30 consecutive days or more are to be temporarily promoted, if Office of Personnel Management qualifications are met. Unless agency instructions require that an employee who is on a temporary promotion be returned to the former position prior to achieving a permanent promotion, an employee on a temporary promotion can receive a permanent promotion (US Office of Personnel Management 2016:14–3).

Wolff and Moser (2009:198) created a 44-item German measure of networking, which reflects the multidimensional construct of networking. They specifically developed six scales using 44 questions: building internal contacts; maintaining internal contacts; using internal contacts; building external contacts; maintaining external contacts; using external contacts. Due to the limited availability of appropriate survey items, this study developed only some of the scales using a single survey item for each.

To provide evidence of convergent and discriminant validity, the intercorrelations of four survey items were examined. Two out of four items converged on the same construct and were summed as one variable. Finally, the intercorrelations of three items demonstrated discriminant validity of the measures.

Gender is a social and cultural construct that describes life experiences, socialization, and social status of the individual, while sex is a biological term that is related to anatomical structure of the individual (Diamond 2002). Even though social scientists argue that gender rather than sex is likely to lead to the differences in an individual’s behaviors and perceptions, they have used sex as a proxy measure of gender because of lack of alternative measures.

According to Burt’s “gender of social capital” (1998), men may expect more from supervisory support for their promotion (as greater beneficiaries from social networking) than women. His argument was not fully supported by the finding of this study, which may explain the influence of governmental policies and managerial effort to establish representative bureaucracy by supporting women with more opportunities for advancement.

Burt (1998) indicated men at senior ranks as strategic partners from which women can borrow social capital. This study, however, showed that women who have same-gender supervisor or female supervisors are more likely to be temporarily promoted than others. This might suggest that supervisors’ willingness to help their subordinates plays a more critical role in generating desirable results for the subordinates than whether supervisors are legitimate members of the majority group in social structures. For example, if male supervisors with legitimacy are not willing to accept their female subordinates as members of their own group, the female subordinates cannot even borrow their supervisors’ social capital.

Laud and Johnson (2013) generated the typography of the 15 tactics used by 187 male and female high-level executives from various organizations (107)—industry, education, government, nonprofits, and military—through interviews. The results showed that female officials are more likely to have used “networking” tactics (e.g., mentors, sponsorship, relationships) than men to accomplish their career advancement goals.

The seemingly neutral human capital inputs and outcomes may not be completely gender-neutral measures but, in fact, are closely intertwined with socio-psychological assumptions, such as women’s gender roles and stereotypes, which often negatively influence the image of women in employment.

Critics argue that the impact of the social construct of gender should not be overlooked when it comes to differentiating women’s and men’s work patterns and behaviors, and preferences and choices (Lips 2013a). For example, to fulfill gendered domestic responsibilities, more women may have shorter average tenures, less work experience, and less opportunities to obtain human capital investments (e.g., education and training) than men (Becker 1985; Blau and Kahn 2000).

Additional information

Notes on contributors

Sungjoo Choi

Sungjoo Choi ([email protected]) is an associate professor in the Department of Public Administration at Kyung Hee University in Seoul, South Korea. Her research interests include innovative work arrangements, diversity management and representative bureaucracy, and gender issues in government. She has published articles in Journal of Public Administration Research and Theory, Public Administration Review, International Public Management Journal, and Review of Public Personnel Administration.