SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Sex, Gender, and Income Attainment
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Biographical Information
  10. Biographical Information
  11. References

Drawing on gender role theory and tournament theory, we examined the effects of sex and organizational culture preferences on the incomes of MBA graduates over an 8-year period. We found that masculine culture preferences led to higher income 4 years after graduation and, in contrast to previous research, the effect was stronger for women. By 8 years after graduation, however, men's rate of income increase was significantly higher than women's, an effect that was mediated by hours worked per week. These findings clarify some of the conflicting results of previous research on the effects of gender roles on women's careers and suggest that a tournament-like aspect of careers may account for higher incomes in organizations. Copyright © 2009 John Wiley & Sons, Ltd.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Sex, Gender, and Income Attainment
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Biographical Information
  10. Biographical Information
  11. References

Wage inequality between men and women has attracted the attention of psychologists (e.g., Lam & Dreher, 2004), sociologists (e.g., Petersen & Morgan, 1995), economists (e.g., Blau & Kahn, 2006), legal scholars (e.g., Browne, 2006), advocacy groups (e.g., Catalyst, 2007), and government institutions (e.g., United Nations, 2005). The fundamental question addressed in this research is: Why do women earn less than men, even after controlling for occupational, organizational, and human capital variables? The answer to this question is important for both practical and theoretical reasons. While the gap between men and women's earnings has narrowed over time (Blau and Kahn, 2006), women are still under-represented in the highest paying jobs and industries (Blackburn, Jarman, & Brooks, 2000; Catalyst, Catalyst, 2007). Although career success can be defined in a myriad of ways, income attainment is important because it provides a quantitative measure of the monetary value assigned to women's labor and allows for comparison across institutional and societal contexts, as well as the examination of historical trends (e.g., Blackburn et al., 2000; Heslin, 2005; Nicholson & De Wall-Andrews, 2005).

One prominent psychological explanation for the male–female gap in earnings is based on gender role theory (Eagly, 1987; Eagly & Karau, 2002; Eagly & Wood, 1991). In this view, gender roles—those socially constructed beliefs that describe how males and females are expected to behave—disadvantage women since masculinity is most often associated with the stereotype of a successful manager while femininity is seen as incongruent with this role (Ely & Meyerson, 2000; Kirchmeyer, 1998). One way in which inequality occurs is from the belief that women are more nurturing and less aggressive than men and, therefore, less prone to act as a leader or be promoted to higher paying management positions (Eagly & Karau, 2002; Johnson, Murphy, Zewdie, & Reichard, 2008). Gender role theory can also apply to beliefs about the masculinity or femininity of specific jobs (e.g., Chatman, Boisnier, Spataro, Anderson, & Berdahl, 2008). Inequality is thought to occur this way when employers factor gender roles into decisions regarding the hiring, promotion, and compensation of women who are seen as incongruent for a particular job or role (e.g., manager), resulting in women earning less than men.

Although this logic suggests a direct link between gender roles and the male–female earnings gap, the empirical evidence is quite mixed. Recent studies indicate that the wage gap in earnings may stem not from gender roles per se, but instead from a complex set of interactions among sex, gender, and organizational context (e.g., Goldberg, Finkelstein, Perry, & Konrad, 2004; Kirchmeyer, 1998; Sools, van Engen, & Baerveldt, Catalyst, 2007; Tharenou & Conroy, 1994; van Vianen & Fischer, 2002). Noonan, Corcoran, and Courant (2005), for example, found that the difference in earnings between male and female attorneys over time was better explained by labor supply (i.e., hours worked) than by sex or human capital variables. However, this study, like many others (e.g., Goldberg et al., 2004; Lyness & Judiesch, 2001; Melamed, 1995; Schneer & Reitman, 1995; Van der Velde, Bossink & Jansen, 2005), only measured sex differences, so it is not possible to examine gender as a potentially confounding effect. Interestingly, when gender is assessed separately from sex, there is evidence that masculinity positively predicts career advancement and femininity negatively predicts career advancement, even after controlling for human capital and promotion opportunities (Baril, Ebert, Mahar-Potter, & Reavy, 1989; Dimitrovsky, Singer, & Yinon, 1989; Tharenou, 2001). Further, the results of several studies indicate that when both sex and gender are investigated, gender may have more important effects than sex alone (Lobel & St. Clair, 1992; Kirchmeyer, 1998; Konrad & Cannings, 1997; Sools et al., Catalyst, 2007; van Vianen & Fischer, 2002).

A key challenge in using gender role theory to explain wage inequality is that gender roles are distal and difficult to capture outside the laboratory context (e.g., Rudman & Glick, 1999; Heilman, Wallen, Fuchs, & Tamkins, 2004; Fiske, Cuddy, Glick, & Xu, 2002). Another challenge is that despite large sample sizes and longitudinal data (e.g., Kirchmeyer, 2002; Lyness & Judiesch, 2001; Melamed, 1995; Schneer & Reitman, 1995; Stroh, Brett, & Reilly, 1992; Tharenou, Latimer, & Conroy, 1994), few replicable results have emerged from field studies of women's income attainment. Even the wage disparity between male and female managers, a basic motivating issue behind many of the studies, is not always confirmed (e.g., Lyness & Thompson, 1997, 2000; Powell & Butterfield, 1994). While there can be little doubt about the overall wage differences between men and women at a societal level (Blau & Kahn, 2006; Petersen & Morgan, 1995), as Phillips and Imhoff (1997) noted over a decade ago, the intermediary links between gender roles and managerial women's income attainment warrant further investigation.

Labor supply may provide a key insight into the process by which gender roles affect wage inequality over time. According to tournament theory (e.g., Lazear & Rosen, 1981; O'Reilly, Main, & Crystal, 1988; Rosenbaum, 1979), careers unfold as a series of tournaments in which employees at lower levels compete with each other for career advancement (Conyon, Peck, & Sadler, 2001; Main, O'Reilly, & Wade, 1993). While gender role congruence may be an important determinant of income earlier in careers, if some employees consistently work more hours than others, it may be that those who work more also earn more over time. If employee work hours differ along sex and gender lines—a reasonable possibility given that women bear a disproportionate share of family work (Konrad & Cannings, 1997; Noonan et al., 2005) and stereotypically masculine organizations require longer work hours (Brett & Stroh, 2003), then over time, women will earn less than men due to differences in labor supply, and less so because of gender role incongruence. In order to test this theory, however, it is necessary to have longitudinal data on gender, sex, and income attainment.

As this brief overview illustrates, studies of women's income attainment leave several important empirical and theoretical gaps. The purpose of this study is, first, to disentangle the separable effects of sex and gender in order to examine the independent and interactive effects of gender roles on wage inequality and, second, to test a theory of gender wage inequality in which gender roles predict wages in the early stages of careers, while tournament processes predict wages in later stages. To accomplish this, we examined relationships among sex, gendered organizational culture preferences, work hours, and income attainment, and demonstrated how these relationships change over an 8-year period following graduation from an MBA program.

Sex, Gender, and Income Attainment

  1. Top of page
  2. Abstract
  3. Introduction
  4. Sex, Gender, and Income Attainment
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Biographical Information
  10. Biographical Information
  11. References

Elucidating the distinction between sex and gender is fundamental for understanding the gap between men and women in income attainment (Abele, 2003; Hyde, 2005; Judge & Livingston, 2008). While sex is biologically determined, gender is a social construction, a product of learning, socialization, and experience (Unger, 1979). The terms “masculine” and “feminine” are often used for characteristics and traits that are considered socioculturally appropriate for males and females (e.g., Bem, 1974; Spence & Helmreich, 1978; Unger, 1979).

Recent research has proposed that a more appropriate assessment of masculinity and femininity should be based on actual sex differences in values, interests and preferences, as represented in the population being studied (Lippa, 1998, 2001; Lippa & Connelly, 1990; Young & Sweeting, 2004). One method for assessing gender using domain-specific preferences that differentiate men from women, while capturing the variation within sex, is the gender diagnosticity approach (Lippa & Connelly, 1990; Lippa, 2001). In this approach, the essence of masculinity and femininity lies in gendered interests and preferences (Lippa & Connelly, 1990; Lippa, 1998, 2005), that is, the specific characteristics that differentiate males and females in a sample population. Masculinity and femininity reflect how “male-like” or “female-like” a person's pattern of values is in comparison to local reference groups of males and females (van Vianen & Fischer, 2002).

Organizational culture preferences provide one such way of capturing gender variation in a managerial population. Extensive research shows that people as well as organizations differ in terms of values, and that a fit between a person's values and an organization's values leads to more positive outcomes for both employees and the organization (O'Reilly, Chatman, & Caldwell, 1991). Values, like other facets of organizational life, are not gender neutral (see Acker, 1990, for a review). Historically, masculinity has been characterized by dominance, aggressiveness, and competitiveness while femininity has been represented by warmth, supportiveness, and caring (Abele, 2003; Helgeson, 1994; Lubinski, Tellegen, & Butcher, 1983; Saragovi, Koestner, Di Dio, & Aubé, 1997). Two factorially independent organizational culture dimensions that correspond to these constructs are aggressiveness and supportiveness (O'Reilly et al., 1991). From a gendered organizational culture perspective, masculinity and femininity can be indexed by the extent to which a person prefers an aggressive organizational culture versus a supportive organizational culture. Using a gender diagnosticity approach, a woman (or a man) who prefers a “masculine” organizational culture (i.e., high on aggressiveness, low on supportiveness) can be differentiated from a woman (or man) who prefers a more “feminine” organizational culture (i.e., high on supportiveness, low on aggressiveness). Importantly, this measurement approach does not rely on ratings of stereotypical traits and behaviors that are less relevant to the careers of men and women in a managerial population.

Gendered organizational culture preferences are potentially important for examining income attainment because previous research has shown that that masculinity, separate from biological sex, is associated with career success (Abele, 2003; Goldberg et al., 2004; Tharenou, 2001). For example, Kirchmeyer and Bullin (1997) found that masculine preferences were associated with higher salaries for both men and women. Fagenson (1990) reported that leaders who were higher in an organization were seen as more masculine, regardless of sex. Finally, Kent and Moss (1994) found that masculinity, more so than sex, was most correlated with leader emergence. Integrating these findings with the evidence that masculinity is characteristically associated with the stereotype of a successful manager (Ely & Meyerson, 2000; Kirchmeyer, 1998), it seems likely that employees who prefer masculine organizational cultures—irrespective of whether they are a man or a woman—are likely to receive the highest salaries.

While seemingly neutral from a societal viewpoint, the effects of gendered organizational culture preferences on income attainment may be less than benign. To the extent that society believes that women prefer work environments that are nurturing, cooperative, and affiliative, they are less likely to be seen as having the potential to succeed at the highest (i.e., masculine) levels of management (Eagly, 1987; Eagly & Karau, 2002; Eagly & Wood, 1991). Thus, women with feminine organizational culture preferences may be blocked from entering stereotypically masculine (and higher paying) career paths. However, the evidence is that men with more feminine preferences may suffer less from this stereotype (e.g., Goldberg et al., 2004; Lobel, 1994). Women with masculine organizational culture preferences face a different problem. Research shows that women who violate societal gender roles are generally disliked and interpersonally sanctioned (Eagly, Karau, & Makhijani, 1995; Heilman et al., 2004; Sools et al., Catalyst, 2007; West & Zimmerman, 1987). According to this logic, societal gender roles could potentially disadvantage all women, regardless of their organizational culture preferences, resulting in women being restricted from access to the highest paying jobs.

While the evidence reviewed thus far indicates that gender roles are important predictors of income, gender role theory does not present a clear picture of the relationships among sex, gender, and income attainment, nor does it specify how these relationships change as careers progress. Tournament theory (e.g., Rosenbaum, 1979; Lazear & Rosen, 1981; O'Reilly et al., 1988) offers additional insights into the relationships among sex, gender, and income attainment over time. According to tournament theory, careers unfold as a series of tournaments in which employees at lower levels compete with each other for career advancement. Just as in actual tournaments, performance is based both on effort and ability (O'Reilly & Chatman, 1994). Initial salary is determined by variables that index human capital and fit with the managerial stereotype, that is, ability and motivation (Noonan et al., 2005; O'Reilly & Chatman, 1994; Tharenou, 2001). Over time, those with less motivation and ability are eliminated, and the remaining participants compete for the top-level positions in the firm.

Based on this logic, masculine organizational culture preferences should be associated with higher income, at least initially, since masculinity is linked to the managerial stereotype as well as overall career success. However, for masculine women, the masculinity advantage may actually be a disadvantage as they are seen as violating both the female gender role and the successful (male) manager stereotype (Eagly et al., 1995; Ely & Meyerson, 2000; Heilman et al., 2004; Rudman & Glick, 2001; Sools et al., Catalyst, 2007). This suggests the following two hypotheses:

Hypothesis 1: Masculine organizational culture preferences will be associated with higher income attainment at Time 2 (4 years after graduation).

Hypothesis 2: Sex will moderate the relationship of masculine organizational culture preferences and income at Time 2 (4 years after graduation), such that masculine organizational culture preferences will amplify men's income attainment more than women's.

As careers continue to unfold, tournament theory suggests that gender roles will become less relevant as effort begins to account for more variance in income; that is, in subsequent rounds of the tournament, less fit contestants are likely to be eliminated and the variance in ability of those remaining will become smaller. Individuals who expend more effort or work longer hours are advantaged in this process (e.g., Becker & Huselid, 1992; Rosenbaum, 1979). Previous research on careers has shown that motivation to succeed and attain top management positions is predictive of actual career attainment (e.g., Bartol & Martin, 1987). Lobel and St. Clair (1992), for instance, found that individuals with salient career identities, regardless of sex, were willing to expend extra effort at work and received higher merit increases. Similarly, in a sample of working MBA graduates, Brett and Stroh (2003) found that those who worked the most hours also earned the most.

Tournament theory implies that high salaries are dependent on how well one plays in the tournament, irrespective of whether the player is a man or a woman. However, the effort put forth in order to compete in the tournament at later stages of careers may differ along sex and gender lines. Historically, men work more hours per week and are less likely to be engaged in intermittent and part-time work (Gabriel, 2005). There are many reasons for this. Research shows that women engage in a greater proportion of work outside the home (Brett & Stroh, 2003; Hochschild, 1989; Lobel & St. Clair, 1992), which prevents them from working longer hours on the job. Other research suggests that women are more likely to prioritize work–life balance and less likely to sacrifice this balance by working long hours. In a survey of 3700 people in Britain and Spain, Hakim (2004) reported that 55 per cent of men and 30 per cent of women were “career obsessed,” but women were more likely than men to be “family centric.” Consistent with tournament theory, Eastman (1998) found both male and female MBA graduates were willing to work more hours than others in order to get ahead, but that women favored working fewer hours then men. These findings suggest that as careers unfold, men will work more hours per week than women.

Working long hours is also associated with masculine organizational norms. In masculine organizational cultures, the ideal worker is someone for whom work is primary, time at work is unlimited, and demands of family, community, and personal life are secondary (Rapoport, Bailyn, Kolb, & Fletcher, 1996). Based on a case study of five multinational corporations, Wajcman (1998) cites working long hours as a “macho” aspect of managerial work (p. 150). Consistent with this observation, Brett and Stroh (2003) found that proportionately more women work over 61 hour per week in the financial services industry—an industry known for its “macho” culture (Antilla, 2002). They attribute this finding to a social norm of extreme work hours and suggest that a tournament process may be at play whereby employers use extreme work hours as a way of sorting out which workers are the most committed.

This logic suggests a mediating hypothesis. Over time, the tournament-like nature of careers should result in larger differences in income as individuals' work commitments become known. Though men and women characterized by masculine organizational culture preferences may be equally ambitious at earlier career stages, they may not be equally willing to sacrifice work–life balance in service of their professional goals in the long term. As careers unfold and the pressure to compete for higher paying jobs increases, qualified women may face the difficult choice of either competing in the face of subtle biases or “opting out” of the competition (Belkin, 2003; Lyness & Judiesch, 2001). The sacrifices and obstacles may result in a withdrawal from the tournament. Although we do not formally hypothesize simple effects, we expect that men will work more hours per week than women. Over time, we predict that men will earn more than women and hours worked will mediate the effects of sex on income attainment.

Hypothesis 3: Women will have lower income attainment than men at Time 3 (8 years after graduation).

Hypothesis 4: Hours worked on the job per week will mediate the effects of sex on income at Time 3 (8 years after graduation).

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Sex, Gender, and Income Attainment
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Biographical Information
  10. Biographical Information
  11. References

Research design and sample

Data for this study were collected at three time periods. The first collection (Time 1) occurred in 1986–1987 (with samples from both years) when respondents were enrolled in the first year of a 2-year full-time MBA program. All first-year students were informed of the opportunity to participate in a weekend personality and management assessment center. Participants were selected to make the sample as representative as possible of the entire MBA cohort. In general, the sample closely mirrored the larger MBA class, except that fewer foreign students participated (11 per cent vs. an average of 15 per cent) and slightly more women participated (43 per cent vs. an average of 34 per cent in the MBA class). Data collection in the first time period was done through a personality and management assessment center. Participants were assessed in groups of 12 over a weekend from Friday evening through Sunday afternoon. Eleven separate weekend assessments were conducted (Craik, Ware, Kamp, O'Reilly, Staw, & Zedeck, 2002).

While the focus of the first data collection was on extensive testing, observation, and ratings of the participants, as well as the results of their initial job search efforts, the second and third data collections were designed to assess career attainment and to document any major life changes such as marriage, divorce, health, and changes in employment status. The second data collection (Time 2) occurred in the fall of 1991. Of the original 132 participants, 105 were successfully contacted, a response rate of 79 per cent. Of these 105, 10 were either voluntarily unemployed or were employed part-time. The third data collection (Time 3) was undertaken in 1995, approximately 7–8 years after graduation. One hundred one of the original 132 participants were surveyed, a response rate of 76 per cent. Of these, 92 were employed full-time, 6 were working part-time, and 3 were unemployed.

Of the 132 original participants, 89 participants returned surveys in both the Times 2 and 3 data collections, a response rate of 67 per cent. We conducted a missing data analysis to determine if there were any differences between participants who responded to both Times 2 and 3 data collections and those who did not respond to one or more of the follow-up data collections. Results indicated that men were less likely to participate than women (t (1, 128) = 2.28, p < .05) and individuals with more prior work experience were more likely to participate than those with less prior work experience (t (1, 128) = −2.28, p < .05). There are no significant differences in organizational culture preferences of male and female non-respondents. Since non-response may be attributable to individuals working more hours per week or working in more demanding positions, our sample may represent a smaller range of incomes, thus, be a more conservative test of our hypotheses.

Dependent measures

Income attainment

In the careers literature, income is a standard objective measure of career success (e.g., Goldberg et al., 2004; Heslin, 2005). Income is particularly relevant for the study of early career attainment because it tends to vary more across individuals than other measures of career attainment such as promotions (Ng, Eby, Sorensen, & Feldman, 2005). In this study, respondents indicated their income at Time 2, 4 years after the initial data collection, and again in response to follow-up surveys conducted 4 years later (Time 3). This permitted us to examine income attainment across two time periods. Income consisted of gross annual salary including bonuses, when relevant. Natural logs were taken of all income measures to reduce the effects of skewness.

Hours worked per week

Participants reported the number of hours they worked on the job per week at Time 2 and again at Time 3.

Independent measures

Sex

Sex was coded at Time 1 based on the subject's participation in the management assessment center.

Organizational culture preferences

To assess organizational culture preferences, we administered the organizational culture profile (OCP) as part of the managerial assessment center at Time 1. The OCP is Q-sort instrument that contains 54 values that can be used to characterize an organization or person's value preferences (O'Reilly et al., 1991). Participants were asked, “How important is it for this characteristic to be part of the organization you work for?” Subsequently, they sorted the values into nine categories ranging from most to least desirable according to the following distribution: 2–4–6–9–12–9–6–4–2.

Control variables

A number of individual, job, and organizational variables may affect career outcome variables and need to be accounted for before examining the effects of sex and gender on career attainment. For purposes of this study, we controlled as covariates the following human capital variables: Date of graduation, age, prior years of relevant work experience, number of children, and underrepresented minority status (white and Asian vs. other). In addition to controlling for individual factors that may affect career outcomes, we included dummy variables for investment banking and education, two industries that differ both in terms of pay and gender type (Ragins & Sundstrom, 1989). Age and number of children did not explain any unique variance in either income at Time 2 or the rate of income increase from Time 2 to Time 3, thus, these variables were dropped from the final model.

Analyses

Assessing masculinity and femininity

The gender diagnosticity approach to assessing masculinity and femininity (Lippa & Connelly, 1990; Lippa, 1991, 2001) is based on the Bayesian probability that an individual is predicted to be male or female on the basis of a set of gender-related indicators. A masculine person is an individual who shows “male-like” preferences in comparison to the reference group (other MBAs) and a feminine person is an individual who shows “female-like” preferences. We classified subjects as feminine or masculine based on organizational culture preferences.

Consistent with previous research investigating sex differences in preferences for occupational interests (e.g., Konrad, Ritchie, Lieb, & Corrigall, 2000; Lippa, 1991, 1998), we expected sex differences in the organizational culture dimensions of aggressiveness and supportiveness. We first created indices for supportiveness and aggressiveness by aggregating OCP items representing these dimensions (O'Reilly et al., 1991). Previous research with large sample sizes has reported good reliabilities for the aggressiveness (α = .75) and supportiveness (α = .87) scales (Sarros, Gray, Densten, & Cooper, 2005). The supportiveness index contained the values of “sharing information freely,” “being supportive,” and “respect for people”; the aggressiveness index contained the values “being aggressive,” “being competitive,” and “being socially responsible” (reverse coded). A t-test for sex differences on these factors revealed that, on average, men (M = 18.8; SD = 5.33) valued aggressiveness in organizations more than women did (M = 16.2; SD = 4.6) (t (1, 128) = −2.90, p < .01) and women (M = 25.4; SD = 3.21) valued supportiveness in organizations more than men did (M = 23.3; SD = 3.53) (t (1, 128) = 3.52, p < .01). There were no significant sex differences in our final sample on any other culture dimensions.

To calculate masculine–feminine organizational culture preferences, we conducted a discriminant analysis with aggressiveness and supportiveness as the independent variables. Individual discriminant scores represent the masculinity or femininity of a person's organizational culture preferences, with higher scores representing more masculine preferences and lower scores representing more feminine preferences. The overall discriminant function was significant (Wilk's λ = .90; χ2 = 13.8; p < .00). Wilk's λ was significant according to the F test for both aggressiveness (Wilk's λ = .94; F (1, 128) = 8.38, p < .01) and supportiveness (Wilk's λ = .91; F (1, 128) = 12.4, p < .01). These variables in the discriminant function explained a third of the variation in the data (Rmath image = .32).

To test the convergent validity of our gender scheme, we conducted supplemental analyses using trained rater observations of masculinity and femininity and self-report data on masculinity–femininity. These analyses reveal that our approach to measuring gender based on sex differences in organizational culture values is consistent with both behavioral and trait measures of gender. In comparison to participants with more feminine culture preferences, participants with more masculine organizational culture preferences were rated higher in masculinity (r = .25, p < .01) and lower in femininity (r = −.19, p < .05) by trained raters over the course of a 2½ day assessment center (Craik et al., 2002). On the personality attributes questionnaire (Spence & Helmreich, 1978), participants with masculine preferences rated themselves higher in masculinity (r = .41, p < .00) and lower in femininity (r = −.45, p < .00) than participants with feminine preferences. Interestingly, when we added these alternative measures of gender to the model, they showed no relationship to income. Only organizational culture preferences remained significant in the equation. One reason may be that our measure of masculinity taps into a domain-specific aspect of gender that directly relates to career choices. Thus, while a gender diagnosticity approach based on organizational culture preferences clearly overlaps with behavioral and trait measures of gender, it also reveals unique information that may be particularly relevant for predicting early career income attainment.

Hypothesis testing

To model the relations among sex, gendered organizational culture preferences, and income across time, and to examine the mediating effect of effort, we used hierarchical linear modeling (HLM version 6.0; Bryk, Raudenbush, & Congdon, 2005). HLM allows the analysis of variables at multiple levels of analysis in a series of regression equations. In the current study, the Level 1 regression included a within-person variable representing income assessed at Times 2 and 3, 4 and 8 years, respectively, after the initial Time 1 data collection. The Level 2 regression included between-person variables representing biological sex (female or male), the discriminant scores reflecting each person's organizational culture preferences, and the interaction of sex and organizational culture preferences. In addition, we included the following between-person control variables: Age, work experience, graduation year, race, and dummy measures for the education industry and investment banking careers. Thus, income across repeated measures was nested within Level 2 between-person variables.

Two analytic steps were undertaken to test main effects (Hypotheses 1 and 3) and moderation (Hypothesis 2) in the current study. We first analyzed repeated measures of income at Level 1 using

  • equation image(1)

We assume that Yti, the observed status at time t for person i and π0i is the income of person i at wti = 0 (4 years after graduation). πpi is the growth rate for person i over the data collection period. Each person is observed on Ti occasions. We assume a simple error structure eti that is independently and normally distributed with a mean of 0 and a constant variance.

An important aspect of Equation (1) is the assumption that the growth parameters vary across individuals. Thus, the second analytic step was to formulate a Level 2 model to represent the change parameters across individuals. Both the intercept and growth-rate parameters are allowed to vary at Level 2 as a function of measured person characteristics. Thus, Equation (2) becomes

  • equation image(2)

Similar to the parameters estimated in OLS regression, the parameter estimates β0q and β1q indicate the relative contribution of the Level 2 control variables (e.g., work experience, investment banking, race) and Level 2 hypothesis-testing variables (i.e., sex, organizational culture preferences, sex*preferences) to the intercept and slope estimates. The logic of analyzing Level 1 coefficients either as intercepts or growth rate parameters is that the Level 2 model describes between-person variables predicting income at Time 2 (π0i), and the change in income from Time 2 to Time 3 (π1i). Factors predicting income at Time 2 are tested by the significance of β0q coefficients. Factors influencing the change in income from Time 2 to Time 3 are tested by the significance of the β1q coefficients. Level 2 variables were grand mean centered.

As with main effects and moderation hypotheses, two analytic steps were undertaken to test mediation (Hypothesis 4) in the current study. We first analyzed repeated measures of income at Level 1 using

  • equation image(3)

We assume that Yti, the observed status at time t for person i and π0i is the income of person i at wti = 0 (4 years after graduation). π1i is the growth rate for person i over the data collection period. π2i is a time varying predictor representing hours worked per week for person i over the data collection period. Each person is observed on Ti occasions.

As with the previous analyses, the second analytic step was to formulate a Level 2 model to represent the change parameters across individuals. Both the intercept and growth-rate parameters are allowed to vary at Level 2 as a function of measured person characteristics. Thus, Equation (2) becomes

  • equation image(4)

Level 2 variables were grand mean centered.

To test for mediation, we followed the steps outlined in Krull and MacKinnon (2001) for multilevel mediational analysis when the mediator variable is a within-person (i.e., Level 1) variable. First, we conducted Level 1 regressions with hours worked per week as a Level 1 predictor of the group intercept in the Level 2 equation. An additional 10 per cent of the variation in income was explained by hours worked per week, an effect that was significantly different from 0 (χ2[1] = 10.4, p < .01). We then compared the results to the parameter estimates without hours per week as a Level 1 predictor. Finally, we regressed the Level 2 predictor variables on hours worked per week.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Sex, Gender, and Income Attainment
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Biographical Information
  10. Biographical Information
  11. References

Table 1 presents the means, standard deviations, and bivariate correlations among the variables. Results confirm that the respondents are well into their careers, with an average annual income at Time 2 of $121 000 (range from $15 000 to $2 million). Respondents were working an average of 50 hours per week (range 17–100). Masculine organizational culture preferences were related to hours worked per week at Time 2. Sex was related to work experience, hours worked per week, both at Times 2 and 3, and income (Time 3). Interestingly, there were no significant relationships between participants' organizational culture preferences and the industry in which they worked.

Table 1. Means, standard deviations, and correlations among variables
 Mean (SD)12345678910
  1. a

    *p < .05;

  2. b

    **p < .01;

  3. c

    ***p < .001.

 1. Relevant work experience0.90 (0.30)          
 2. Education industry0.11 (0.32)−0.59***         
 3. Investment banker0.11 (0.32)0.12−0.13        
 4. Graduation year0.60 (0.49)−0.13−0.000.00       
 5. Nonwhite0.12 (0.33)−0.100.080.08−0.11      
 6. Male0.51 (0.50)0.27**−0.29**0.070.100.10     
 7. Masculine preferences −0.01 (1.01)0.08−0.11−0.020.090.000.29**    
 8. Log income (Time 2)11.0 (0.41)0.12−0.28**0.53***−0.10−0.160.110.02   
 9. Log income (Time 3)11.6 (0.72)0.22*−0.25*0.63***0.02−0.130.24**0.020.71***  
10. Work hours (Time 2)49.3 (9.48)0.16−0.23*0.38***0.180.020.31**0.23**0.44***0.47*** 
11. Work hours (Time 3)49.6 (11.7)0.09−0.120.23*0.070.060.36***0.150.200.38***0.31**

Partitioning of variance components

Before proceeding to test the linkages in the hypothesized model, we investigated whether systematic within- and between-person variance existed in the criterion variables by estimating a null model for each variable where the within-person variance represents the change in income across time. The null model partitions the total variance of the dependent variable into within- and between-individual components, and the intercept for each null model represents the average level of that variable across individuals. If no within-individual variance exists in the criterion variable, then HLM is not appropriate (i.e., there is only one level of analysis). The null model results indicated that there was significant between-individual variance in income (χ2 = 158.4, p < .001), and that a substantial portion of the variance was within individuals. By dividing the within-person variance over the sum of the within-person (ρ2 = 0.11) and between-person variance (τ00 = .29), we concluded that a substantial proportion of the total variance was within individuals (27 per cent), thus, there was significant change in income across time.

Hypothesis tests—Time 2

The effects of sex, gender, and the interaction of sex and gender at Time 2 (4 years after graduation) are shown by the Level 1 (π0i) equation in Table 2. Parameter estimates representing control variables reveal that income at Time 2 was negatively related to work experience (β01) and jobs in the education industry (β02).

Table 2. Hierarchical linear modeling results of income attainment
VariableParameter estimateSEt
  • Note: R2 = variance explained by π0i and π1i (including male, masculinity, and the interaction term). ΔR2 = change in variance explained with the addition of male, masculinity, and the interaction terms to π0i and π1i. Proportions were computed as the proportional reduction in the Level 1 variance component of income.

  • a

    Individual-level predictors of income 4 years after graduation.

  • b

    Individual-level predictors of change in income from 4 to 8 years after graduation.

  • *

    p = .05;

  • **

    p = .01.

Level 1 π0ia
 Intercept (β00)10.640.10108.5**
 Work experience (β01)−0.550.22−2.49*
 Education (β02)−0.610.23−2.62**
 Investment banking (β03)−0.030.37−0.08
 Graduation year 1987 (β04)−0.190.13−1.50
 Nonwhite (β05)−0.160.13−1.22
 Male (β06)−0.150.12−1.25
 Masculinity (β07)0.180.072.55**
 Male*masculinity (β08)−0.220.11−1.97*
Level 2 π1ib
 Intercept (β10)0.080.024.39**
 Work experience (β11)0.100.033.06*
 Education (β12)0.060.041.75
 Investment banking (β13)0.180.062.92**
 Graduation year 1987 (β14)0.030.021.41
 Nonwhite (β15)−0.020.03−0.58
 Male (β16)0.050.022.36*
 Masculinity (β17)−0.010.01−1.15
 Male*masculinity (β18)0.010.020.51
R2.54  
ΔR2.05**  

Hypothesis 1 predicted that more masculine culture preferences would be associated with higher income at earlier career stages. The β07 coefficient reveals that, on average, increases in masculine organizational culture preferences lead to higher income at Time 2 (β07 = 0.18, t (8, 81) = 2.55, p < .01). All else being equal, a one-unit increase in masculine organizational culture preferences corresponded to an $8238 increase in salary. While this only corresponds to a 1 per cent increase in the proportion of variance explained, a post-hoc test of the effect revealed that β07 was indeed significantly different from 0 (χ2[1] = 6.50, p < .01).

Hypothesis 2 predicted an interaction between sex and masculinity at earlier career stages. Specifically, we expected that sex would moderate the effect of masculine organizational culture preferences on income at Time 2, with masculine preferences having a stronger effect on men's income. As shown in Table 2, the β08 coefficient for the full interaction is significant (β08 = −0.22, t (8, 81) = −1.97, p < .05). The addition of this variable increased the proportion of variance explained by 3 per cent and a post-hoc test of the effect indicated that β08 was significantly different from 0 (χ2[1] = 3.87, p < .05). Unexpectedly, and contrary to our hypothesis, masculine organizational culture preferences initially appear to have a more positive effect on women's income than on men's. While masculine organizational culture preferences had no effect on men's income (t (8, 81) = −0.54, ns), women whose preferences for masculine organizational cultures were one standard deviation above the mean earned $8028 more than women whose preferences were closer to the mean (t (8, 81) = 2.55, p < .01). Taking into consideration the unexpected form of the interaction, Hypothesis 2 was not supported.

Although Hypotheses 3 and 4 predicted relationships at later career periods, it is worth noting that there were no sex differences in income attainment at Time 2, as evidenced by the non-significant coefficient (β06), and hours worked per week did not mediate the effect of masculinity and the interaction of sex and masculinity at Time 2. Although the addition of hours worked per week as a within-person variable predicting income did reduce the effects of masculine organizational culture preferences (β06) and the interaction of sex and masculine organizational culture preferences (β07) to non-significance in the Level 1 (π0i) equation, after taking into account differences in hours worked per week due to the organizational culture preferences and the interaction of sex and organizational culture preferences, the effect of hours worked per week on income was actually negative (β26) and the interaction (β27) was non-significant. A repeated measures analysis of Level 2 variables regressed on hours worked per week revealed that masculinity did not have an independent effect on hours worked per week (β06 = −1.38, t (8, 81) = −0.98, ns) and while the interaction did have a significant effect on hours worked per week (β07 = 3.63, t (8, 81) = 2.07, p< .04), the pattern of results did not support our theory. Women with masculine organizational culture preferences worked an average of 9 hour less than men with masculine organizational culture preferences, while women with feminine organizational culture preferences worked only an average of 6 hours less than men with masculine organizational culture preferences (F (3, 86) = 4.89, p < .01).

Hypothesis tests—Time 3

The effects of sex, gender, and the interaction of sex and gender on change in income from Time 2 (4 years after graduation) to Time 3 (8 years after graduation) are shown by the Level 2 (π1i) equation in Table 2. Examination of the Level 2 (π1i) control variables reveals that the rate of change in income from Time 2 to Time 3 was significantly related to work experience (β11) and investment banking careers (β13).

Hypotheses 1 and 2 predicted that the effects of sex and the interaction of sex and gender on income would be strongest at earlier career stages. As Table 2 illustrates, the significant effects of masculinity and the interaction of sex and masculinity at Time 2 had disappeared by Time 3, as evidenced by the non-significant coefficient β17 and β18. This pattern of results supports our theory, since congruence with the masculine managerial gender role appears to lead to higher income initially, but its importance lessens as careers unfold.

Hypothesis 3 predicted a sex difference in income at Time 3. In support of this hypothesis, Table 2 shows that the rate of change for men at Time 3 is significantly steeper than the rate of change for women (β16 = 0.05, t (8, 81) = 2.34, p < .02), as indicated by the β16 coefficient. This difference amounts to an additional $1337 income increase for men above the average rate increase of $3479. Although the change in the proportion of variance explained was only 1 per cent, a post-hoc test of the effect revealed that β16 was significantly different from 0 (χ2[1] = 5.56, p < .02). As the importance of sex increases later in careers, this pattern of results also supports our theory.

Hypothesis 4 predicted that hours worked per week would mediate the relationship of sex and income at Time 3. As shown in Table 2, the coefficient β16 representing the effect of sex on change in income from Time 2 to Time 3 was significant. As Table 3 shows, when we included hours worked per week as a within-person variable predicting income, the direct effect of sex (β16) was no longer significant (t (8, 81) = 1.30, ns). A repeated measures analysis of Level 2 variables regressed on hours worked per week reveals that men increased the number of hours worked per week from Time 2 to Time 3 at a significantly higher rate than women (β16 = 0.92, t (8, 81) = 3.25, p < .01). Particularly striking is the fact that masculine men worked an average of 12 more hours per week than masculine women (F (3, 86) = 6.55, p < .001). As shown in Table 3, taking into account sex differences in hours worked per week, the effect of hours worked per week on income remained significant (β26 = 0.02, t (8, 81) = 2.28, p < .02). A Sobel test using the critical values recommended by MacKinnon, Lockwood, Hoffman, West, and Sheets (2002) revealed that the indirect effect of sex on income was indeed significant (z′ = 1.14, p < .05). Overall, this pattern of results supports Hypothesis 4. Consistent with tournament theory, hours worked per week became more important than gender roles in explaining income differences at the later period of individuals' careers.

Table 3. Hierarchical linear modeling results of income attainment given hours worked per week
VariableParameter estimateSEt
  • Note: R2 = variance in income explained by π0i, π1i, and π2i. ΔR2 = change in variance explained with the addition of Level-2 variables predicting π2i. Proportions were computed as the proportional reduction in the Level 1 variance component of income.

  • a

    Individual-level predictors of income 4 years after graduation.

  • b

    Individual-level predictors of change in income from 4 to 8 years after graduation.

  • c

    Effect of hours worked per week on income, taking into account individual-level differences in hours worked per week.

  • *

    p = .05;

  • **

    p = .01.

Level 1 π0ia
 Intercept (β00)10.630.1199.7**
 Work experience (β01)−0.650.19−3.44**
 Education (β02)−0.880.20−4.48
 Investment banking (β03)−0.740.34−2.16*
 Graduation year 1987 (β04)−0.170.15−1.17
 Nonwhite (β05)−0.150.19−0.78
 Male (β06)−0.060.15−0.41
 Masculinity (β07)0.140.081.77
 Male*masculinity (β08)−0.180.14−1.28
Level 2 π1ib
 Intercept (β10)0.080.023.98**
 Work experience (β11)0.110.033.22**
 Education (β12)0.090.034.04**
 Investment banking (β13)0.220.063.88**
 Graduation year 1987 (β14)0.040.021.62
 Nonwhite (β15)−0.010.03−0.42
 Male (β16)0.030.031.30
 Masculinity (β17)−0.010.01−0.83
 Male*masculinity (β18)0.000.02−0.06
Level 2 π2ic
 Intercept (β20)0.010.002.37*
 Work experience (β21)−0.010.001.52
 Education (β22)−0.020.00−1.90
 Investment banking (β23)0.040.022.65**
 Graduation year 1987 (β24)−0.010.02−1.68
 Nonwhite (β25)−0.030.01−4.70**
 Male (β26)0.020.012.28*
 Masculinity (β27)−0.010.00−2.99**
 Male*masculinity (β28)0.000.01−0.05
R2.74  
ΔR2.10**  

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Sex, Gender, and Income Attainment
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Biographical Information
  10. Biographical Information
  11. References

Summary of results

Masculinity, indexed by preference for a masculine as opposed to feminine organizational culture, had an initial positive impact on income attainment for both men and women; that is, consistent with some previous research (Fagenson, 1990; Goldberg et al., 2004; Kirchmeyer, 1998; Wong, Kettleman, & Sproule, 1985; Tharenou, 2001; van Vianen & Fischer, 2002), gender role mattered more for early career success than sex. This effect may reflect the masculine bias associated with the successful managerial stereotype (Eagly & Karau, 2002), but does not suggest any overt discrimination against women. Indeed, biological sex had no effect on income attainment initially, suggesting that it was not a significant determinant of early career success. It did have an effect at later periods, but this was mediated by the number of hours worked. As shown in our sample and in earlier research, women work fewer hours as their careers progress. Since previous research has found effort to be positively related to earnings (Lobel & St. Clair, 1992; Stroh et al., 1992; Tharenou, 2001), this result is consistent with tournament theory (Noonan et al., 2005; O'Reilly & Chatman, 1994; O'Reilly et al., 1988) in that, over time, effort matters more than gender role congruence.

Contrary to previous research, the interaction of sex and gender role did not provide an advantage to masculine men or a disadvantage to masculine women. Instead, it initially appeared to work in favor of women with a strong preference for masculine organizational cultures. This could be result of the small number of women who conform to the masculine managerial stereotype or the distinctiveness of this group (Chatman et al., 2008; Taylor & Fiske, 1975). Nevertheless, this finding runs counter to the backlash effect against agentic women found in earlier studies (Eagly et al., 1995; Ely & Meyerson, 2000; Heilman et al., 2004; Rudman & Glick, 2001; Sools et al., Catalyst, 2007). Given the fact that this effect on income attainment declined over time, it may be role incongruence helps women succeed at lower levels of the tournament but disadvantages them as the competition intensifies at higher levels.

Our results highlight the importance of empirically disentangling sex and gender and of studying the impact of sex and gender on income attainment longitudinally. Studies that fail to explore these joint effects run the risk of conflating the separable effects—and possibly overstating the effects of biological sex. Our approach to disentangling sex and gender is unique in that we were able to distinguish masculine and feminine organizational culture preferences by assessing how “male-like” and “female-like” individuals were in comparison to other men and women in the sample. Interestingly, a traditional measure of gender roles did not account for variance in income attainment beyond the gender diagnosticity approach. Although our approach overlapped with other measures of gender identity, gendered organizational culture preferences revealed an aspect of societal gender roles that uniquely relates to income attainment. Hierarchical linear analyses allowed us to examine the independent and interactive effects of sex and gender over time by modeling both within- and between-person variation in the same equation. In addition, we were able to control for many individual and structural factors that might also affect income, including previous work experience and industry, resulting in more rigorous tests of our hypotheses.

Theoretical implications

By demonstrating the effects of sex and gender on income attainment longitudinally, this study bridged two distinct theoretical perspectives on the male–female gap in income attainment, namely, gender role theory and tournament theory. Bridging these perspectives allowed us to extend current theory as well as clarify some of the ambiguities in previous research on women's career attainment. While gender roles can be an important determinant of initial career success, gender role theory alone appears insufficient for explaining the male–female gap in earnings. Our results suggests that career success, as indexed by income attainment, has tournament-like properties and that effort (indexed here by the number of hours worked) is a more significant determinant of success later in careers. Given this finding, future research might focus more on those factors that determine career effort than crude measures of human capital. For example, studies might examine how more subtle or unconscious biases around gender role incongruency occur, how these affect career salience and motivation to succeed at the top levels of competition, and how organizations can reduce career competition based on effort, since it appears to disproportionately impact women.

The fact that women earn less over time—despite the initial success of women who conform to the masculine managerial gender role—has two interpretations, with divergent implications for theory. One picture to emerge is one of individuals making trade-offs between flexibility and income attainment (e.g., Brett & Stroh, 2003; Burke, 1999; Judiesch & Lyness, 1999; Rothbard, Phillips, & Dumas, 2005). According to this interpretation, women who initially prefer masculine organizational cultures make a trade-off between career success and work-life balance that results in some simply “opting out” of the career tournament (Belkin, 2003; Brett & Stroh, 2003; Chusmir & Parker, 1991; Hakim, 2004; Schneer & Reitman, 1993). Another possible interpretation is that masculine women are being penalized for succeeding in male-dominated jobs, causing them to withdraw or exert less effort as the tournament continues (e.g., Heilman et al., 2004; Rudman & Glick, 1999, 2001; Sools et al., Catalyst, 2007). These two interpretations suggest different underlying mechanisms: On the one hand, a personal choice not to compete in the tournament and, on the other hand, systematic gender bias and discrimination. In reality, both may be happening simultaneously, with educated women facing the difficult choice of either competing in the face of subtle biases or “opting out” when the pressure to compete conflicts with other life goals.

Managerial implications

Shattering the glass ceiling is a complicated problem. The dearth of women at the highest paying levels of management is well documented (e.g., Blackburn et al., 2000; Catalyst, Catalyst, 2007). Organizations that invest in and reward top female talent face a significant financial loss when women decide to “opt out” of the tournament. Research has shown that organizations that have more women in senior management are those that are also able to retain younger managerial women (Goodman, Fields, & Blum, 2003), suggesting that a failure to nurture younger women professionals can have long-term consequences. To be successful, organizations must address the root of the problem: Time is not unlimited and most workers (men as well as women) have commitments to their family, community, or personal life (Rapoport et al., 1996). Although tournament processes may seem gender-neutral, the likelihood is that they disadvantage women, who shoulder the disproportionate share household and childcare responsibilities. A tournament system of careers, while providing an efficient mechanism for sorting out the most committed workers (Lazear & Rosen, 1981), can also lead to other adverse consequences such as reduced productivity and strain on worker mental and physical health (Ng & Feldman, 2008; Sparks, Cooper, Fried, & Shirom, 1997). This suggests that if organizations want to construct a level playing field in which women are not disadvantaged by the requirement to work long hours and employees are able to thrive mentally and physically, organizational policies and culture need to emphasize career flexibility over long hours (Hooks, 1996); that is, the rules of the tournament should be constructed in a way that does not privilege excessive effort. At the same time, educational institutions should encourage students to examine how their values may change over the course of their careers, and whether this change in values should influence their early career decisions.

Limitations

Although the results presented here shed some light on the conflicting findings from previous studies, several important limitations are worth noting. First, our sample is comparatively small, restricted to MBA graduates from a single institution, and included people tracked for only 8 years. There may be aspects of the sample (e.g., the geographic location of the school, the nature of students enrolled, the time period studied) that limit the generalizability of the findings. Second, some of our variables (e.g., hours worked per week) are self-reported and, thus, may be impacted by social desirability bias. Third, although an attempt was made to control for a set of individual and organizational variables that can affect careers, there are clearly other factors that could be controlled. However, given the large number of such potential influences, it seems unlikely that any single study will provide definitive answers.

A final limitation may provide a fruitful area for future research. Although we used organizational culture values to differentiate masculine and feminine preferences, we did not assess actual fit with the organizations or industries in which participants worked. Since fit is generally associated with more positive outcomes for employees (i.e., Kristof-Brown, Zimmerman, & Johnson, 2005), we would expect higher incomes when there was a better fit between an individual's culture preferences and the culture of the organizations or industries in which they worked. Interestingly, we found no association between organizational culture preferences and the industries in which participants worked immediately after graduation. However, supplemental analyses revealed interactions between organizational culture preferences and industry whereby incongruence between organizational culture preferences and sex-type of the industry appeared to lead to higher incomes, at least initially. These results reveal both the complexity of understanding career dynamics as well as the importance of research that can disentangle the effects of sex, individual gender roles, industry and job effects, and career stage (e.g., Goldberg et al., 2004; Judge & Livingston, 2008).

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Sex, Gender, and Income Attainment
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Biographical Information
  10. Biographical Information
  11. References

This study demonstrated the importance of considering the independent and interactive effects of sex and gender longitudinally when trying to understand income attainment. While it is certainly true that bias may exist within organizations and that it can work against women, we find that congruence with masculine gender roles ultimately matters less than the willingness and ability to work long hours per week. This raises the intriguing possibility that success may result not from fitting masculine leadership stereotypes or gender role congruence, but from compliance with tournament-like expectations of working long hours, which require difficult trade-offs to be made. This conclusion does not rule out more subtle biases stemming from gender role incongruence. Biases could account for the fewer number of hours worked by women in the study when compared to men. In either case, it appears that hours worked is a central causal variable affecting income attainment. Insofar as careers are defined narrowly in terms of income attainment, any person—male or female—who chooses to put in fewer hours is likely to be disadvantaged. In our view, this does not constitute a “failure,” but rather a choice not to play that particular version of the career tournament.

Biographical Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Sex, Gender, and Income Attainment
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Biographical Information
  10. Biographical Information
  11. References

Olivia (Mandy) O'Neill is a Visiting Assistant Professor of Management at The Wharton School, University of Pennsylvania. She received her PhD in Organizational Behavior from the Stanford University Graduate School of Business. Her interests include organizational culture, gender, emotions, and change.

Biographical Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Sex, Gender, and Income Attainment
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Biographical Information
  10. Biographical Information
  11. References

Charles O'Reilly is the Frank E. Buck Professor of Management at the Graduate School of Business at Stanford University. His research interests include organizational demography, leadership, culture, and change.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Sex, Gender, and Income Attainment
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Biographical Information
  10. Biographical Information
  11. References
  • Abele, A. E. (2003). The dynamics of masculine-agentic and feminine-communal traits: Findings from a prospective study. Journal of Personality and Social Psychology, 85, 768776.
  • Acker, J. (1990). Hierarchies, jobs, bodies: A theory of gendered organizations. Gender & Society, 4, 139158.
  • Antilla, S. (2002). Tales from the boom-boom room: Women vs. Wall Street. New York: HarperBusiness.
  • Baril, G. L., Elbert, N., Mahar-Potter, S., & Reavy, G. C. (1989). Are androgynous managers really more effective? Group & Organization Studies, 14, 234249.
  • Bartol, K. M., & Martin, D. C. (1987). Managerial motivation among MBA students: A longitudinal assessment. Journal of Occupational Psychology, 60, 112.
  • Becker, B. E., & Huselid, M. A. (1992). The incentive effects of tournament compensation systems. Administrative Science Quarterly, 37, 336350.
  • Belkin, L. (2003). The opt-out revolution. The New York Times, October 26.
  • Bem, S. L. (1974). The measurement of psychological androgyny. Journal of Counseling and Clinical Psychology, 42, 155162.
  • Blackburn, R. M., Jarman, J., & Brooks, B. (2000). The puzzle of gender segregation and inequality: A cross-national analysis. European Sociological Review, 16, 119135.
  • Blau, F. D., & Kahn, L. M. (2006). The U.S. gender pay gap in the 1990s: Slowing convergence. Industrial and Labor Relations Review, 60, 4566.
  • Brett, J. M., & Stroh, L. K. (2003). Working 61 plus hours a week: Why do managers do it? Journal of Applied Psychology, 88, 6778.
  • Browne, K. R. (2006). Evolved sex differences and occupational segregation. Journal of Organizational Behavior, 27, 143162.
  • Bryk, A. S., Raudenbush, S. W., & Congdon, R. T. (2005). HLM. Hierarchical linear and nonlinear modelling with the HLM6 programs. Chicago, IL: Scientific Software International.
  • Burke, R. J. (1999). Workaholism in organizations: Gender differences. Sex Roles, 41, 333345.
  • Catalyst. (2007). Women's earnings and income.
  • Chatman, J. A., Boisnier, A. D., Spataro, S. E., Anderson, C., & Berdahl, J. L. (2008). Being distinctive versus being conspicuous: The effects of numeric status and sex-stereotyped tasks on individual performance in groups. Organizational Behavior and Human Decision Processes, 107, 141160.
  • Chusmir, L. H., & Parker, B. (1991). Gender and situational differences in managers' values: A look at work and home lives. Journal of Business Research, 23, 325336.
  • Conyon, M. J., Peck, S. I., & Sadler, G. V. (2001). Corporate tournaments and executive compensation: Evidence from the U.K. Strategic Management Journal, 22, 805815.
  • Craik, K. H., Ware, A. P., Kamp, J., O'Reilly, C. A., Staw, B., & Zedeck, S. (2002). Explorations of construct validity in a combined managerial and personality assessment program. Journal of Occupational and Organizational Psychology, 75, 171193.
  • Dimitrovsky, L., Singer, J., & Yinon, Y. (1989). Masculine and feminine traits: Their relation to suitedness in training for traditionally masculine and feminine army functions. Journal of Personality and Social Psychology, 57, 839847.
  • Eagly, A. H. (1987). Sex differences in social behavior: A social-role interpretation. Hillsdale, NJ: Erlbaum.
  • Eagly, A. H., & Karau, S. J. (2002). Role congruity theory of prejudice toward female leaders. Psychological Review, 109, 573598.
  • Eagly, A. H., & Wood, W. (1991). Explaining sex differences in social behavior: A meta-analytic perspective. Personality and Social Psychological Bulletin, 17, 306315.
  • Eagly, A. H., Karau, S., & Makhijani, M. (1995). Gender and the effectiveness of leaders: A meta-analysis. Psychological Bulletin, 117, 125145.
  • Eastman, W. (1998). Working for position: Women, men, and managerial work hours. Industrial Relations, 37, 5166.
  • Ely, R. J., & Meyerson, D. E. (2000). Theories of gender in organizations: A new approach to organizational analysis and change. Research in Organizational Behavior, 22, 105153.
  • Fagenson, E. A. (1990). Perceived masculine and feminine attributes examined as a function of individuals' sex and level in the organizational power hierarchy: A test of four theoretical perspectives. Journal of Applied Psychology, 75, 204211.
  • Fiske, S. T., Cuddy, A. J., Glick, P., & Xu, J. (2002). A model of (often mixed) stereotype content: Competence and warmth respectively follow from perceived status and competition. Journal of Personality and Social Psychology, 82, 878902.
  • Gabriel, P. (2005). The effects of differences in year-round, full-time labor market experience on gender wage levels in the United States. International Review of Applied Economics, 19, 369377.
  • Goldberg, C. B., Finkelstein, L. M., Perry, E. L., & Konrad, A. M. (2004). Job and industry fit: The effects of age and gender matches on career progress outcomes. Journal of Organizational Behavior, 25, 807829.
  • Goodman, J. S., Fields, D. L., & Blum, T. C. (2003). Cracks in the glass ceiling: In what kinds of organizations do women make it to the top? Group & Organization Management, 28, 475501.
  • Hakim, C. (2004). Choosing to be different: Women, work, and the family. London: Centre for Policy Studies.
  • Heilman, M. E., Wallen, A. S., Fuchs, D., & Tamkins, M. M. (2004). Penalties for success: Reactions to women who succeed at male gendered-typed tasks. Journal of Applied Psychology, 89, 416427.
  • Helgeson, V. S. (1994). Relation of agency and communion to well-being: Evidence and potential explanations. Psychological Bulletin, 116, 412428.
  • Heslin, P. A. (2005). Conceptualizing and evaluating career success. Journal of Organizational Behavior, 26, 113136.
  • Hochschild, A. R. (1989). The second shift. New York: Penguin.
  • Hooks, K. L. (1996). Diversity, family issues and the big 6. Journal of Accountancy, 182, 5158.
  • Hyde, J. S. (2005). The gender similarity hypothesis. American Psychologist, 60, 581592.
  • Johnson, S. K., Murphy, S. E., Zewdie, S., & Reichard, R. J. (2008). The strong sensitive type: Effects of gender stereotypes and leadership prototypes on the evaluation of male and female leaders. Organizational Behavior and Human Decision Processes, 106, 3960.
  • Judge, T. A., & Livingston, B. A. (2008). Is the gap more than gender? A longitudinal analysis of gender, gender role orientation, and earnings. Journal of Applied Psychology, 93, 9941012.
  • Judiesch, M. K., & Lyness, K. S. (1999). Left behind? The impact of leaves of absence on managers' career success. Academy of Management Journal, 42, 641651.
  • Kent, R. L., & Moss, S. E. (1994). Effects of sex and gender roles on leader emergence. Academy of Management Journal, 37, 13351346.
  • Kirchmeyer, C. (1998). Determinants of managerial career success: Evidence and explanation of male/female differences. Journal of Management, 24, 673692.
  • Kirchmeyer, C. (2002). Change and stability in manager's gender roles. Journal of Applied Psychology, 87, 929939.
  • Kirchmeyer, C., & Bullin, C. (1997). Gender roles in a traditionally female occupation: A study of emergency, operating, intensive care and psychiatric nurses. Journal of Vocational Behavior, 50, 7895.
  • Konrad, A., & Cannings, C. (1997). The effects of gender role congruence and statistical discrimination on managerial advancement. Human Relations, 50, 13051328.
  • Konrad, A. M., Ritchie, J. E., Lieb, P., & Corrigall, E. (2000). Sex differences and similarities in job attribute preferences: A meta-analysis. Psychological Bulletin, 126, 593641.
  • Kristof-Brown, A. L., Zimmerman, R. D., & Johnson, E. C. (2005). Consequences of individuals' fit at work: A meta-analysis of person-job, person-organization, person-group, and person-supervisor fit. Personnel Psychology, 58, 281323.
  • Krull, J. L., & MacKinnon, D. P. (2001). Multilevel modeling of individual and group level mediated effects. Multivariate Behavioral Research, 36, 249277.
  • Lam, S. S., & Dreher, G. E. (2004). Gender, extra-firm mobility, and compensation attainment in the United States and Hong Kong. Journal of Organizational Behavior, 25, 791805.
  • Lazear, E., & Rosen, S. (1981). Rank-order tournaments as optimum labor contracts. Journal of Political Economy, 89, 841864.
  • Lippa, R. A. (1991). Some psychometric characteristics of gender diagnosticity measures: Reliability, validity, consistency across domains, and relationship to the Big Five. Journal of Personality & Social Psychology, 61, 10001011.
  • Lippa, R. A. (1998). Gender-related individual differences and the structure of vocational interests: The importance of the people-things dimension. Journal of Personality and Social Psychology, 74, 9961009.
  • Lippa, R. A. (2001). On deconstructing and reconstructing masculinity-femininity. Journal of Research in Personality, 35, 168207.
  • Lippa, R. A. (2005). Subdomains of gender-related occupational interests: Do they form a cohesive bipolar M-F dimension. Journal of Personality, 73, 693706.
  • Lippa, R. A., & Connelly, S. C. (1990). Gender diagnosticity: A new Bayesian approach to gender-related individual differences. Journal of Personality and Social Psychology, 59, 10511065.
  • Lobel, T. E. (1994). Sex typing and the social perception of gender stereotypic and nonstereotypic behavior: The uniqueness of feminine males. Journal of Personality and Social Psychology, 66, 379386.
  • Lobel, T. E., & St. Clair, L. (1992). Effects of family responsibilities, gender, and career identity salience on performance outcomes. Academy of Management Journal, 35, 10571069.
  • Lubinski, D., Tellegen, A., & Butcher, J. A. (1983). Masculinity, femininity, and androgyny viewed and assessed as distinct concepts. Journal of Personality and Social Psychology, 44, 428439.
  • Lyness, K. S., & Judiesch, M. K. (2001). Are female managers quitters? The relationship of gender, promotions, and family leaves of absence to voluntary turnover. Journal of Applied Psychology, 86, 11671178.
  • Lyness, K. S., & Thompson, D. E. (1997). Above the glass ceiling? A comparison of matched female and male executives. Journal of Applied Psychology, 82, 359375.
  • Lyness, K. S., & Thompson, D. E. (2000). Climbing the corporate ladder: Do female and male executives follow the same route? Journal of Applied Psychology, 85, 86101.
  • MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., & Sheets, V. (2002). A comparison of methods to test mediation and other intervening variable effects. Psychological Methods, 7, 83104.
  • Main, B. G., O'Reilly, C. A., & Wade, J. (1993). Top executive pay: Tournament or teamwork? Journal of Labor Economics, 11, 606628.
  • Melamed, T. (1995). Career success: The moderating effects of gender. Journal of Vocational Behavior, 47, 3560.
  • Ng, T. W., & Feldman, D. C. (2008). Long work hours: A social identity perspective on meta-analysis data. Journal of Organizational Behavior, 29, 853880.
  • Ng, T. W. H., Eby, L. T., Sorensen, K. L., & Feldman, D. C. (2005). Predictors of objective and subjective career success: A meta-analysis. Personnel Psychology, 58, 367408.
  • Nicholson, N., & De Wall-Andrews, W. (2005). Playing to win: Biological imperatives, self-regulation, and trade-offs in the game of career success. Journal of Organizational Behavior, 26, 137154.
  • Noonan, M. C., Corcoran, M. E., & Courant, P. N. (2005). Pay differences among the highly trained: Cohort differences in the sex gap in lawyers' earnings. Social Forces, 84, 853872.
  • O'Reilly, C. A., & Chatman, J. A. (1994). Working harder and smarter: A longitudinal study of managerial success. Administrative Science Quarterly, 39, 603627.
  • O'Reilly, C. A., Chatman, J. A., & Caldwell, D. F. (1991). People and organizational culture, a profile comparison approach to assessing person-organization fit. Academy of Management Journal, 34, 487516.
  • O'Reilly, C. A., Main, B. G., & Crystal, G. (1988). CEO compensation as tournament and social comparison: A tale of two theories. Administrative Science Quarterly, 33, 257274.
  • Petersen, T., & Morgan, L. A. (1995). Separate and unequal: Occupation-establishment sex segregation and the gender wage gap. American Journal of Sociology, 101, 329365.
  • Phillips, S. D., & Imhoff, A. R. (1997). Women and career development: A decade of research. Annual Review of Psychology, 48, 3159.
  • Powell, G. N., & Butterfield, D. A. (1994). Investigating the “glass ceiling” phenomenon: An empirical study of actual promotions to top management. Academy of Management Journal, 37, 6886.
  • Ragins, B. R., & Sundstrom, E. (1989). Gender and power in organizations: A longitudinal perspective. Psychological Bulletin, 105, 5188.
  • Rapoport, R., Bailyn, L., Kolb, D., & Fletcher, J. K. (1996). Rethinking life and work: Toward a better future. New York: Ford Foundation.
  • Rosenbaum, J. (1979). Tournament mobility: Career patterns in a corporation. Administrative Science Quarterly, 24, 220241.
  • Rothbard, N. P., Phillips, K. W., & Dumas, T. L. (2005). Managing multiple roles: Work-family policies and individuals' desire for segmentation. Organization Science, 16, 243258.
  • Rudman, L. A., & Glick, P. (1999). Feminized management and backlash toward agentic women: The hidden costs to women of a kinder, gentler image of middle managers. Journal of Personality and Social Psychology, 77, 10041010.
  • Rudman, L. A., & Glick, P. (2001). Prescriptive gender stereotypes and backlash toward agentic women. Journal of Social Issues, 57, 743762.
  • Saragovi, C., Koestner, R., Di Dio, L., & Aubé, J. (1997). Agency, communion, and well-being: Extending Helgeson's (1994) model. Journal of Personality and Social Psychology, 73, 593609.
  • Sarros, J. C., Gray, J., Densten, I. L., & Cooper, B. (2005). The organizational culture profile revisited and revised: An Australian perspective. Australian Journal of Management, 30, 159182.
  • Schneer, J. A., & Reitman, F. (1993). Effects of alternative family structures on managerial career paths. Academy of Management Journal, 36, 830843.
  • Schneer, J. A., & Reitman, F. (1995). The impact of gender as managerial careers unfold. Journal of Vocational Behavior, 47, 290315.
  • Sools, A. M., van Engen, M. L., & Baerveldt, C. (2007). Gender career-making practices: On doing ambition or how managers discursively position themselves in a multinational corporation. Journal of Occupational and Organizational Psychology, 80, 413435.
  • Sparks, K., Cooper, C., Fried, Y., & Shirom, A. (1997). The effects of hours of work on health: A meta-analytic review. Journal of Occupational and Organizational Psychology, 70, 391408.
  • Spence, J. T., & Helmreich, R. L. (1978). Masculinity and femininity: Their psychological dimensions, correlates, and antecedents. Austin, TX: University of Texas Press.
  • Stroh, L. K., Brett, J. M., & Reilly, A. H. (1992). All the right stuff: A comparison of female and male managers' career progression. Journal of Applied Psychology, 77, 251260.
  • Taylor, S. E., & Fiske, S. T. (1975). Point of view and perceptions of causality. Journal of Personality and Social Psychology, 32, 439445.
  • Tharenou, P. (2001). Going up? Do traits and informal social processes predict advancing in management? Academy of Management Journal, 44, 10051017.
  • Tharenou, P., & Conroy, D. (1994). Men and women manager's advancement: Personal or situational determinants? Applied Psychology: An International Review, 43, 531.
  • Tharenou, P., Latimer, S., & Conroy, D. (1994). How to make it to the top? An examination of influences on women's and men's managerial advancement. Academy of Management Journal, 37, 899931.
  • Unger, R. K. (1979). Toward a redefinition of sex and gender. American Psychologist, 34, 10851094.
  • United Nations. (2005). The world's women 2005: Progress in statistics.
  • Van der Velde, M. E., Bossink, C. J., & Jansen, P. G. (2005). Gender differences in the determinants of the willingness to accept an international assignment. Journal of Vocational Behavior, 66, 81103.
  • van Vianen, A. E., & Fischer, A. H. (2002). Illuminating the glass ceiling: The role of organizational culture preferences. Journal of Occupational and Organizational Psychology, 75, 315337.
  • Wajcman, J. (1998). Managing like a man. University Park, PA: Penn State Press.
  • West, C., & Zimmerman, D. H. (1987). Doing gender. Gender & Society, 1, 125151.
  • Wong, P. T., Kettlewell, G., & Sproule, C. F. (1985). On the importance of being masculine: Sex role, attribution, and women's career achievement. Sex Roles, 12, 757769.
  • Young, R., & Sweeting, H. (2004). Adolescent bullying, relationships, psychological well-being, and gender-atypical behavior: A gender diagnosticity approach. Sex Roles, 50, 525537.