If the shoe fits : Gender role congruity and evaluations of public managers

ole congruity theory suggests that perceived incongruity between the roles and attributes typically associated with women and the roles and attributes typically associated with effective managers will result in less favorable attitudes towards women managers (Eagly & Karau, 2002). Since women are expected to exhibit communal characteristics (e.g., kindness, compassion, nurturing) and men are expected to exhibit agentic characteristics (e.g., independence, ambition, confidence)—characteristics traditionally associated with effective managers— women are perceived to be less congruent with management positions (Eagly, 1987; Martell, Parker, Emrich & Crawford., 1998). The “think manager – think male” phenomenon has been well documented (Schein, 1973; Schein, 1975): When describing the characteristics of a successful manager, individuals more often list the agentic characteristics traditionally associated with men rather than the communal characteristics traditionally associated with women. In other words, when people think of managers, they inevitably think of men. Perceived incongruity between women and management is not confined to private corporations. Public organizations have been characterized as inherently masculine as well (Ferguson, 1984; Stivers, 2002). Indeed, there is a large literature examining the role of gender in the public sector (for examples, see Dolan, 2004; Funk, Silva, & Escobar-Lemmon, forthcoming; Guy & Newman, 2004; Keiser, Wilkins, Meier & Holland, 2002; Meier & Funk, 2017). The public dimension of public organizations is precisely what makes these organizations masculine because, historically, the public sphere has been understood as men’s domain and the domestic sphere as women’s domain (Stivers, 2002). This, combined with both the legal and practical constraints placed on women’s public participation throughout history, contributes R Abstract: Traditional stereotypes about both gender and managers make women appear to be incongruent with management positions; a potential consequence of which is less favorable evaluations of women managers. Using an original survey experiment, this study tests whether and how gender role incongruity, combined with the gender-typing of organizations and gender of the evaluator, affect evaluations of hypothetical public managers. Results suggest that men and women public managers are evaluated equally favorably overall; however, men evaluators perceive women managers less favorably than do women evaluators. This is largely driven by the gender-typing of organizations. While men evaluators rate women managers in feminine organizations as favorably as do women evaluators, they rate women managers in masculine organizations less favorably compared to women evaluators. Indeed, men evaluators rate women managers in masculine organizations lower compared to all other groups of comparison, including all other possible combinations of evaluator, manager, and organization gender. Findings indicate that though perceived incongruity between women and management positions may have diminished over time, there is evidence that gender biases still remain problematic for women managers’ careers, especially in masculine gender-typed organizations.

ole congruity theory suggests that perceived incongruity between the roles and attributes typically associated with women and the roles and attributes typically associated with effective managers will result in less favorable attitudes towards women managers (Eagly & Karau, 2002).Since women are expected to exhibit communal characteristics (e.g., kindness, compassion, nurturing) and men are expected to exhibit agentic characteristics (e.g., independence, ambition, confidence)-characteristics traditionally associated with effective managerswomen are perceived to be less congruent with management positions (Eagly, 1987;Martell, Parker, Emrich & Crawford., 1998).The "think managerthink male" phenomenon has been well documented (Schein, 1973;Schein, 1975): When describing the characteristics of a successful manager, individuals more often list the agentic characteristics traditionally associated with men rather than the communal characteristics traditionally associated with women.In other words, when people think of managers, they inevitably think of men.
Perceived incongruity between women and management is not confined to private corporations.Public organizations have been characterized as inherently masculine as well (Ferguson, 1984;Stivers, 2002).Indeed, there is a large literature examining the role of gender in the public sector (for examples, see Dolan, 2004;Funk, Silva, & Escobar-Lemmon, forthcoming;Guy & Newman, 2004;Keiser, Wilkins, Meier & Holland, 2002;Meier & Funk, 2017).The public dimension of public organizations is precisely what makes these organizations masculine because, historically, the public sphere has been understood as men's domain and the domestic sphere as women's domain (Stivers, 2002).This, combined with both the legal and practical constraints placed on women's public participation throughout history, contributes Funk, 2019 to the conceptualization of public management as more congruent with men than with women.In order for the perceived incongruity between women and management roles to diminish, there must be a change in conceptualizations of women's prescriptive roles and attributes, conceptualizations of successful managers, or both.

Women and Public Management: Role Incongruity or Female Leadership
Advantage?
There are reasons to suspect these changes might have occurred over the last several decades in the United States.Collectively, women have accumulated decades of experience in the paid workforce and are increasingly represented in public management positions as well.There is also growing evidence that women public managers are just as effective as-or perhaps even more effective than-men public managers.Recent studies suggest that increasing women's numeric representation at both the lower level (Keiser et al., 2002;Meier, Mastracci, & Wilson 2006) and upper levels (Johansen, 2007) of public organizations can improve organizational performance.
Studies of the perceived effectiveness of managers in the public sector and beyond also suggest that women are perceived to be equally effective (Eagly, Karau, & Makhijani, 1995) or even more effective (Paustian-Underdahl, Walker, & Woehr, 2014) than men.
Moreover, some argue that ideas about effective leadership practices have changed in recent years, with more importance being placed on stereotypically feminine attributes, such as collaboration, mentoring, and transformational leadership (Koenig, Eagly, Mitchell & Ristikari., 2011).Since women are assumed to possess such attributes and leadership styles (Funk, 2015), this has led to the creation of the idea of a "female leadership advantage" (Helgesen, 1995;Rosener, 1995;Vecchio, 2002).Proponents of this idea suggest that since perceptions of effective management practices are now more closely aligned with feminine attributes rather than masculine ones, women should have the upper hand when it comes to management.
Yet, current indicators of gender equality suggest that there remain significant gender disparities in wages, an underrepresentation of women in leadership positions, and substantial gender segregation in the workforce.Recent figures suggest that the gender pay gap in the United States remains quite large: Women are paid around 80% of what men are paid (AAUW 2018).Moreover, women leaders in both the public and private sectors are rare.As of January 2018, only 27 Fortune 500 companies had women CEOs (Fortune Editors 2017).With respect to political leadership, only a handful of countries have had women heads of state (Jalalzai, 2013) and women's legislative representation remains low compared to their share of the population (Inter-Parliamentary Union 2018), including at the local government level (Escobar-Lemmon & Funk 2018).Furthermore, people often state they prefer to have a male boss rather than a female one (Carroll, 2006;Riffkin, 2014; but see Brenan, 2017).All of these indicators suggest that women likely still face a great deal of gender discrimination and double standards; thus, women may not have a new leadership advantage after all.
Part of the explanation for why these observed disparities still exist may be that perceived incongruity between women and management roles is still active in the minds of many.Even if the leadership styles commonly associated with women do lead to better outcomes, and even if women are actually more effective managers, prescriptive gender stereotypes about how women ought to behave can elicit backlash for women who are successful in male gender-typed roles, including management positions (Heilman, 2012;Heilman, Wallen, Fuchs & Tamkins, 2004;Rudman & Glick, 2001).Prescriptive gender stereotypes create "normative standards for behavior that induce disapproval and social penalties when they are directly violated or when violation is inferred because a woman is successful" (Heilman, 2012, p. 113).This can lead to women being denied credit for their successes, penalized for their competence, and underappreciated for improving performance (Greenhaus & Parasuraman, 1993;Heilman, 2012).
Thus, though ideas about management may have changed recently, resulting in a de-emphasis on stereotypically masculine attributes, "a good manager is still perceived as predominately masculine" (Powell, Butterfield, & Parent, 2002, p. 177).Rather than feminine traits replacing masculine ones in order of importance, today, a good manager may be expected to display both feminine and masculine characteristics (Kark, Waismel-Manor & Shamir, 2012;Vecchio, 2002).For all these reasons, women managers may be perceived less favorably than their men counterparts, even in successful organizations.
Gender Role Incongruity Hypothesis: Women public managers are evaluated less favorably than men public managers.

Women in Masculine Gender-Typed Organizations
Perceived incongruity between women and management is likely exacerbated as the mismatch between women's gender and the gender-typing of an organization increases.In other words, women managers may be evaluated even less favorably if they work in traditionally male-dominated organizations.Evidence from Eagly, Karau, & Makhijani's (1995) seminal meta-analysis suggests women leaders are equally effective as men in the aggregate; however, women are less effective in highly masculinized leadership roles and organizations numerically dominated by men.More recent research suggests that when women do succeed in masculinized roles, they often elicit negative reactions (Heilman et al., 2004)-likely because it is assumed these women lack communal attributes (Heilman & Okimoto, 2007).
The gender-typing of organizations occurs because "gender is present in the processes, practices, images and ideologies, and distributions of power in the various sectors of social life" (Acker, 1992, p. 567; see also Acker, 1990).Organizations become more feminine in orientation the more they align with women's traditional roles as caregivers, nurturers, and childcare providers, and more masculine the more they correspond with men's traditional roles as defenders, breadwinners, and power-holders.Organizations can be characterized by gender by examining both the nature of the work done by the organization as well as the gender composition of the organization's employees.Oftentimes, these two indicators are highly correlated.Organizations that operate in policy areas more closely aligned with women's traditional roles (e.g., healthcare, education, social services) tend to have more women employees, while organizations in policy areas closely aligned with men's traditional roles (e.g., economics, defense, finance) tend to have more men employees.In U.S. local governments, for example, 90% of elementary school teachers are women, while around 88% of police officers are men (Guy, 2017).
If women are perceived more favorably when they work in organizations that are congruent with their traditional gender roles, and less favorably when they work in organizations perceived to be incongruent with their gender, we would expect women managers in masculine organizational contexts to be evaluated less favorably compared to their colleagues who are men.
Organizational Incongruity Hypothesis: Women public managers are evaluated less favorably than men public managers when they work in masculine organizational contexts.

Gender of the Evaluator
Another potential source of variation in perceptions of women managers is the gender of the evaluator (Eagly & Karau, 2002).In the 1970s, both men and women managers perceived management to be more congruent with men's stereotypical attributes than with women's stereotypical attributes (Schein, 1973;1975).Nearly two decades later, research emerged suggesting women had changed their minds.While men still associated management with masculine characteristics, women no longer sex-typed managerial jobs (Brenner, Tomkiewicz & Schein, 1989).Further research indicates that men view women as less qualified and less likely to possess the characteristics required to be a successful manager, while women tend to be significantly less prejudiced against women managers (Bowen, Chieh-Chen, Swim & Jacobs, 2000;Rudman & Kilianski, 2000;Schein, 2001;Schein, 2007).Thus, perhaps the "think managerthink male" paradigm is more alive in the minds of men than it is in the minds of women.
Gender of the Evaluator Hypothesis: Women public managers are evaluated less favorably than men public managers by men evaluators, but evaluated equally by women evaluators.

Research Design and Sampling Strategy
To test these hypotheses, a survey experiment was implemented using Amazon Mechanical Turk (MTurk). 1 MTurk potentially provides a more diverse and representative sample than more conventional sampling strategies, such as those involving university students (Stritch, Pedersen & Taggart, 2017).Indeed, the sample used for this study includes greater variation in age, income, and educational attainment than is usually present among university students (see supplemental materials for details).The sample was limited to adults in the U.S. since conceptions of gender identity and traditional gender roles varies across countries.
Respondents were asked to read a hypothetical end-of-the-year report about a public organization that experienced good performance over the past year.The reports varied two key pieces of information: whether the manager is a man or a woman and whether the organization falls within a stereotypically masculine or feminine domain.Thus, respondents were randomly assigned to read a report with either a (1) woman manager in a feminine organization, (2) man manager in a feminine organization, (3) woman manager in a masculine organization, or (4) man manager in a masculine organization.The manager's gender was conveyed five times throughout the report by using the manager's name, gendered titles, and gender pronouns.The manager was identified as "Lisa" or "Michael" R. Jones. 2 The gender-typing of the organization was also conveyed five times through use of the organization's name, Children's Health First or Economic Development First.Both children's issues and healthcare correspond to women's traditional roles in the domestic sphere and fall under the purview of stereotypically feminine issues, thus Children's Health First was selected as the feminine organization.Likewise, economic development issues correspond to men's traditional roles in the public sphere and fall under the purview of stereotypically masculine issues, so Economic Development First was selected as the masculine organization.Table 1 presents the gender distribution of respondents into each of the four treatment conditions.

Dependent Variable: Evaluations of the Manager
Following the report, respondents were asked to evaluate the manager by indicating the extent to which they agree with the following statements: (1) the manager performed well, (2) the manager is competent, (3) the manager provided good leadership for the organization, (4) the manager seems to be a good fit for this organization, and (5) I would renew the manager's contract for another year.Respondents could select a value between 1 and 10, where 1 indicates "strongly disagree" and 10 indicates "strongly agree."The evaluation measures were combined us-ing factor analysis to generate a standardized evaluation measure, which is used as the dependent variable in the analyses presented below. 3The evaluation variables are highly and statistically significantly correlated with one another, and each have correlation coefficients of 0.75 or higher.Each evaluation variable loads positively onto a single factor with loading scores of 0.89 or higher (Eigenvalue = 4.33).The resulting variable is continuous and standardized, with a mean of zero and standard deviation of one, and ranges from -3.62 to 1.61.Higher values indicate more favorable evaluations of the manager, while lower values indicate less favorable evaluations.

Testing the Role Incongruity Hypothesis
To test whether perceived role incongruity leads to less favorable evaluations of women managers, the effect of manager gender on evaluations of the manager's performance is examined using regression analysis.The results suggest that women managers are evaluated no more or less favorably than men managers, all else equal.In other words, the manager's gender does not have a statistically significant effect on overall evaluations of the manager.This finding is robust to the inclusion of additional variables controlling for the gender of the organization, respondent demographics, and manipulation checks, and holds for both the standardized and individual evaluation variables.Table 2 provides a summary of the regression results.Based on these analyses, we might conclude that there is neither a female leadership advantage nor a perceived role incongruity that results in women managers being evaluated more or less favorably than men managers.

Testing the Organizational Incongruity Hypothesis
However, these results alone are insufficient for determining whether women are evaluated less favorably when they manage organizations in masculine contexts (or more favorably when they manage organizations in feminine contexts).For this reason, another set of regression models are estimated accounting for the interaction between the manager's gender and the gender-typing of the organization.(See the supplemental appendix for the full regression table .) Figure 1 shows the predicted standardized evaluations of men and women managers in both feminine-and masculine-typed organizations.Comparing each of the scenarios, we see that the average predicted value for women managers of masculine organizations is slightly smaller than the predicted values for women managers of feminine organizations, and men managers of both types of organizations.However, pairwise comparisons of the predicted effects suggest that the only statistically significant difference among the scenarios is observed when comparing evaluations of women managers in masculine organizations to women managers in feminine organizations (p=0.072). 4Thus, there is partial support for

Figure 1 Predicted Evaluations of Public Managers by Organization Gender-Typing
the hypothesis that women managers are evaluated worse when they manage organizations in masculine contexts, but this is relative to evaluations of women managers in feminine organizations.There is not a statistically significant difference between evaluations of women managers in masculine organizations and evaluations of men managers in either type of organization when pooling across respondent gender.

Testing the Gender of the Evaluator Hypothesis
The third hypothesis suggests that the gender of the evaluator may also affect evaluations of men and women managers, with men evaluators in particular rating women managers more harshly.To test this hypothesis, the gender of the manager is interacted with the gender of the respondent. 5Figure 2 shows predicted standardized evaluations of a (1) man manager given by a man respondent, (2) man manager given by a woman respondent, (3) woman manager given by a man respondent, and (4) woman manager given by a woman respondent.Comparing predicted values across the four scenarios, we see that women managers are indeed evaluated less favorably by men evaluators compared to how women evaluate women managers.This difference is statistically significant, which can be seen in the fact that the 95% confidence intervals do not overlap (p<0.000).Pairwise comparisons of the predicted effects also suggest that men evaluators rate women managers less favorably compared to how women evaluators rate men managers (p=0.006). 6 There is not, however, a statistically significant difference between how men evaluate men managers and how men evaluate women managers (p=0.104).The important distinction here is that the statistically significant difference lies in the comparison of men and women respondents and their evaluations of women managers, not in the comparison of men's evaluations of men managers to men's evaluations of women managers (though this pairwise comparison does approach conventional levels of statistical significance).These findings lend support to the hypothesis that the gender of the evaluator matters for evaluations of women managers; however, the significant difference lies in how men evaluate women managers and how women evaluate women managers, rather than how men evaluate women managers compared to how men evaluate men managers.

Examining the Gender of the Evaluator across
Managers and Organizations Figure 3 presents predicted standardized evaluations generated from the results of a regression model including a three-way interaction between the gender of the manager, gender of the organization, and gender of the evaluator to test whether men evaluate

Figure 2 Predicted Evaluations of Public Managers by Respondent Gender
women managers in masculine organizations especially harshly compared to women evaluators.Note that the X-axis in Figure 3 is the gender of the organization, while the gender of the respondent is identified using the markers in the graph's legend.Again, we see statistically significant differences in the ways that men and women respondents evaluate women managers; however, the differences appear to be largely driven by men evaluators' low ratings of women managers in masculine-typed organizations.
Pairwise comparisons of the predicted effects suggest that men evaluate women managers in masculine organizations less favorably compared to how women evaluate both men and women managers across masculine and feminine organizations (statistically significant at p<0.05 for all four comparisons).Moreover, the pairwise comparisons test also reveals that men evaluate women managers in masculine organizations less favorably compared to men's evaluations of: women managers in feminine organizations (p=0.007),men managers in feminine organizations (p=0.016), and men managers in masculine organizations (p=0.1).In other words, men respondents evaluate women managers in masculine organizations significantly less favorably compared to all other combinations of manager, organization, and evaluator gender. 7 Thus, the gender difference observed among men and women evaluators in the previous analysis appears to be driven primarily by men's perceptions of women managers in masculine organizations.While women respondents evaluate both men and women managers equally favorably regardless of the type of organization they manage, men evaluate women managers less favorably in masculine gendertyped organizations (compared to both women evaluators and to men evaluators in other treatment groups).Consequently, it appears that men evaluate women managers less favorably when women manage organizations that appear to be incongruent with their gender.

Discussion and Conclusions
These findings add important nuance to role congruity theory and the "think manager -think male" paradigm.Results of this study suggest perceived incongruity between women and management positions in general may have diminished over time, so much so that women managers are now evaluated equally with men managers in the aggregate.However, findings

Figure 3 Predicted Evaluations of Public Managers by Respondent
Gender and Organization Gender-Typing also indicate that perceived incongruity likely does still affect evaluations of women managers, but the perceived incongruity is between women and management positions in masculine-typed organizations, rather than management positions in general, and may only exist in the minds of men evaluators.These findings also add to our understanding of perceptions of women in management roles throughout history.In the 1970s, both men and women perceived women to be incongruent with management positions (Schein, 1973;1975).By the late 1980s, women had changed their minds, but men still held this view (Brenner et al., 1989).Today, men seem to be coming to terms with women in management positions in general, but still seem to struggle with perceived incongruity between women and management positions in traditionally masculine organizations.
In substantive terms, the results indicate that the differences in how men and women evaluate women managers are relatively small.However, these small differences could amount to very large disparities in the real world, especially considering that men are oftentimes placed in the role of the evaluator in realworld settings.Management positions, and positions of power more generally, are typically held by men.Thus, men are more likely to be in positions where they are providing evaluations that matter for women's careers (i.e., as managers, supervisors, members of hiring or promotion committees, etc.).In addition, the discounting of women's job performance relative to men's, though seemingly minute, can snowball over the course of a woman's career.For example, if salary raises and promotions are based on performance evaluations, and if women are consistently evaluated worse than men, then women will be less likely to receive raises and promotions compared to men, leading to even wider gender gaps in wages and the representation of women in management positions over time.
Yet, the results of this study also leave room for optimism: Perceived role incongruity may only negatively affect evaluations of women managers in masculine organizations and could perhaps be countered by using women as evaluators.This also implies that women managers should be evaluated equally favorably as their men counterparts in organizations that are less clearly gendered (i.e., organizations that are not clearly masculine).Still, those placed in positions of power and decision-making roles in public organizations should be cognizant of the effects of perceived gender incongruity when using performance evaluations to make decisions about promotions, raises, or other issues likely to affect the trajectory of women's careers.
7. Pairwise comparisons also suggest that men evaluators rate men managers in masculine organizations lower than women evaluators rate women managers in both feminine organizations (p=0.033) and masculine organizations (p=0.035).

Table 1 Number of Respondents in Each Treatment Condition, by Gender
Note: Survey includes 280 respondents: 113 women and 167 men.Respondents were randomly assigned to groups.