Does the perceived prestige of a political office affect support for women candidates?

Prestige as a barrier for women political candidates

Previous scholarship suggests that support for women candidates tends to decrease as the level of office increases (Duerst-Lahti 2007), and higher level offices such as the presidency are thought to be more prestigious than lower level offices, such as a city councilor (Huddy and Terkildsen 1993; Rosenwasser and Dean 1989). Data on women’s presence in political institutions documents that as the level of office increase, women’s numeric representation also decreases. ² Other empirical work shows that people express more willingness to vote for women in less prestigious offices (Ono and Burden, 2018). As Crowder-Meyer et al. (2015) explain, women candidates will have a higher chance of victory for “offices that are seen as lower prestige and more secretarial in nature than offices that are perceived as having a higher level of responsibility and prestige” (p. 321). Here, Crowder-Meyer et al. (2015) distinctly connect the gendered dynamics of prestige for women candidates to the level of power associated with the office so that more prestige implicitly equates to more power.

Prestige can pose a gendered barrier through two connected pathways: gender role incongruity and links between prestige and masculinity. Gender role incongruity explains how political leadership stereotypes are associated with masculine traits (Eagly and Karau 2002; Sweet-Cushman 2022), and voters tend to see women political candidates as lacking the masculine qualities associated with politics (Van der Pas, Aaldering, and Bos 2024). Feminine stereotypes and women’s communal social roles (i.e., roles as caregivers or mothers) are incongruous with the masculine perceptions of political offices (Schneider and Bos 2019). If the prestige of a political office is associated with masculinity, then political offices perceived as more prestigious may be more difficult for women to obtain.

Outside of politics, economics research on occupational segregation by gender shows that jobs dominated by women are seen as less prestigious and classified as more feminine relative to jobs dominated by men (Alonso-Villar et al. 2012; Busch 2020). The link between prestige and masculinity is such that masculine traits, such as assertiveness or power, are thought to be required for high-prestige jobs (Bose and Rossi 1983). Recent work tested how occupational prestige might vary for women candidates entering the political pipeline at the local level (Bernhard and Holman 2025). If occupational prestige is shaped by perceptions of masculinity, that is, high prestige jobs are more masculine jobs, then prestige can affect how voters evaluate the perceived fitness of women for political office. Bernhard and Holman (2025) found no link between occupational prestige and masculinity based on the types of career paths for women who run for local office.

Higher-level political offices are not only considered more prestigious, but they are also considered the most masculine (Conroy 2015; Duerst-Lahti 2007; Gunderson et al., 2025). If prestige and masculinity are connected, then women may face multiple barriers among voters. In other words, both the high prestige and high masculinity can jointly affect whether voters will support a woman candidate. Past work does not offer clear empirical tests about whether the masculinity of a political office affects the prestige rating of the office. Are the most prestigious offices considered prestigious because of their association with masculinity? If the most prestigious political offices are also rated the most masculine, then women candidates may face greater barriers pursuing elected office.

Data description

I use original data collected in a 2018 CCES (now CES) module. The module asked respondents to rate the prestige of the following political offices: city council, governor, president, senate, and state legislature—with the order randomized. Another question then asked participants to rate the femininity-masculinity of these same offices. I asked participants how willing they would be to support a woman candidate with the level of the political office randomized so that each person only saw one of the following three political offices including city council, the senate, or the presidency. My expectation is that if there is a link between prestige and support for women candidates, then support for a women candidate will be highest at the least prestigious level of office and should decrease as the level of office increases. The online appendix includes the full question wordings and response options of these key variables, and Table A1 of the online appendix includes descriptive data about the CCES sample.

The average prestige rating across all the levels of office was M = 3.116 (SD = 0.657) on a 1–4 scale with higher values indicating higher prestige (see Table A2). This average rating is above the mid-point of the scale and suggests that people, in general, see holding political office as somewhat prestigious. Figure 1 shows the percentage of respondents who rated each office as very prestigious, somewhat prestigious, somewhat not prestigious, or not very prestigious. The descriptive data shows that as the level of office increases so too does the prestige. Only about 6% of respondents rated a city council seat as very prestigious while the vast majority of respondents rated a city council seat as somewhat prestigious. But, moving up to a state legislative office 21% of respondents rated a state legislative seat as very prestigious. Overall, perceived prestige steadily increases with the level of political office with 76% of respondents rating the presidency as the most prestigious level of office.OPEN IN VIEWER

I next turn toward examining if there is a relationship between the perceived prestige of a political office and the perceived masculinity of an office. The average feminine-masculine stereotypic rating across all the political offices was M = 2.810 (SD = 0.863) on a 1–4 scale with higher values indicating higher levels of masculinity. Figure 2 displays the masculine-feminine rating for a city council seat, senate seat, and the presidency (see Table A2). As the level of office increases so does the perceived masculinity of the office, though it is important to note that most respondents for the city council, state legislature, governor, and senate fall into the somewhat masculine category for these offices. The masculinity ratings mirror the prestige ratings most clearly for the presidency as the most masculine and most prestigious office and city councils as the least masculine and least prestigious office.OPEN IN VIEWER

To further examine if there is a relationship between perceived prestige and masculinity, I conducted polychoric correlations between prestige and masculinity for city councils, the senate, and the presidency. Polychoric correlations are appropriate as the two variables of interest, prestige and masculinity, are ordinal variables. The correlations show no significant relationship between the masculinity ratings and the perceived prestige of the political office for city councils (r = .0177 p = .4675), U.S. Senate (r = .0888, p = .2161), and the presidency (r = .3078, p = .3687). A strong body of theoretical research suggests that there should be a connection between masculinity and political offices, especially at high levels such as the senate or the presidency (Conroy 2015; Duerst-Lahti 2007; Gunderson et al., 2025; Rosenwasser and Dean 1989). The lack of a significant relationship at any level of office between these two characteristics of political offices suggests that if prestige serves as a barrier for women candidates it may not be through masculinity.

I next turn to identifying whether the perceived prestige and/or masculinity of a political office affects people’s expressed willingness to support a woman candidate. These analyses, which are exploratory, examine the relationship between the perceived prestige of a political office and expressed electoral support for women candidates. The expressed vote support question only asked about a city council seat, a senate seat, and the presidency and each respondent only answered this question for one of these three offices with the office randomly assigned. The overall level of vote support for a woman candidate was M = 3.509 (SD = 0.711) with the scale ranging from 1 to 4 and higher values indicating more electoral support. Figure 3 shows the descriptive patterns of support for women candidates at these levels of office. There is little variation in the levels of support for a woman candidate based on the level of political office. About 63% of respondents said they were very likely to support a woman for a city council seat, the least prestigious level of office. At the level of the presidency, 53% of respondents said they would be very likely to support a women candidate at the most prestigious level of political office. While majorities say they would vote for a woman, the 10-point decline in support from city councils to the presidency fits the theoretical expectation that high prestige reduces support for women.OPEN IN VIEWER

To consider the relationship between prestige and masculinity on the expressed support for a woman candidate, I estimated a series of ordered logistic regression models as my outcome variable is categorical with a rank-ordering in the response options. I included an interaction between prestige and masculinity to test how these dynamics together and separately affect expressed support for a woman candidate. The models incorporate a number of attitudinal control variables to capture attitudes voters have about women in politics and in public life including feminist political identities, attitudes about the importance of electing women to political office in general, gender favoritism in public policy measures (see Goldman (2018)), and benevolent and hostile sexism measures (Bracic et al. 2019; Cassese and Holman 2019). Along with these attitudinal measures, I also include controls for standard political attitudes including party identification, political ideology as well as standard demographic variables. All the survey questions and response options in my module are in the online appendix along with the corresponding CCES common content questions.

The full results are in Table A3 of the online appendix. If prestige, tied to masculinity, is a barrier for women candidates then the interaction term should be significant and negative. The interaction term between prestige and masculinity is insignificant for city council offices (β = 0.3705, SE = .2582, p = 0.151) and for the presidency (β = 0.0409, SE = 0.1745, p = 0.814). The interaction term for the senate shows a significant, positive effect (β = 0.7994, SE = 0.2645, p = 0.003). This unexpected positive relationship between prestige and masculinity for the senate suggests that higher prestige and higher masculinity might increase the likelihood of a respondent saying they would be more likely to support a woman candidate. This result is interesting, but should be interpreted cautiously, given the significant negative main effects of prestige (β = −1.4338, SE = 0.7940, p = 0.071) and for masculinity (β = −3.2443, SE = 0.8335, p < 0.001). These negative main effects are in line with the theoretical expectations from past work, but it is not clear from these initial exploratory analyses why the interactive effect between prestige and masculinity would lead to higher levels of support for women.

These initial results suggest that prestige may not pose an electoral barrier for women candidates; however, these results should be interpreted cautiously given the exploratory nature of these analyses. The relatively high levels of respondents falling into the somewhat or very likely category of supporting a woman candidate suggest ceiling effects. Additionally, social desirability biases may be at play (Burden et al. 2017; Krupnikov et al. 2016). Unfortunately, the CES did not include self-monitoring questions to assess the extent to which these biases affect support for women candidates.

Curiously, across all the models women respondents have a statistically significant and lower predicted probability of expressing positive vote support for a woman city council, senate, and presidential candidate relative to men. This finding runs counter to the gender affinity literature (Dolan, 2008; Plutzer and Zipp, 1996; Sanbonmatsu, 2002; Zipp and Plutzer, 1985). To better assess this unexpected effect, I replicated the models in Table A4 with a three-way interaction that adds gender to prestige and masculinity. Women who rate Senate offices high on prestige and masculinity have lower levels of expressed support for a woman candidate, but this relationship does not occur in the city council or presidency models. Future work should replicate these findings with an aim toward identifying the causal factors that might be behind these patterns.

I also examined how partisanship may affect prestige and masculinity ratings and people’s willingness to support a woman candidate. Three-way interaction models between partisanship, masculinity, and prestige (see Table A5 of the online appendix) show no significant differences in support for a woman based on respondent partisanship, prestige, and masculinity ratings.

Discussion & conclusion

There are several key findings from this research note: First, voters see higher levels of office as both more prestigious and more masculine relative to lower levels of political office. Second, while there is a theoretical link between prestige and masculinity, the findings here do not show a strong statistical association between these two concepts. Third, high levels of prestige and masculinity do not appear to diminish support for women candidates. Of course, these results should be interpreted cautiously as they are an exploratory first step in examining prestige and support for women.

This study comes with limitations. The effect of prestige on support for women candidates may occur via an indirect path in the ways that voters search for, receive, and process information about candidates (Ditonto 2019; Ditonto et al. 2014). Prestige may pose a barrier for women at other stages of the electoral process such as making decisions to enter the political pipeline (Crowder-Meyer 2020) or party nomination processes (Crowder-Meyer 2013). This research did not consider how intersectionality affects support for women candidates for high prestige offices (Matos et al. 2020) or support for LGBTQ candidates (Naunov 2025; Strode 2025).

While the results did not show that prestige negatively effects support for women, this does not mean that prestige never plays a role in how voters evaluate women candidates.

Moreover, these results certainly do not mean that women do not face any challenges when running for political office (see, e.g., Dovi (2025)) and prestige may still pose a barrier through other mechanisms or in other stages of the candidate evaluation process not investigated here. Even if prestige or masculinity has an indirect effect on support for women, untested in this research, there still exists a perception among some that women are not suited for serving in the most powerful political roles. And, indeed, women’s absence in the highest and more powerful political positions in the U.S. is striking. Identifying and understanding the barriers women may face in reaching high levels of political power sheds light on the conditions under which women’s political representation may improve.

Supplemental material