Continuing our new series of collaborations with political science journals, we are pleased to present the following guest post from University of Wisconsin Milwaukee political scientist Kathleen Dolan discussing her article “Gender Stereotypes, Candidate Evaluations, and Voting for Women Candidates: What Really Matters?” that appears in the current issue of Political Research Quarterly. In conjunction with this post, SAGE will make the article freely available to all for the next 30 days; you can download it here.
As the latest round of speculation about Hillary Clinton’s intentions regarding the 2016 presidential race begin to heat up, we are reminded by some critics that she is just too darned old to be president. Of course, attacking a candidate’s perceived vulnerabilities is a tried and true element of politics. But negative attention to the age and appearance of a woman candidate has a certain resonance in our politics because it stirs concerns of sexism and condescension toward these women.
We can easily conjure up anecdotal examples of high-level women candidates who have been subjected to criticism and attacks because of their age (too young or too old), appearance (too beautiful or too plain), or family status (whether mothers or childless). Ask Sarah Palin, Jari Askins, or Kelly Ayotte what is it like for women at times on the campaign trail. For support, consult the extensive literature in political science that examines the gender stereotypes people hold about what women and men candidates are like and what they can successfully do in office. Research on female and male stereotypes often warns that these attitudes can work against women candidates if voters perceive them not to possess the “right” skills and abilities for office.
However, look closer and you will see that we have less evidence of the negative impact of gender stereotypes on women candidates than we thought. Much of the existing data comes from experiments and hypotheticals, rather than from measures of attitudes connected with actions toward real candidates in real contests. In 2010, in order to bring new data to bear on these issues, I conducted a survey of 3150 U.S. adults in 29 states, designed to gather information about abstract gender stereotypes, specific evaluations of candidates running for office, and vote choice. These data allow me to link the gender stereotypes people may (or may not) hold with their specific actions – candidate evaluations and vote choice – in these elections.
I find very little support for the concern that abstract gender stereotypes hurt, or help, women candidates when they run against men. For Republican women candidates, none of a range of policy and trait stereotypes is significantly related to any of the candidate evaluations voters make about their traits or abilities. That is to say that stereotyped ideas about women’s superiority on education issues or men’s greater leadership abilities had no bearing on evaluating the policy abilities or traits of the specific Republican women candidates in these House races. For Democratic women, the only negative impact of stereotypes is that respondents who see women as less able to handle “male” policies areas like the economy and military in the abstract evaluated the Democratic woman candidate in their House race as less well-suited to deal with these issues than her male opponent.
However, this limited impact for stereotypes on evaluations doesn’t translate into an impact on the thing that matters most to candidates – vote choice. Examining the impact of abstract stereotypes on vote choice alongside traditional political influences such as political party, incumbency status, candidate spending clearly demonstrates that stereotypes have no direct impact on vote choice. Stereotypes also have no indirect impact through candidate evaluations. Instead, the factors that predict whether a voter will choose a woman candidate for the House are the same things that predict vote for a man. Overwhelmingly, people vote for the candidate of their political party, regardless of the sex of the candidate. In every analysis, a shared party identification between the respondent and woman candidate was the most important influence in determining vote for a woman and no stereotyped attitudes were significantly related to vote choice.
One additional thing to note from this analysis is that the political party of a woman candidate is key to how she is perceived by voters. A glimpse of this is visible in Figures 1 (above) and 2 (below). Each figure indicates the direction of the relationship between abstract stereotypes and evaluations of the candidates. For Democratic women, people are more likely to hold the expected stereotype that they are better at female policies and less well-suited for handling male policies. For Republican women, the slopes reveal exactly the opposite, that they are perceived as better able to handle male policy areas and less well-suited for female issues. This suggests that even when gender stereotypes are not related to impressions of women candidates, political party interacts with candidate sex in meaningful ways to shape the way these women are seen by the public.
The analysis of House elections in 2010 reminds us that party matters and demonstrates that stereotypes do not. This first finding should not be surprising. The second should allow us to have a clearer picture of the dynamics of elections involving women candidates. Anecdotal focus on a woman candidate’s hair or family status or ability to be commander in chief should not distract us from the evidence that abstract stereotyped thinking about women and men doesn’t play a central role in the fates of women candidates.