Poli-Sci Perspective is a weekly Wonkblog feature in which Georgetown University's Dan Hopkins and George Washington University's Danny Hayes and John Sides offer an empirical perspective on the issues dominating Washington. In this edition, Hayes summarizes research showing that bias against female political candidates in both the media and among voters is weakening.
When the 113th Congress convenes in January, women will occupy more seats than ever before. Eighty-one in the House, and 20 in the Senate. Amy Klobuchar better get ready for more traffic jams.
Why did women, most of whom were Democrats, do well in 2012? Observers have offered numerous explanations, including controversial comments about rape and abortion by “self-immolating” Republican candidates, an election-year focus on women’s health issues, redistricting, or a favorable electoral environment for the Democratic Party.
But here’s another: gender bias – either by the media or the voters – is no longer the impediment to female candidates that it once was. That is the conclusion that Jennifer Lawless of American University and I draw in a recent investigation of media coverage of and voter attitudes toward U.S. House candidates. We find that neither news coverage nor voters’ assessments of female candidates reflects the kind of gender stereotyping that has typically pervaded contemporary American politics.
For decades, observers have noted that women running for political office are portrayed in the media in ways that are consistent with gender stereotypes. Long before Sarah Palin, research had shown that female candidates tended to receive less attention in the news and were often covered in a fluffier fashion – with an emphasis on their appearance (Elizabeth Dole’s nails, anyone?), personality, family roles, or “feminine” traits, such as compassion and honesty. Men, on the other hand, were more likely to be portrayed as experienced, competent leaders.
Survey research and experimental studies have shown that voters also evaluate candidates through a gendered lens. Taken together, news coverage and public attitudes suggest that it has been more difficult for women to persuade voters that they are suited for the rough-and-tumble of Washington politics. It’s not that women can’t successfully make the case. But it has been thought to be tougher than for men.
But is this still true? Against a backdrop of changes in public opinion about women in politics and the rise of party polarization – the growing ideological divide between the parties, which may make candidate sex less salient to voters – Lawless and I undertook a two-part study of gender stereotyping during the 2010 midterm elections.
We first conducted a detailed analysis of local newspaper coverage of House races in nearly 350 congressional districts across the country. Analyzing 4,748 articles, we found virtually no gender differences whatsoever.
News coverage of women was just as common as coverage of men. And the content of campaign stories was nearly indistinguishable across candidate sex. The frequency with which reporters referred explicitly to candidates’ sex or gender – for instance, noting how they dressed or their family roles – was the same for men and women. Paul Ryan, hustling to the tailor to get that suit taken in, wouldn’t be surprised.
Mentions of candidates’ personal characteristics also did not fall along stereotypical gender lines. Women were just as likely as men to be portrayed as possessing competence and leadership skills and no more likely to be covered as trustworthy or warm.
The dots in the figure below show the proportion of candidates in 2010 whose news coverage included at least one mention of four broad traits: competence, leadership, integrity, and empathy. The data are also broken down by whether the mention was positive (for example, “competent”) or negative (such as “incompetent”). Regardless of whether the traits are stereotypically “male” or “female,” or positive or negative – the coverage was the same for men and women. For instance, 25 percent of female candidates received positive mentions of their competence, while 24 percent of male candidates did. Across the traits, there were no statistically significant gender differences.
But even if media coverage isn’t gendered, voters’ assessments could still be. If they are, then we would expect women to be judged more favorably than men on empathy, trustworthiness, and integrity, but less favorably on leadership and competence. In the second part of the study, Lawless and I examined data from the 2010 Cooperative Congressional Election Study, a nationally representative survey of citizen attitudes toward U.S. House candidates.
The chart below shows the effect of candidate sex on voters’ assessments of the candidates’ traits, controlling for various factors that shape political evaluations (party identification, ideology, incumbency status, etc.). The dots represent regression coefficients for candidate sex, with 95 percent confidence intervals. A dot to the right of the vertical dashed line indicates a more favorable evaluation for female candidates on average. A dot to the left indicates that female candidates were assessed less favorably. Evaluations of Democratic and Republican candidates are shown separately.
Candidate sex had no effect on voters’ assessments. Even setting aside the absence of statistical significance, the pattern of results is wholly inconsistent with gender stereotyping. Female Democrats were assessed as slightly weaker leaders, less empathetic, less trustworthy, less competent, and less likely to possess integrity than male Democrats. Republican women were viewed as stronger leaders, more empathetic, more trustworthy and equally competent and possessing integrity as men. Please disperse, nothing to see here.
If not gender, then what does drive media coverage and voter assessments of candidates? The usual suspects. For the media, competitive campaigns receive more coverage, and the tone of coverage varies by whether a candidate is an incumbent.
For voters, party identification, ideology and incumbency shape attitudes toward candidates. We also found that news coverage of candidates can influence voters’ opinions. But because the media don’t engage in pervasive stereotyping, coverage doesn’t systematically harm or help women. Reporters and voters react in similar ways to candidates, regardless of their sex.
This is not to say that gender inequalities in American political campaigns are relics of the past. Far from it. As Lawless and Richard Fox have written about extensively, women will remain underrepresented until more of them run for political office. And the barriers to increasing political ambition among women and encouraging their entrance into the electoral arena are formidable.
But once they decide to run for office, the terrain female candidates navigate seems to be more level than we might have previously thought. Indeed, the findings from our study of 2010 suggest that one reason women did well in 2012 is that they likely did not face the kind of systematic gender stereotyping that has traditionally made it harder for female candidates to succeed. It might seem surprising that the media and voters aren’t to blame for women’s underrepresentation. But the data make it hard to argue otherwise.