The gender gap in political representation in this country is not exactly news. From the national to the local level, the proportion of women in leadership positions is depressingly low: Women occupy just 18.2 percent of seats in the House of Representatives and 20 percent of the seats in the Senate. Only 12 of the hundred largest cities in America are governed by female mayors. According to calculations cited by The Nation, it will be another 500 years before women achieve equal representation in government.
The question of why the disparity is so persistent can feel like a chicken-and-egg debate. Are women less inclined to run for office because they believe they’re unlikely to win? Are they disengaged from politics because they feel under-represented? New research may offer a clue into breaking the cycle. In a paper in the Journal of Politics, Arizona State University political scientists Kim Fridkin and Patrick Kenney find that when women are represented by a female senator, they become significantly more informed and active in politics.
Fridkin and Kenney analyzed data from the 36,500 respondents to the nationwide Cooperative Congressional Election Survey in 2006, to see how male and female respondents’ political knowledge and engagement shifted when a female senator was elected. Being represented by a female senator, they predicted, would increase citizens’ political engagement.
According to the “novelty hypothesis,” both men and women should know more about female than male senators. In 2006, when the data was collected, only 14 senators were female. “When individuals are confronted with a politician who does not fit the prototype … the politician is more noteworthy,” write Fridkin and Kenney. “Similarly, a woman senator is more likely to ‘stand out’ because the woman senator looks quite different than the majority of her male colleagues in the chamber.”
Another hypothesis—“saliency of self hypothesis”—holds that only women would be better informed about female senators. Women in the electorate may have a reasonable expectation that female politicians will be more sympathetic to policies that are important to women: “Women senators are much more likely to have lived similar lives,” write the authors, “producing similar attitude toward politics and policy with women citizens.” And women might just naturally be more attuned to other women: “Information that is highly self-descriptive … is processed more quickly and with greater emotion,” they write.
Fridkin and Kenney also asked respondents to try to identify their senators’ party affiliation and asked them how they voted on several issues, such as immigration reform and the withdrawal of troops from Iraq. (About 34 percent of participants were able to identify their senators’ party affiliation.) Overall, women were less likely than men to answer these questions correctly—but when they were asked about female senators, the number of incorrect answers decreased. After adjusting for factors like age and education, they found that women represented by female senators were not just better-informed than women represented only by male senators, but more likely to be politically engaged in other ways, too—like voting, donating money to a candidate, registering with a political party, or trying to convince someone else to vote for a certain candidate. Men’s political activity, meanwhile, did not increase when they were represented by a female senator. “People are more interested in people who are like them,” says Jeffrey Koch, a political scientist at SUNY-Geneso who has also published papers on the gender gap in politics. “I’ve never seen this studied in relation to race, but I can imagine it.”
This is not the first study to examine the effect of female representation on women in the electorate, though it’s one of the largest. The literature on this topic is “somewhat inconsistent,” says Kathleen Dolan, a political scientist at the University of Wisconsin-Milwaukee who has studied similar issues. “It makes intuitive sense—that women would pay attention, be more motivated when they have a woman senator—but it’s not clear-cut.” One possible explanation for the increase in engagement, she points out, is that the effect is “driven by the fact that women senators tend to represent relatively affluent, liberal states. There could be something different about the people in those states.”
“This is certainly plausible to me,” said Kay Schlozman, co-author of The Private Roots of Public Action: Gender, Equality, and Political Participation (2001). Schlozman’s research examined the effect on women of being represented by a woman in the House, as a senator or as a governor. When she analyzed data from 1975 to 2013, she found a similar effect as Fridkin and Kenney—women were better informed when represented by a woman—but she found that the effect peaked in the 1990s, and basically disappeared after 2000. “The effect was concentrated at a time when women were rapidly increasing their proportion in the House and Senate,” she said. “The ‘Year of the Woman’ in 1992 was critical. Women began to think that politics was for them.” One potential limitation of Fridkin and Kenney’s study is their exclusive focus on women senators: “It’s not clear why key is the Senate as opposed to the House or the governor,” says Schlozman. “I’d want to be slightly cautious about the meaning of those findings.”
One worrying finding, though, is consistent across the literature. In study after study, men significantly outperform women on tests of political knowledge. Dolan suspects that this may have to do with the types of questions pollsters and sociologists ask: They tend to focus on national and international politics, rather than local and state-level issues where women might have an edge. Other researchers think it has to do with the way women answer questions. When asked open-ended questions about politics, they’re more likely to say they don’t know and refuse to answer, whereas men may be rewarded for guessing.
All of which makes Rebecca Traister’s argument that feminists had better find a woman to elect president, pronto, seem even more urgent.