Political scientists Eric R.A.N. Smith (UCSB) and Richard Fox surveyed undergraduates at UC Santa Barbara and the University of Wyoming and found that 1 in 10 respondents at the latter campus displayed an unmistakable bias against women candidates. According to Smith, the findingswhich will be published in the spring issue of the journal Political Psychologyhave broad policy implications.
"The fact is, it's tougher for women to get elected, and that has a direct impact on the legislation authored in Washington and in state capitals. If you change the sort of people in Congress you change the laws that come out of Congress, because politicians tend to be more sensitive to issues to which they can personally relate. For instance, breast cancer received a great deal more attention once women in Congress began pushing for additional research funding," he said.
Smith and Fox asked students in two large undergraduate courses, one each at UCSB and UW, to rate an assortment of fictional political candidates on thebasis of each candidate's positions on a variety of issues. Accompanying each candidate description was a fictional name.
When the researchers randomly switched the male and female first names accompanying the descriptions (from Carl to Karen or from John to Joan, for example), support for the female candidates eroded a full 10 percent among University of Wyoming respondents. (There was no discernible difference among UCSB respondents.) Smith says that while Wyoming is a more conservative state, there are other forces at work as well.
"Our research suggests that voters look at candidates and see stereotypes. They fill in missing information with stereotypes, which can lead them to make decisions. For instance, the assumption is that women are effective on such issues as education and daycare, but less so on crime, immigration, and military expenditures. Women are perceived as being too soft and unable to negotiate effectively as legislators," he said.
Smith is quick to point out that the study results don't necessarily describe the voting tendencies of Wyoming's entire population, or even those of UW's student body. Even so, he says the findings are significant, especially considering the age and background of the survey's respondents.
"College students are among the most tolerant people in the country. They're young, and the young are more tolerant than the old, and they're well-educated, which also typically lends itself to greater tolerance. If we look at older groups or groups that are less well-educated in general, we should expect to find larger biases."
Smith and Fox plan to conduct additional studies on other college campuses and analyze national survey data. They hope their research eventually yields abook or monographand new attitudes among voters.
"Information of this type might lead to more sensitivity. If voters realize they're biased on some level perhaps they'll be able to correct for it and go to the polls with more impartiality," said Smith, an associate professor of political science at UCSB.
"Unfortunately, there is not a lot of academic literature on this subject. The studies that have been done are very small and methodologically flawed. They support the idea that voters aren't biased against women candidates because they're based on public opinion surveys in which people were asked point-blank if they would vote against a qualified woman candidate for president. That's like asking people to admit that they're blatantly bigoted against women, and most voters aren't willing to do that. Besides, such questions completely skirt the issue of what constitutes a qualified candidate. We think our research is likely to show at least some systematic bias against women."