By Emily Reynolds
Politics in the UK is becoming increasingly diverse. But there is still a way to go. When it comes to gender, the proportion of women in the House of Commons is at an all time high — but at 35%, is still far from representative of the population.
A new study, published in PNAS, looks at the barriers to women being elected. And the Stanford University team finds even voters who would prefer a female candidate show a level of “pragmatic bias”: if they believe that women candidates face barriers that make them less electable, they are less likely to vote for them.
Participants in the first study were Democrat voters. First, they were asked whether they felt it would be easier or harder for a woman to win the 2020 election against Trump compared to a man, and how ready they felt the US population was for a woman president. Next, they assessed the electability of specific candidates: Elizabeth Warren, Kamala Harris, Joe Biden and Bernie Sanders. They were also asked about their personal preference for presidential candidate and who they would actually vote for in the Democratic primary.
Across the measures, participants consistently felt that women candidates were less electable: 76% believed it would be harder for a woman to win the 2020 election than a man, while only 18% believed Americans were extremely ready for a woman president. Half of the participants also believed that Harris and Warren were less electable than Biden and Sanders, while just 27% felt they were more electable.
Unsurprisingly, participants whose preferred candidate was a woman were more likely to say they’d vote for a woman candidate. But if they believed that male candidates were more likely to be elected, then many of these participants indicated that they’d shift their actual vote to a man. That is, concerns that women were less electable seemed to lead people to vote for men even when they really wanted to vote for a woman. A second study replicated these findings.
A third study found that simply giving people information about women’s electability was not enough to counteract this effect. Democratic primary voters were assigned to one of three conditions. In the true information condition, participants saw a correct figure showing that 52.5% of voters are very or extremely ready for a woman president; in the misperception condition, participants saw a figure showing only 15.7% are ready; and in the control condition, participants read general information about the election. They were then asked their likelihood of voting for a woman candidate. However, providing candidates with true information about the electorate’s willingness to elect a woman was not enough to overcome their pragmatic bias.
In the final studies, the team instead gave participants evidence that women are actually as successful as men in elections. They read that women are just as likely to win general elections as men, that they receive slightly more support than men, and that they are as successful as men at primaries. This time, the intervention increased their likelihood of intending to vote for a woman presidential candidate. And in a follow-up study, those who had seen this evidence were still more likely to say they’d vote for women candidates a month later, suggesting that sharing information about electability can increase voting intentions over a longer period of time.
Overall, the study suggests that a belief that women candidates are not electable ironically makes people less likely to vote for them — even voters who they profess to support these candidates. Future research could look at different levels of politics — does this stand for council elections, for example, or non-governmental elections, or was the executive level position explored here relevant? And would it apply to Republicans or those outside of the United States?
The team notes that in order to overcome pragmatic bias, it is not enough to believe others are simply ready for a female president but that “others will act on that readiness and vote accordingly”. Thinking and acting collectively could, then, shift attitudes and change representation.
Emily Reynolds is a staff writer at BPS Research Digest