In a primary election, if voters believe it is too difficult or impossible for a female candidate to win a general election, they will instead support a male candidate from their party – even if they personally preferred a female, according to a new study from Stanford. .
Fortunately, there is a way to counter the assumption that women are less eligible than men, researchers report in a paper published February 1 in the newspaper Proceedings of the National Academy of Sciences.
When voters are presented with evidence showing that political women candidates garner as much support as men in the U.S. general election, voters’ intentions to support female presidential candidates increased by about 3 percentage points, according to data from the researchers.
“Although the effects of our short intervention were small, primary elections are often decided by just a few percentage points,” said Robb Willerlead author of the paper and professor of sociology at Stanford’s School of Humanities and Sciences.
As the 2022 primary elections approach and many women prepare to run for important positions in government, the researchers’ findings are particularly applicable, as this is where voters are most concerned about the possibility of winning candidates, Willer said.
The “electability trap”
To better understand voters’ perceptions of eligibility and how it may undermine female candidates at the polls, researchers collaborated with women’s leadership organization LeanIn.org to conduct a survey during the 2020 Democratic presidential primary election.
The Democratic primaries provided a unique opportunity for academic study: never before in a U.S. presidential election had there been multiple competitive candidates who were women, said Christianne CorbettPhD student in sociology and one of the main co-authors of the article.
“In most previous US presidential primaries, all candidates have been men. When there were competitive female candidates, there was only one, so it was very difficult to disentangle the gender from all the other characteristics of the specific candidates,” Corbett said.
Since there were multiple female candidates to study – Elizabeth Warren, Kamala Harris, Amy Klobuchar and Kirsten Gillibrand – the researchers were better able to gauge the impact of gender bias and what drove stereotypical voter assumptions.
In a survey administered to a sample of nearly 1,000 likely Democratic primary voters, researchers asked a series of questions such as “Do you think it will be harder or easier for a woman to win the 2020 election versus President Trump, compared to a man? ” and “If he were guaranteed to win, who would you most want to be the next president? and provided 12 choices that included all of the top contenders at the time of the survey.
They found that overall, 76% of respondents thought it would be harder for a woman to beat then-President Donald Trump in the general election.
Some of the reasons cited by voters for perceiving a female candidate as less eligible included the idea that Americans were not yet ready to elect a female president and that gender biases in society would hurt her campaign – such as being held to a higher standard than a man, face biased media coverage, and face harsher or more effective attacks from the opposition. In sum, many Democratic voters did not consider it practical for a woman to run if they wanted their party to win the election.
Additionally, the researchers also observed a clear pattern of voters who, although they may have personally preferred a female candidate as their party’s candidate, nevertheless changed their voting intentions in favor of a male candidate when ‘they perceived the female candidate as being less eligible.
Social scientists have long documented the personal biases faced by female political leaders. Here, the researchers identified another, more subtle barrier: the “pragmatic bias”. Pragmatic bias is different from personal bias in that it is based in part on what other people can think of a particular group and what a barrier that may be for members of that group.
The authors argue that pragmatic bias is not limited to gender alone and can extend to other characteristics, such as race.
An example of pragmatic bias emerged in the 2008 primaries when Barack Obama ran for President of the United States. At the time, many speculated – including michelle obama – if the United States was ready to elect a black man as president.
“That’s part of why the Obama campaign centered the ‘yes, we can’ idea in 2008. They wanted to convince Democratic Party primary voters that a black man could win the presidential election,” he said. Willer said.
These assumptions run the risk of turning into what researchers have called a “eligibility trap.” Assumptions about group behavior can become a self-fulfilling prophecy: if voters don’t support a candidate because they think other Americans won’t support the candidate, they have the power to make their assumption true and the candidate won’t win.
“One aspect of pragmatic bias that is particularly pernicious is that people who themselves are gender-neutral — and who may even be actively trying to promote the advancement of women — might still act against women at because of what they see around them. They can see that women candidates still struggle with gender bias and conclude that a woman won’t win and decide to vote for a man instead,” Corbett said.
Overcoming pragmatic biases
While women are still vastly underrepresented in politics, academic research has shown that in general elections, and U.S. congressional races in particular, female candidates receive roughly the same level of support as male candidates. The researchers wanted to know if any interventions could be implemented to reduce bias among voters and help them overcome their assumptions about eligibility.
In an experiment with likely Democratic primary voters, they found that simply reporting the percentage of Americans saying they were ready for a female president was not compelling enough.
What influenced people’s intentions to vote for a female candidate, however, was providing strong evidence, rooted in academic literature, showing that female and male candidates are equally eligible. Participants received a short 300-word summary describing how research shows that voters are as likely to vote for a female candidate as a male candidate in the US general election. (In the control group, participants received a generic essay about the 2020 election).
Providing behavioral evidence was key to making the intervention effective, said the paper’s co-lead author and sociology PhD student. Jan Volkel. “It might be more persuasive to tell people what other people are actually doing rather than what other people think,” he said. Voters not only have to believe that a majority of other voters want to see a woman candidate win, they also have to believe that a majority of voters will show up at the polls and vote for her, Voelkel stressed.
Furthermore, Willer also speculated that contrary perceptions are likely to be better received by voters if they come from a source other than the candidate herself.
Slow-down social progress
Pragmatic biases are seen by scholars as impeding social progress and perpetuating existing inequalities that many are trying to overcome.
“People base their idea of what’s possible on what they’ve seen happen in the past. Often that makes a lot of sense, but it can also be an overly conservative force, slowing down the social changes that most people would like,” Willer said.
“For women to have equal access to political leadership, it is important that voters update their idea of what is possible for women running for office. Until they do, women will face this extra barrier,” he added.
This research was supported by the VMWare Women’s Leadership Innovation Lab and funded by the Stanford Center on Philanthropy and Civil Society. The data for the first study presented in the PNAS article was collected in collaboration with LeanIn.org.