Democratic presidential candidate Sen. Elizabeth Warren (D-MA), speaks during a house party in Windham, New Hampshire, on June 14. | Christopher Evans/MediaNews Group/Boston Herald via Getty Images
The complicated gender politics behind “likability,” explained.
New Hampshire voters approve of the job Sen. Elizabeth Warren is doing; they just don’t like her all that much. Same goes for Sen. Kamala Harris.
Despite Harris’s recent bump in New Hampshire following her performance at the first Democratic debate, data in a recently released CNN/UNH Survey Center poll of likely New Hampshire voters found good favorability numbers for both Warren and Harris (67 percent for Warren, 54 percent for Harris). But when pollsters asked, “Which Democratic candidate do you think is most likable?” the numbers for both women were bleak.
Just 4 percent of likely voters polled said they found Warren “likable,” and 5 percent said the same about Harris. The candidates they liked better were all men: 20 percent found both Vice President Joe Biden and Sen. Bernie Sanders likable, while 18 percent found South Bend, Indiana, Mayor Pete Buttigieg likable. (Warren’s favorability rating was the same as Sanders’s and higher than either Biden’s or Buttigieg’s.)
Likability is a tricky, highly subjective political term. Pollsters used to get at the same question by asking, “Who is the candidate you’d rather grab a beer with?” But the question of who voters think is the most likable is difficult to pin down because different people have vastly different ideas of what it means, according to Kathleen Hall Jamieson, director of the Annenberg Public Policy Center at the University of Pennsylvania.
“The construct is an unclear construct,” she told Vox. “We are making it relevant without asking why should it be. We don’t know what it is, anyway.”
One thing is clear: Likability applies differently to male and female candidates. But female candidates need to be liked in order to be elected, research has found.
“This likability dimension is a real barrier for women,” Democratic pollster Celinda Lake told Vox. “Voters are perfectly willing to vote for a man for executive office that they think is qualified that they don’t like, but they are not willing to vote for a woman they think is qualified that they don’t like.”
Rather than questioning the experience or qualifications of female presidential candidates, pollsters, pundits, and voters are now homing in on a much more subjective question.
“Nobody questions whether these women are qualified, so now they’re on to likability,” Lake said.
Likability is hard to separate from gender
Voters must weigh numerous factors decide whether they’ll vote for a candidate: experience, competence, political ideology, policy positions. But research shows the vast majority of voters also want to like the candidate they vote for.
Recent research also shows that voters both put a higher importance on women candidates to be likable and hold them to a different standard on what it means.
“I liken it to walking a tightrope because the recommendations are so subjective,” said Amanda Hunter, research and communications director for the Barbara Lee Family Foundation. “Use humor but not too much humor, dress well but not too opulently. So much of it is based on someone’s actual opinion.”
The Barbara Lee Family Foundation’s research showed that 84 percent of men and 90 percent of women said liking a candidate is an important factor in deciding to vote for them. Beyond the gender breakdown, people with strong partisan views, less educated voters, Latinas, and Southern women are the most likely to say likability is important to them.
But there are a lot of complex, gendered things wrapped up in that question, including a woman’s tone of voice, her appearance, and her demeanor. Voters of both genders judge women candidates more heavily on all these things. And women candidates have to strike an incredibly tricky balance in the public eye: being confident without being boastful, being stylish without being either too garish or too frumpy, and appearing strong without being overbearing. Women pollsters and political researchers told Vox it’s an exhausting list of requirements to meet.
“It comes down to maybe internalized sexism for people,” Hunter said. “A lot of voters have a stereotype for how women should act.”
Lake remembered how difficult it was for the first wave of women governors in the 1990s, some of whom shied away from debates with their male opponents because they saw their poll numbers go down after participating.
“Debates are an arena where people want to see are you tough enough,” Lake said. “But when women would be tough in debates, their likability would go down.”
As Vox’s Li Zhou wrote, the conversation around electability that’s been dominating the 2020 race so far is also wrapped up in gender politics. That’s in large part because there’s never been a woman president before:
More often than not, however, the expectation of who can win is inextricably wrapped up in the knowledge of who has won. What that feeling looks like for each voter could well be influenced by the kinds of candidates voters have seen win races before. And in the case of the presidency, that mold consists overwhelmingly of older white men, a precedent that could hurt candidates who don’t fit those characteristics.
There’s an interesting recent contradiction here: The central theme of the 2018 midterms was that Democratic female candidates overperformed their male counterparts by about 15 percent in primaries, the Cook Political Report’s Dave Wasserman found. So how do we square the current conversation we’re having about electability in 2020 with the Year of the Woman in 2018, when Democratic women were elected to the US House in droves?
The biggest difference is most of the 2018 races that women won were for legislative seats, not executive ones (although a few women governors were elected as well). The presidency remains the biggest glass ceiling for women.
“Most of the gains in 2018 were legislative positions, but women face a much higher climb when they seek executive office,” Hunter said.
The New Hampshire poll demonstrates why likability is highly subjective
The concept of likability is confusing. Case in point: Andy Smith, the UNH Survey Center political science professor and pollster who oversaw the recent poll of New Hampshire voters, wondered aloud how Bernie Sanders reached the same percentage of likability as Biden. Sanders is widely seen as grumpy (although that’s a character trait his core supporters love).
“I can’t explain how Bernie Sanders is the most likable one,” Smith said. “Joe Biden has always been ‘slap him on the back.’ Buttigieg has emphasized [likability] ... if you read about his candidacy and his campaign, that is emphasized.”
Warren’s single-digit likability number stood in stark contrast to her overall favorability numbers (favorability is more of a generic polling question). Sixty-seven percent of voters said they had a favorable opinion of her, while 18 percent said unfavorable (Sanders had the same number). Meanwhile, 57 percent of voters said they had a favorable opinion of Biden, while 25 percent said unfavorable. Harris had a 54/14 percent favorable/unfavorable rating, while Buttigieg had 48/11 percent.
Warren has gone out of her way to connect one on one with voters. She stays at the end of every campaign event to take selfies with whoever in the audience wants one and has become known for calling people who donate to her campaign to thank them personally.
But Smith said he reads Warren’s low likability number as a reflection of her campaign emphasizing policy ideas over being a likable candidate.
“I can’t recall reading in a newspaper article how likable Elizabeth Warren is,” Smith said. “I don’t think she’s ever run as a warm and fuzzy candidate. The things she is known for ... are the policy issues, the things that show up in other questions.”
Warren’s likability number was the only piece of bad news in an otherwise sterling poll for her. The poll found that while Biden is still at the top of the Democratic field in New Hampshire, with 24 percent of voters saying they’d support him for the nomination, Sanders and Warren appear to be closing in on the former vice president; they are tied for second, with 19 percent of voters supporting each.
Warren’s New Hampshire polling numbers have gotten a significant boost; she jumped from 5 percent support in April to her current 19 percent. That’s better news than for Sanders, whose support among New Hampshire voters fell in the same time period.
Warren is clearly on the rise in New Hampshire. The question is whether her likability numbers will hurt her, or if they will matter at all.