It’s axiomatic that in the 2016 presidential campaign, which pitted an infamous philanderer against the first woman nominated by a major party, voters’ views on gender played an unprecedented role. “Sexism powerfully predicted vote choice even after controlling for authoritarianism, partisanship, and other predispositions,” wrote the authors of a study published in Public Opinion Quarterly earlier this year. They looked at “the impact of sexism in recent presidential elections,” and found that “2016 was the only year in which it played a large and significant role.”

The #MeToo movement, which soon followed, may be an outpouring of pent-up anger and disgust after revelations of serial sexual abuse by high-profile men like Harvey Weinstein, but it’s also a backlash against the election of Donald Trump, a man who has been accused of various forms of sexual misconduct by almost two dozen women. An audio recording of Trump bragging about getting away with sexual assault because he’s famous may have helped him lose the popular vote by 3 million ballots, but it didn’t derail his candidacy.

With Donald Trump leading the GOP, it’s perhaps unsurprising that the debate over gender equality has become inexorably intertwined with partisanship. Tufts University political scientist Brian Schaffner presented a paper to the American Political Science Association at the end of August that suggests that these dynamics appear to have made Republicans more comfortable with expressions of what researchers call “hostile sexism.”

Schaffner tells me that during the 2016 campaign, he “felt like Trump was getting a pass for saying things that people might not otherwise be willing to stomach.” After the election, he became “interested in how his victory might have changed what people were willing to say after he had won.” He set out to answer both questions.

First, he conducted a straightforward experiment. He asked one group of respondents how they would react to an acquaintance who called a woman “a dog” and referred to his wife as “a beautiful piece of ass.” Schaffner then asked them to rate their response on a scale ranging from “very comfortable” to “very uncomfortable.” He asked the other group the same question, but this time he attributed the boorish statements to Donald Trump. (Trump has said both of these things.)

The results, when respondents were sorted by political party, were significant. Sixty-three percent of Republicans (and Republican-leaning independents) said those statements would make them very or somewhat uncomfortable if they came from an acquaintance, but only 39 percent said the same when they were attributed to Donald Trump.

Democrats and Dem-leaning independents were more likely to feel queasy about those statements when they were attributed to Trump than if they were attributed to an acquaintance, but the effect wasn’t statistically significant. Schaffner says that’s probably because they were more likely than Republicans to be uncomfortable with these kinds of statements regardless of who supposedly uttered them. (Long before Donald Trump came on the scene, the GOP had its share of rank misogynists like Todd Akin, the 2012 Missouri Senate candidate who blew up his campaign by claiming that women were unlikely to get pregnant as a result of a “legitimate rape.”)

Schaffner then set out to determine whether Trump’s victory had validated the sexist sentiments he expressed during the campaign. He had conducted a series of surveys with the same groups of respondents just before the election, in March of 2017, and then finally this past July. The survey used a panel of questions that’s common in such research. Designed to measure how much resentment people express toward women fighting for equality, it asks respondents whether they agree with statements like, “Many women are actually seeking special favors, such as hiring policies that favor them over men, under the guise of asking for ‘equality,’” “Women seek to gain power by getting control over men,” and “When women lose to men in a fair competition, they typically complain about being discriminated against.”

The surveys found that Republicans were significantly more likely to agree with these statements after Trump’s victory than they were prior to the election. Democrats were slightly less so, but again the shift on their side was small. Interestingly, there was a gap in these perceptions between Democratic men and women, but, Schaffner says, “Republican women score almost as high on this hostile-sexism measure as Republican men do.” Does that mean that partisanship trumps Republican women’s own lived experiences? “It’s not to say that gender or lived experiences aren’t important,” says Schaffner, “but a lot of that’s already baked into what party they identify with in the first place. The women who are still in the Republican Party at this point are perfectly fine with these kinds of sentiments.”

(In an earlier study, Schaffner and two colleagues found that there was a significant education gap seen in white voters’ expression of both sexism and racial bigotry. Non-college-educated whites—the most commonly cited Trump constituency—scored higher on both measures than better-educated whites.)

Schaffner says this is consistent with what researchers call the “justification-suppression model of prejudice.” In a nutshell, we all harbor some prejudices, but we tend to suppress them, either because we don’t want to see ourselves as being prejudiced, or because we don’t want to appear that way to others. But we also may take cues that tell us it’s all right to lower our guard and say the quiet parts out loud in certain situations. It’s why locker-room talk tends to be a lot more offensive than boardroom talk.

Schaffner found a similar dynamic with open expressions of racial animus earlier this year. White respondents who viewed Trump’s rant about Mexico sending us their “rapists” were more likely to express hostility toward all other groups than those who saw more traditional campaign messages. But that effect, unlike the one in this new study, was seen in both Trump and Clinton voters.

“Absent partisan politics, we might feel like maybe we shouldn’t say things that are so sexist or express views that are so sexist. But when sexism becomes viewed through a partisan lens, which I think is especially happening right now, right in this moment, then that sort of pushes in the other direction and says, well, my party is basically the party that’s expressing this hostility towards women and, therefore, by endorsing these [hostile] statements I am supporting my team,” says Schaffner.

And partisan passions lead people to engage in “motivated reasoning”—the tendency to reject information that conflicts with one’s strongly held views and accept that which reinforces them, even if that information is poor or from dubious sources. We’re seeing an excellent example of that dynamic right now, with conservatives responding to Brett Kavanaugh’s contentious confirmation to the Supreme Court by embracing the idea that men are at high risk of being falsely accused of assault. That story requires not only ignoring the fact that victims often face dire consequences for coming forward with their allegations—Christine Blasey Ford is still in hiding—but also dismissing studies showing that false accusations are rare.

Some degree of backlash against the #MeToo movement was probably inevitable, but partisan animosity tends to make people more entrenched in their views. But backlashes are cyclical, and while Schaffner didn’t find that partisanship had a significant effect on Democrats’ comfort expressing sexist attitudes, there’s plenty of evidence from other surveys that Republicans’ embrace of Trump’s misogyny is motivating Democratic-leaning women in a big way. A CNN poll released this week found a whopping 35-point gender gap on the generic congressional ballot among likely voters. That may be an outlier, but not by much—a Washington Post survey of 69 of the most competitive House districts found a 19-point gender gap and a Fox News poll taken conducted nationwide before the Kavanaugh hearings found a 21-point gap.

The upcoming midterms will be a test of which party’s base is more pissed off, and gender politics will play a big role. At present, Democrats appear to have an advantage.