In this 2005 frame from a video that was recently released, Donald Trump prepares for an appearance on “Days of Our Lives” with actress Arianne Zucker (center). He is accompanied to the set by “Access Hollywood” host Billy Bush. (Obtained by The Washington Post)

In the past two weeks, several women have publicly accused Republican presidential candidate Donald Trump of groping and forcing himself on them, behavior he had described on a hot mic in 2005. Polls now reveal the largest gender gap in support for Trump’s Democratic opponent ever recorded at the presidential level. Moreover, news outlets have documented sexist paraphernalia for sale at Trump rallies around the country.

This raises a key question: How much do attitudes about gender and women affect attitudes toward Donald Trump? Our research shows that these attitudes do matter — over and above factors that others have widely noted, such as authoritarianism, ethnocentrism and anxiety about economic stagnation. Moreover, the anger so visible in this emotionally charged campaign may be helping to make sexism more of a political force.

The impact of sexism on Trump support predates the famous tape

In June 2016, we conducted a nationally representative survey of 700 U.S. citizens. They were asked whether they agreed with statements such as “Most women interpret innocent remarks or acts as being sexist” and “Many women are actually seeking special favors, such as hiring policies that favor them over men, under the guise of asking for equality.” An index based on these statements is widely used in social science research on sexism and gender attitudes.

We found that sexism was strongly and significantly correlated with support for Trump, even after accounting for party identification, ideology, authoritarianism and ethnocentrism. In fact, the impact of sexism was equivalent to the impact of ethnocentrism and much larger than the impact of authoritarianism. Again, this was in June — well before the “Access Hollywood” tape was released and several women came forward to accuse Trump of unwanted touching or kissing.

How anger — not fear — makes sexism more important

The election has also been one of the most emotionally charged in recent memory. A significant amount of conventional wisdom suggests that fear is a special catalyst of support for Trump. Interestingly, however, recent research suggests that fear will often dampen rather than boost the impact of attitudes like sexism, ethnocentrism and authoritarianism.

Anger, on the other hand, may have very different political consequences. It emerges when out-groups are seen as violating long-held norms and disrupting preexisting social hierarchies. For this reason, it tends to make attitudes like sexism or ethnocentrism more important. Anger also has other important consequences distinct from fear: It can powerfully mobilize voters and lead them to take greater risks and reject attempts to correct their misperceptions or process new information.

This is exactly what our research shows regarding sexism and support for Trump. In a February 2016 experiment, we first asked random subsets of respondents to think of a time in their life when they were either scared, angry or relaxed. This is called an emotion induction manipulation, and it causes respondents to feel these emotions keenly.

After priming these emotions, we asked individuals how much they supported Donald Trump. Among respondents who were primed to feel afraid, the impact of sexism on support for Trump was smaller, compared with respondents primed to feel angry or relaxed. In contrast, among respondents primed to feel angry, the impact of sexism was slightly larger than those primed to feel relaxed.


Many political observers have assumed that fear — of changing demographics and declining economic conditions — are motivating support for Trump, especially among those with less favorable views of certain groups. But our research suggests that the role of racial prejudice or sexism may be catalyzed more by anger.

Is this unique to Trump?

Trump is not the only candidate whose support is linked to attitudes about women. In our June survey, we found that respondents scoring higher in sexism were, unsurprisingly, significantly less supportive of Hillary Clinton.

Moreover, other research has shown that sexism is associated with support for many political figures in both parties — including, in 2012, Barack Obama (negatively) and Mitt Romney (positively). Although partisanship often matters more than attitudes about gender in determining preferences in general election matchups between Democratic and Republican candidates, our and others’ research suggests that sexism may be baked into how Americans view the political parties.

For that reason, a dominant narrative in this election — about “fearful authoritarians” who are resentful of immigrants and other minorities — misses out on a crucial dynamic of this campaign. And in a campaign that promises to feature even more angry rhetoric in its closing weeks, it seems likely that sexism will continue to affect how Americans see both Donald Trump and Hillary Clinton.

Carly Wayne is a doctoral candidate in political science at the University of Michigan. Nicholas Valentino is a professor of political science at the University of Michigan. Marzia Oceno is a graduate student in political science at the University of Michigan.