Study finds partisan hostility drives protest participation in the United States

A new study published in Social Forces suggests that in the United States, people are more likely to protest not simply because they identify with a political party, but because they feel anger or threat toward the opposing one. Drawing from three large survey-based studies conducted between 2014 and 2022, political scientist Seth Warner found that partisan animosity—feelings of hostility toward the other party—consistently predicted whether someone participated in a protest. In contrast, merely identifying strongly with a party had little or even negative effects on protest participation.

In the past two decades, protest activity has grown significantly across developed democracies, and the United States has seen some of the largest nonviolent mobilizations in its history. The reasons why some individuals take to the streets while others do not has long interested political scientists and sociologists, but explanations have often centered on digital organizing, ideological commitments, or shared group identities.

Warner was particularly intrigued by the growing overlap between protest movements and partisan politics in the U.S. Though many demonstrations are not explicitly organized by political parties, they tend to align with party-affiliated agendas. For instance, the 2020 Black Lives Matter protests were often perceived as supportive of the Democratic Party, while Tea Party protests aligned with Republican priorities.

Warner observed that even when protests are framed in terms of social justice or policy concerns, many participants appear to be motivated by their feelings toward political opponents. This raised the question: is it loyalty to one’s party that drives protest behavior, or is it antagonism toward the opposing side?

“The idea for the study came to mind even before I went to graduate school, in 2017. The first big protest event of the Trump presidency was the Women’s March, which occurred the day after President Trump’s first inauguration,” said Warner, an assistant professor at the University of Connecticut.

Free daily newsletter

“I remember thinking that even though the protest was ‘about’ women’s issues, a lot of people were probably there to protest Donald Trump more generally. Then, when I began working on this study for my dissertation, I noticed a lot of conservative protests as well that seemed to be more partisan than anything. People’s signs expressed anger at Joe Biden, or support of Donald Trump, but did not have a lot of policy substance otherwise.”

Warner conducted three empirical studies, each designed to assess the influence of partisan identity and partisan animosity on protest participation. These studies used different datasets and methodologies but were unified by the same goal: to test whether identifying with a political party or harboring negative feelings toward the other side better explained participation in protests.

Study 1 used data from the American National Election Study (ANES) panel, tracking 2,839 respondents from 2016 to 2020. Warner examined whether levels of partisan identity and animosity in 2016 could predict protest participation four years later. He accounted for prior protest activity and a range of other factors, including ideological extremism, news interest, and demographic characteristics.

In this study, partisan animosity was measured by how coldly respondents rated the opposing party on a 0 to 100 scale. Strength of partisan identity was based on how strongly they aligned with their own party. The analysis revealed that partisan animosity significantly predicted protest participation in 2020, even after controlling for earlier behavior. By contrast, strong party identification was associated with a lower likelihood of protest participation.

Study 2 looked more closely at specific protest movements. Drawing on data from the Pew Research Center’s American Trends Panel, Warner analyzed participation in protests tied to the Black Lives Matter movement, the climate movement, and the Tea Party. Each dataset included information about participants’ views on the protest’s central issues, their partisan identity, and their feelings toward the other party.

Here again, Warner found that animosity toward the opposing party was a robust predictor of participation. In the case of Black Lives Matter and Tea Party protests, feelings of hostility toward the opposing party accounted for about half as much variation in protest behavior as policy alignment. For climate protests, partisan animosity was even more predictive than issue concern. Meanwhile, being a strong Democrat or Republican had little influence on whether someone participated in a protest related to their party’s general platform.

Study 3 explored how the political atmosphere in a respondent’s local community influenced protest behavior. Using data from the Cooperative Election Study and a method known as multilevel regression with poststratification, Warner estimated the level of partisan identity and animosity across U.S. counties. He then tested whether living in a county with higher levels of animosity or partisanship made people more likely to protest.

The results indicated that local environments shaped by high levels of partisan animosity increased the likelihood of protest participation. Interestingly, it was not just the animosity expressed by one’s own party that made a difference. Animosity from the opposing party—what people perceived as hostility directed at them—was especially influential. Warner found that living in a high-animosity county increased a person’s odds of protesting by nearly 50 percent. The presence of many like-minded partisans in the area also slightly increased protest participation, though to a lesser extent.

“Partisan polarization is a big reason why we see so many protests today,” Warner told PsyPost. “The number of protests has gone up a lot in recent years, and even though many of them touch on perfectly valid policy concerns (like racial inequality or the extent of COVID-19 restrictions), polarization is part of what has led so many people to take to the streets.”

While Warner’s findings are supported across multiple studies, they come with some limitations. The research relies on self-reported survey data, which can be subject to memory errors or social desirability bias. Participants may have misremembered or misrepresented their involvement in protests, and the studies may not fully capture the wide range of protest experiences or motivations behind participation.

The data also reflect a uniquely American political context, shaped by a polarized two-party system. In countries with multi-party systems or different protest traditions, the relationship between identity, animosity, and protest may look different. Warner encourages future research to examine whether similar dynamics exist in contexts where hostility is directed at other groups, such as far-right factions or political elites. His emphasis on anger as a driving emotion also leaves open questions about how other feelings—like fear, hope, or sadness—might influence political behavior in different ways.

The study, “Partisan animosity and protest participation in the United States,” was published May 30, 2025.