As America shifted left under Trump, emotional hostility widened the partisan gap

Over the course of Donald Trump’s first presidential term, American public opinion generally shifted in a more liberal direction, driven primarily by left-leaning and independent voters changing their minds on specific issues. At the same time, emotional animosity between the two major political parties intensified, a shift that appears connected to the widening gap in policy preferences. These findings were recently published in American Politics Research.

Political scientists often track changes in public opinion using a concept known as the public mood. This mood tends to operate like a thermostat. When a conservative administration is in power, the public typically reacts by moving to the left, and when a liberal administration takes over, the public usually shifts to the right.

Beneath these broad swings, there is also a phenomenon called partisan divergence. This refers to the growing gap between Democrats and Republicans on specific policy issues, such as health care, taxation, and immigration. Over the past few decades, the two parties have moved further apart on these topics.

This policy divide has been accompanied by a rise in affective partisanship, which is the emotional dislike or hostility that voters feel toward the opposing party. Theoretical models help explain how these factors interact. For instance, a 2024 study covered by PsyPost demonstrated that a stronger connection between political disagreement and personal dislike tends to drive people into two opposing ideological camps.

Understanding how public opinion shifts requires looking at individual voters. The authors of the new study note that a study covered by PsyPost in 2023 found that party identification remained remarkably stable between 2011 and 2020. This indicates that when the public mood changes, it is often because people who already identify with a party are changing their minds on issues, rather than simply switching their party allegiance.

Free daily newsletter

The research was conducted by Alan Yang, a senior lecturer at the School of International and Public Affairs at Columbia University, and Robert Y. Shapiro, a professor in Columbia’s Department of Political Science and School of International and Public Affairs. “We were interested in whether the election of an ‘untraditional’ Republican president in 2016 may have upended or accelerated past patterns of opinion change, partisan divergence, and ‘affective partisanship,'” the researchers told PsyPost.

Because Donald Trump’s platform did not always align with traditional conservatism, it was an open question whether the expected liberal reaction would occur. The authors wanted to see if partisan gaps continued to widen, and whether the emotional animosity of the era played a role in how individuals changed their minds on policy issues.

To explore these questions, the researchers analyzed data from the 2016-2020 American National Election Study (ANES) and the 2016-2020 General Social Survey (GSS). Both datasets are uniquely useful because they use panel designs, meaning they surveyed the exact same individuals at multiple points in time. This setup allows researchers to track whether an individual actually changed their answer from one election year to the next.

The team examined 32 specific policy questions from the ANES and 40 questions from the GSS. These questions covered a wide range of topics, including government spending, racial equality, immigration, foreign affairs, and social issues like abortion and gay rights. The researchers tracked how responses changed between 2016 and 2020 across the entire sample and within specific political groups.

To measure affective partisanship, the surveys used a feeling thermometer scale. Respondents were asked to rate their feelings toward Democrats and Republicans on a scale from 0 to 100 degrees, with higher numbers indicating warmer, more positive feelings.

The analysis indicates a noticeable overall shift in a liberal direction between 2016 and 2020 across many policy areas. For example, support for keeping or increasing welfare spending, providing more government services, and expanding immigration levels grew substantially. “We found an (expected) liberal rebound in aggregate opinion after the election of a Republican president, which was an open question given that Donald Trump was not a mainstream conservative Republican,” the researchers noted.

However, this overall shift masked a growing divide between the political parties. The partisan divergence continued to expand, and it did so asymmetrically. The liberal shift in the broader public mood was driven almost entirely by Democrats and Independents adopting more progressive positions on issues like government spending and racial equality.

In contrast, Republican opinions generally remained stable on these issues, or moved slightly to the right on specific topics like scientific research and alternative energy. The findings are in line with research covered by PsyPost in 2026, which found that widening issue divisions were largely driven by the left shifting in a more progressive direction. It is worth noting a slight measurement difference, as the previous study clustered respondents purely by their policy beliefs, whereas the new study measured divergence between self-identified partisans.

The emotional temperature of the country also shifted notably. Overall affective partisanship increased, but the most dramatic changes occurred among Republicans. Between 2016 and 2020, Democratic panelists maintained a steady level of warmth toward their own party and cooled by roughly four degrees toward Republicans.

Republican panelists, on the other hand, grew over 10 degrees warmer toward their fellow Republicans and cooled by about eight degrees toward Democrats. The gap between how much Republicans liked their own party versus the opposing party grew by nearly 19 degrees, which was a much larger jump than the three-degree increase observed among Democrats.

When the researchers looked at individual changes, they found that these short-term increases in emotional polarization predicted changes in actual policy preferences. For both Democrats and Republicans, individuals whose partisan animosity intensified were more likely to change their minds on specific high-profile issues. At the individual level, the researchers found that an “increased affective partisanship level was associated with Republican partisans moving rightward and Democratic partisans moving leftward across a number of highly salient policy preferences during the period of the first Trump presidency.”

For example, voters who developed stronger negative feelings toward the opposing party were more likely to adopt their own party’s typical stance on issues like immigration, the environment, and racial justice.

As with all research, there are a few things to keep in mind. The study looks at a relatively short four-year window that was dominated by a highly unusual political environment and the onset of the COVID-19 pandemic. It is hard to know whether the specific dynamics of the Trump presidency will continue into future election cycles.

The relationship between emotional animosity and policy opinions is also highly complex. The data shows an association between the two, but untangling which one causes the other is a challenge. People might develop stronger negative feelings toward the opposing party because they disagree on policy, or they might change their policy views because they want to distance themselves from a party they dislike. Both processes likely happen at the same time.

To see if these dynamics persisted after Trump left office, the authors plan to expand their timeline. “We see as a next step for us and others to extend this analysis to 2020 to 2024 using the available panel data for this period,” Yang and Shapiro said.

The study, “Individuals and Public Opinion Change: The Liberal Rebound, Continued Partisan Divergence and Affective Partisanship 2016–2020,” was authored by Alan Yang and Robert Y. Shapiro.