What the Research Says About the Perception Gap

What this note covers

This note summarizes, in plain language, a body of social-science research on two related phenomena: affective polarization, meaning the extent to which partisans dislike and distrust members of the other party, and the perception gap, meaning the difference between what partisans believe the other side thinks and what the other side actually thinks. It is written for readers who want the findings without the jargon. Where the research is uncertain, we say so. The sources we draw on are listed at the end of the piece.

Affective polarization: what has changed

Political scientists distinguish between ideological polarization — how far apart the parties are on policy — and affective polarization — how much partisans dislike each other. The two are related but not identical, and the second has grown more sharply than the first.

Survey measures going back several decades ask Americans to rate the parties on a "feeling thermometer" from zero to one hundred. Ratings of one's own party have stayed roughly stable. Ratings of the other party have fallen substantially. The gap between how partisans feel about their own side and the other side has widened considerably over the last several decades. The Pew Research Center's long-running work on partisan attitudes documents rising shares of both Democrats and Republicans who describe the other party not merely as wrong but as closed-minded, dishonest, or a threat to the nation's well-being.

Several explanations compete, and the honest answer is that they probably all contribute. The parties have become more sorted by ideology, religion, and geography, so that partisan identity now overlaps with other identities rather than cutting across them. Media environments allow people to consume mostly content that confirms what they already believe. Campaigns have found that negative messages about the opposition motivate turnout more reliably than positive messages about one's own side. And elite behavior — what politicians and commentators model — filters downward.

The perception gap: what each side gets wrong

A second line of research asks not how partisans feel about the other side but what they believe the other side believes. The findings here are, in some ways, more hopeful, because they suggest that a portion of affective polarization rests on misinformation that can, in principle, be corrected.

The "Perception Gap" study published by the research organization More in Common asked Democrats and Republicans to estimate what share of the other party held various views, and then compared those estimates to what members of the other party actually reported. Both sides overestimated the prevalence of extreme views on the other side by wide margins. Democrats believed that far more Republicans held hard-line positions on immigration and race than actually did. Republicans believed that far more Democrats held hard-line positions on policing and patriotism than actually did. In many cases the actual share holding the "extreme" view was well under half of what the other side estimated.

Two findings from that study are worth dwelling on. First, the perception gap was larger among the most politically engaged and among those who consumed the most news — not smaller. Following politics closely, in the current media environment, appears to make one's picture of the other side less accurate rather than more. Second, the gap was smaller among people who reported having friends or family in the other party. Personal contact corrected what media consumption distorted.

Related academic work has found similar patterns using different methods. Partisans overestimate the share of the other party that belongs to stereotyped demographic groups, overestimate how much the other side dislikes them, and overestimate how willing the other side is to endorse rule-breaking. Review articles in the political science literature summarize this work and its limits.

What this does and does not mean

It would be a mistake to read these findings as showing that Americans do not really disagree. They do. Ideological polarization is real, the parties have moved apart on important questions, and no correction of misperceptions will make those disagreements vanish.

It would also be a mistake to conclude that the gap is symmetric in every respect or that the two sides are identical. The research measures misperception, not truth; it does not tell you which side's policies are better. And the studies have limitations: they rely on self-report, they measure attitudes at a moment in time, and the questions chosen shape the results.

What the research does support is narrower and, for a coalition like this one, more useful. A meaningful portion of the dislike partisans feel for each other rests on beliefs about the other side that are measurably false. The people most confident in their picture of the opposition are, on average, the least accurate. And the single most reliable corrective identified so far is not more information but more contact — knowing, personally, someone on the other side.

The most confident partisans are, on average, the least accurate about the people they oppose.

Open questions

Researchers continue to debate how durable the effects of correcting misperceptions are, whether interventions that work in a survey setting work in daily life, and whether reducing affective polarization changes political behavior or merely changes reported attitudes. We will update this note as the evidence develops.

About the sources

The sources listed with this note include the Pew Research Center's ongoing work on partisan attitudes, the More in Common Perception Gap report, and the Annual Reviews series in which the leading academic review of affective polarization appeared. One entry is a placeholder for additional peer-reviewed work and should be replaced with a verified citation before publication.