Not all stereotypes are equal: Consequences of partisan stereotypes on polarization

Download data and study materials from OSF

Principal investigators:

Ethan C. Busby

Brigham Young University



Adam J. Howat

Oberlin College



Jacob E. Rothschild

Reality Check Insights



Richard M. Shafranek

HIT Strategies



Sample size: 2018

Field period: 06/07/2018-12/17/2018

Politics in the United States has become tribal and personalized. Partisan conflict has grown to influence people’s everyday lives, not just the political realm. Supporters of the two major parties increasingly dislike each other on an emotional level, even in the absence of ideological or policy-based disagreements. Party affiliation has come to function as a social identity in its own right, comparable in its importance to race, religion, or gender. Many individuals can call to mind specific images of what typical partisans “are like,” and these stereotypes can have significant consequences for how the parties and their supporters interact. Our study investigates the impact of thinking about everyday partisans in three distinct ways—in terms of individual traits, group memberships, or issue priorities—on both affective and ideological polarization. We find that, relative to other kinds of stereotypes, a focus on issue-based stereotypes tends to de-polarize the electorate, with important implications for mitigating partisan conflict.
H1: Trait-based stereotypes will lead to greater partisan polarization relative to an emphasis on groups or issues.
H2: Issue-based stereotypes will lead to lower polarization compared to other kinds of stereotypes.
H3: Group-focused stereotypes will lead to more moderate amounts of polarization that fall between issue- and trait-based stereotypes.
Experimental Manipulations
Subjects were first asked to write two short paragraphs—one each for a typical Democratic and Republican Party supporter—describing such individuals in terms of “their personality and character traits,” “the political issues they find important,” or “the social groups they belong to.” To increase engagement with the treatments, participants then viewed a checklist of 15 items for each party, corresponding to the same stereotype domain as the writing task, and were asked to select which items described typical Democratic and Republican Party supporters. The order in which the parties were presented was randomized.

Affective polarization: The difference in feeling thermometer ratings for the Democratic Party and Republican Party.

Personal ideological polarization: A 7-point scale describing the respondent’s ideological views from extremely liberal to extremely conservative, folded into a 4-point scale where higher values indicate greater ideological extremity.

Perceived ideological distance between the parties: The difference in 7-point ideology ratings of “people who support the [Democratic/Republican] Party” and of “[Democrats/Republicans] in Congress.”

Partisan identity importance: a 5-point scale from “not important at all” to “extremely important” indicating the importance of the respondent’s party identification.

Preference for partisan social distance: The degree to which the subject agreed with the statement “You would be displeased if a family member dated someone who supports the [Democratic/Republican] Party” on a 5-point scale from “agree strongly” to “disagree strongly.” (Partisans were asked about the opposing party, independents about both parties.)

Summary of Results

Treatments showed no significant effects with respect to affective polarization. However, subjects in the traits condition proved more ideologically polarized than those in the issues condition; interestingly, a partisan breakdown of these results revealed significant effects among Democrats and independents but not Republicans. Moreover, Democrats in the groups condition showed a greater degree of ideological polarization than those in the traits condition.

With respect to perceived polarization, participants in the issues condition perceived less ideological distance between both rank-and-file and congressional partisans. Those in the traits condition perceived greater polarization than those in the control group.

Subjects in the issues condition also expressed less desire for social distance from outpartisans. They also rated their partisan social identity as less important.

Ethan C. Busby, Adam J. Howat, Jacob E. Rothschild, and Richard M. Shafranek. The Partisan Next Door: Stereotypes of Party Supporters and Consequences for Polarization in America. Forthcoming, Cambridge Elements Series in American Politics.