Correcting Misperceptions at a Glance: Using Data Visualizations to Reduce Political Sectarianism

Douglas Markant, Wenwen Dou, Subham Sah, Alireza Karduni, Milad Rogha, My T. Thai2025

TL;DR

This study explores how different data visualizations of opposing political groups' views can reduce partisan misperceptions and, consequently, lower support for extreme political stances. They found that visualizing the full distribution of out-party views, not just the average, was most effective in correcting these misperceptions and improving recall, suggesting that how we visualize uncertainty matters.

What Problem Does This Paper Address?

Political sectarianism, a growing problem in many democracies, is often fueled by "false polarization." This phenomenon occurs when people overestimate how extreme their political opponents' views are and underestimate the diversity within those opposing groups. These misperceptions can lead to increased animosity, justify extreme actions by one's own party, and hinder compromise. While previous research has shown that correcting these misperceptions can reduce support for partisan violence and anti-democratic actions, there's a limited understanding of how the design of these correction interventions impacts their effectiveness. Specifically, most prior work has focused on presenting only the average view of the out-party, potentially obscuring important nuances about the range of opinions.

Methods

The researchers conducted an online experiment with U.S. participants who identified as either Democrats or Republicans. The study involved five steps:

  1. Political Identity: Participants stated their party affiliation.
  2. Prediction of Out-Party Views: Participants predicted how members of the opposing party would respond to questions about support for political violence and undemocratic practices.
  3. Visualization Conditions: Participants were randomly assigned to one of four groups:
  4. Control: Saw their own predictions but no data about the out-party.
  5. Mean-Only: Saw their predictions and the average response of the out-party.
  6. Mean+Interval: Saw their predictions, the average response, and an interval representing the middle 75% of out-party responses.
  7. Mean+Points: Saw their predictions, the average response, and a visualization of every individual out-party response (a dot plot).
  8. Self-Views: Participants reported their own level of support for the same political violence and undemocratic practices.
  9. Post-Test: Participants answered questions about their perceptions of the data (surprise, trust) and completed recall tasks to assess how accurately they remembered the out-party views presented.

Datasets

The study utilized baseline survey data from a previous study ([8]) which included responses from 218 Democrats and 174 Republicans collected in June 2022. This data was used to generate the visualizations presented to participants. No new datasets were introduced.

Key Insights and Ideas

  • Visualizing the Full Picture is Powerful: Simply showing the average view of the "other side" can be helpful, but visualizing the entire spectrum of their opinions (the Mean+Points condition) had the strongest effect in reducing participants' own support for extreme political actions.
  • Uncertainty Can Be Tricky: While the goal was to convey the range of views, the "Mean+Interval" condition, which showed the average plus a summarized range (75% of responses), was actually less effective than showing just the average. This suggests that how uncertainty is represented matters significantly.
  • Surprise Isn't Enough: Participants who were more surprised by the out-party data (meaning their initial guesses were way off) tended to report lower trust in the data. This indicates that simply confronting people with their misperceptions isn't always enough if they don't believe the corrective information.
  • Recall Matters: The visualization that showed the full distribution of responses (Mean+Points) also led to the best recall of the out-party data, suggesting that a more detailed visualization aids understanding and retention.

Key Results

  • Reduced Extremism: Compared to the control group, all correction conditions showed some reduction in participants' self-reported support for political violence and undemocratic practices.
  • Mean+Points Dominates: The Mean+Points condition (full distribution) showed the strongest reduction in support for extreme views.
  • Mean+Interval Underperforms: The Mean+Interval condition showed weaker correction effects and lower accuracy in post-test recall compared to other conditions.
  • Overestimation is Rampant: Participants significantly overestimated the support for political violence and undemocratic practices among the out-party. For example, Democrats overestimated support for partisan violence by an average of 40.5 points.
  • Recall Accuracy: Participants in the Mean+Points condition were most accurate in recalling the average out-party response and the proportion of support.

Limitations and Future Work

  • Generalizability: The authors acknowledge that their findings are based on U.S. political contexts and specific visualization types. Future work should explore how these findings generalize to other political divides or data types.
  • Visualization Design: The specific design of the "Mean+Interval" visualization might have contributed to its weaker performance. Further research could explore different ways to represent uncertainty.
  • Sensemaking: The study found that participants who doubted the data's credibility (due to large discrepancies with their beliefs) were less receptive to corrections. Future work should investigate how to better support participants' sensemaking process, perhaps through narrative framing, explanations for misperceptions (like media bias), or highlighting commonalities between parties.
  • Trust and Surprise: The interplay between surprise, trust, and attitude change needs further exploration. Understanding how to build trust in corrective information is crucial.

Summary updated

Abstract

Political sectarianism is fueled in part by misperceptions of political opponents: People commonly overestimate the support for extreme policies among members of the other party. These misperceptions inflame partisan animosity and may be used to justify extremism among one’s own party. Research suggests that correcting partisan misperceptions—by informing people about the actual views of outparty members—may reduce one’s own expressed support for political extremism, including partisan violence and anti-democratic actions. However, there remains a limited understanding of how the design of correction interventions drives these effects. The present study investigated how correction effects depend on different representations of outparty views communicated through data visualizations. Building on prior interventions that present the average outparty view, we consider the impact of visualizations that more fully convey the range of views among outparty members. We conducted an experiment with U.S.-based participants from Prolific (N=239 Democrats, N=244 Republicans). Participants made predictions about support for political violence and undemocratic practices among members of their political outparty. They were then presented with data from an earlier survey on the actual views of outparty members. Some participants viewed only the average response (Mean-Only condition), while other groups were shown visual representations of the range of views from 75% of the outparty (Mean+Interval condition) or the full distribution of responses (Mean+Points condition). Compared to a control group that was not informed about outparty views, we observed the strongest correction effects (i.e., lower support for political violence and undemocratic practices) among participants in the Mean-only and Mean+Points condition, while correction effects were weaker in the Mean+Interval condition. In addition, participants who observed the full distribution of out-party views (Mean+Points condition) were most accurate at later recalling the degree of support among the outparty. Our findings suggest that data visualizations can be an important tool for correcting pervasive distortions in beliefs about other groups. However, the way in which variability in outparty views is visualized can significantly shape how people interpret and respond to corrective information.
BibTeX
@misc{markant2025correcting,
  author = {Douglas Markant and Wenwen Dou and Subham Sah and Alireza Karduni and Milad Rogha and My T. Thai},
  title = {{Correcting Misperceptions at a Glance: Using Data Visualizations to Reduce Political Sectarianism}},
  year = {2025},
  eprint = {2508.00233},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2508.00233},
}
Figure