Article Summary
Introduction
While concerns over partisan polarization have been growing over the last decades, so too have concerns about whether increasing partisan animosities affect individuals’ beliefs in facts. For instance, many partisans appear to subscribe to factually incorrect statements, such as about climate change or voter fraud. Previous research has suggested two main explanations to account for this behavior. First, from a motivated-reasoning perspective, citizens accept false information if it supports their ideological preferences. Second, according to the partisan cheerleading approach, partisans only report subscribing to wrong information without actually believing in this information. This is because they want to show support for their own party and thereby recite information that party leaders and partisan media disseminate. The authors thus ask: Do partisans follow motivated reasoning or partisan cheerleading when answering factual information questions incorrectly?
Analytical Approach
The authors conducted an online experiment to investigate whether motivated reasoning or party cheerleading could account for misinformed responses. To achieve this, they randomly assigned respondents to one of two treatment conditions. In the first condition, researchers provided incentives for correct responses to factual information questions, offering $0.50 for each accurately chosen answer. In the second condition, respondents were not incentivized. A table outlining the treatment conditions is presented below. Before making their choices, participants were given the option to read news from politically-leaning, expert, or mainstream sources.
| Treatment condition | Incentive |
|---|---|
| Incentivized | $0.50 for correctly answered question |
| Not incentivized | No incentive |
If there is a difference in search behavior and correctly answered responses between treatment conditions, this pattern would support the cheerleading hypothesis. This is because partisans would have an exogenous incentive to report their correct beliefs in the survey instead of adhering to incorrect partisan information. If there were no differences between the conditions, the results would support the motivated reasoning hypothesis, as respondents would continue to report incorrect responses despite having an incentive to select the correct answer. The authors also validated their news consumption measure with real-world search data from a subset of respondents. These respondents had consented to the researchers analyzing their search history, allowing the researchers to determine which news sources respondents accessed.
Main Findings
The authors discover robust support for the motivated reasoning hypothesis: the incentivized and unincentivized treatment groups exhibit only marginal differences in their responses to the factual information questions. Likewise, the selection of news sources does not vary substantially between the treatment groups. The validation using real search data further demonstrates that the source selection within the survey experiment aligns with real-world search behavior, thereby enhancing the validity of the news selection measure.
Implications
The study’s findings imply that partisans do not exhibit misinformed behavior solely to demonstrate support for their party; rather, they genuinely believe in incorrect information. This suggests that polarization is a deeply ingrained phenomenon and not merely a strategic behavior displayed by partisans to publicly affirm their allegiance to their party. Consequently, providing incentives and rewards for accepting facts is unlikely to reduce polarization, as partisans appear genuinely convinced of misinformation.
Questions left unanswered
The authors raise the possibility that the reward for reporting correct information in the survey might still be too low. Future research could therefore explore whether stronger incentives reveal that some partisans engage in cheerleading rather than motivated reasoning.
Another avenue for future research involves examining the sequence in which partisans adopt factually incorrect information. Do partisans first select partisan-leaning news sources and, as a result, adopt factually wrong information? Alternatively, are partisans interpreting information from any source in light of pre-existing misinformed beliefs?
Methods and Analysis
Was the study and its analyses pre-registered?: No
Did the study rely on proxy variables to measure polarization?: Yes
The authors use responses to knowledge items and media selection behavior as outcome variables.
Were standard p-value thresholds used (p<.05 or 95% Confidence Intervals that don’t overlap zero)?: Yes
- Largest p-value presented as significant: 0.05
Were correlational results interpreted with causal language?: NA
Limitations / Weaknesses
While the authors transparently discuss the possibility that the monetary incentives chosen in the experiment might be too low for some partisans to refrain from party cheerleading, they could have explored other incentives to detect insincere reporting of misinformation. For instance, priming respondents about the potential drawbacks of misinformation for democratic stability might incentivize abandoning misinformed responses beyond materialistic rewards. Furthermore, it remains unclear how respondents reconcile ample evidence that their response to the information question is wrong with their misinformed behavior. Hence, future research may investigate how respondents justify their incorrect responses with factual arguments.
Open Data & Analyses
Does the article make the replication data publicly available?: Yes
Does the article make the replication analysis scripts publicly available?: Yes
Article Citation
Peterson, E., & Iyengar, S. (2021). Partisan Gaps in Political Information and Information-Seeking Behavior: Motivated Reasoning or Cheerleading? American Journal of Political Science, 65(1), 133–147.
Bibtex
1
2
3
4
5
6
7
8
9
10
Bibtex entry: @article{peterson2021,
title = {Partisan Gaps in Political Information and Information-Seeking Behavior: Motivated Reasoning or Cheerleading?},
shorttitle = {Partisan Gaps in Political Information and Information-Seeking Behavior},
author = {Peterson, Erik and Iyengar, Shanto},
date = {2021},
journaltitle = {American Journal of Political Science},
volume = {65},
number = {1},
pages = {133--147}
}