(Mis-)perceptions of politically motivated reasoning

How to cite

Yanay, Guy; Musolff, Robin. 2025. "(Mis-)perceptions of politically motivated reasoning." AEA RCT Registry. AEARCTR-0015498. https://doi.org/10.1257/rct.15498-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Guy Yanay* , Robin Musolff

* Primary investigator

Registry ID
AEARCTR-0015498
DOI
https://doi.org/10.1257/rct.15498-1.0
First registered
Last updated
Status
In development
Trial dates
2025-03-10 – 2025-06-30
Countries
United States of America ()
JEL codes
D9 D8
Registry record
http://www.socialscienceregistry.org/trials/15498

Abstract

Mass political polarization has been on the rise in The US since the 1990s. Beyond the general distrust people have for their political opponents, it seems a common conception that there is little agreement on basic facts.
This perception of disagreement is often attributed to biased news sources, but has also been attributed to politically motivated information processing. We look at how people revise their beliefs about politically charged factual questions when presented with feedback on their initial guesses.
We ask the following:
* Do people respond differently to information when it aligns/misaligns with their political opinions?
* If this asymmetry exists, does it differ by individual or issue characteristics?
* Are People's perceptions of these behaviors correct?

Experimental design

We conduct an online survey of 800 US Republicans and Democrats. Participants are presented with factual statistical questions and are asked to give their best estimates.
Following their initial guess, participants are told (for each question) whether the real statistic is lower or higher than their initial estimation.
They are then asked to give a revised estimate. A subset of participants estimate others' responses.

Design details

Randomization method
Participants in the first study are randomly assigned political and non-political (control) questions for estimation. For half of the political questions, a high answer is aligned with a Republican political stance, and for the other half, it is aligned with a Democrat political stance. The key randomization is thus whether a question is political or neutral, and for political questions whether a higher answer is aligned with a Republican or Democrat view. For the second and third studies, participants are asked to estimate the guesses of earlier participants which are Republican/Democrat and received pro-party/counter-party news. Whether the earlier participants are identified as Republicans or Democrats and whether they received pro-party or counter-party news is randomized.
Randomization unit
Random questions per participant
Clusters
800 participants
Observations
2800 actual belief updates 2000 expectations of updates (regarding others)
Arms
--
Primary outcomes
actual and perceived belief movement in response to information

Cited by

No citation data available for this registration yet.

Keywords