Political Polarization: Beliefs vs. Values

Last registered on August 11, 2025
View Trial History

Pre-Trial

Trial Information

General Information

Title
Political Polarization: Beliefs vs. Values
RCT ID
AEARCTR-0016538
Initial registration date
August 08, 2025

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
August 11, 2025, 10:08 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Primary Investigator

Name
Kadeem Noray
Affiliation
Harvard University
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Savannah Noray
PI Affiliation
Harvard Kennedy School
Contact Investigator
PI Name
Stephen Zhu
PI Affiliation

Additional Trial Information

Status
In development
Start date
2025-08-08
End date
2026-12-31
Keywords
Behavior, Governance, Other
Additional Keywords
political polarization
JEL code(s)
D79, D83, D91
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
This project aims to understand the underlying determinants of political disagreements and whether misperceptions about the true determinants contribute to political polarization. Specifically, we plan to document whether political disagreements are rooted in fundamental values (e.g., valuing equality of opportunity) or beliefs about the consequences of policies (e.g., believing that a policy is effective in achieving equality of opportunity). To do so, we will ask respondents whether they agree with a set of value statements related to policies and then ask them to predict responses from individuals in the opposite political party. Then, to a randomly selected subset of participants, we will reveal the truth about their political opponent’s values and test whether doing so leads them to exhibit more warmth towards their political opponents.
External Link(s)

Registration Citation

Citation
Noray, Savannah , Kadeem Noray and Stephen Zhu. 2025. "Political Polarization: Beliefs vs. Values." AEA RCT Registry. August 11. https://doi.org/10.1257/rct.16538-1.0
Sponsors & Partners
Experimental Details

Interventions

Intervention(s)
Intervention Start Date
2025-08-09
Intervention End Date
2025-08-15

Primary Outcomes

Primary Outcomes (end points)
For descriptive analysis, we collect the following three types of variables from each participant:

Indicators of agreement with various political/policy values (i.e. the Value Statements referenced in the Intervention description)

(Mis)perceptions of whether others with the same or differing political affiliations agree with those political/policy values

Indicators of support for a range of public policies

As outcomes for our experiment, we collect four indicators of warmth towards the opposite party:

(a) The first outcome measures respondents’ willingness to direct a donation to a bipartisan organization rather than one aligned with their own political party. Specifically, we inform respondents that we plan to donate $50 to a left-leaning, right-leaning, or bipartisan organization, and that the recipient of the donation will be determined by randomly selecting one respondent’s choice.

(b) The second is one’s stated warmth (on a scale 0-100) towards people in the opposite party,

(c) The third is one’s indication of interest in signing up for a newsletter from Braver Angel’s, an organization devoted to bridging political divides and reducing political polarization.

(d) The fourth is one’s willingness to donate part of a $500 lottery prize to Braver Angels. Specifically, we inform the respondents that one respondent will be randomly selected to receive a $500 lottery prize. They are asked if, in the event that they won, how much of their prize ($0-500) they would like us to donate to Braver Angels.
Primary Outcomes (explanation)
Descriptively, will use measures of value-agreement and perceptions to measure (1) the overlap in value-agreement between Republicans and Democrats and (2) the accuracy of predictions they make about each other and their own party members. Overlap will be measured as the distance between average Republican and Democrat agreement rates by policy (and aggregated over policies). Misperceptions will be measured as the distance between mean value agreement of a party and average perceptions of mean value agreement by the same or the opposing party.

We examine these outcomes both individually and as part of an index, constructed by first averaging the four standardized variables (each with mean 0 and standard deviation 1), and then re-standardizing the resulting composite measure.

Secondary Outcomes

Secondary Outcomes (end points)
We may also explore secondary outcomes, such as the propensity for the respondent to predict a high willingness of people in the opposite party to switch their view if the policy had (or didn’t have) certain effects.
Secondary Outcomes (explanation)
After the information treatment, we also ask respondents to report what upsides/downsides they believe to be true about a policy (e.g. the policy would achieve its goal, have some negative side effect, etc.). Then, we ask whether knowing that the upside/downside were certainly true/false would cause them to switch their policy stance. Finally, we ask them to predict how someone with the opposing policy stance would answer these same sets of questions (they are paid bonus payments for accuracy, which we evaluate by comparing to answers from the pre-survey).

Experimental Design

Experimental Design

We will randomly select a subset of respondents to receive an information treatment. The control group receives no information.
Experimental Design Details
Not available
Randomization Method
Randomization is done by the survey software, Qualtrics.
Randomization Unit
We randomize individuals into either treatment or control.
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
1000 in Full Survey and 500 in Pre-Survey. Both surveys are evenly split between Republican and Democrat.

Sample size: planned number of observations
1000 in Full Survey and 500 in Pre-Survey. Both surveys are evenly split between Republican and Democrat.
Sample size (or number of clusters) by treatment arms
(Treatment is only applicable to the Full Survey.) 500 individuals from the Full Survey will be treated, and 500 will be in the control group.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
Harvard University-Area Committee on the Use of Human Subjects
IRB Approval Date
2025-07-10
IRB Approval Number
IRB00000109
Analysis Plan