Trust, Truth, and Selective Exposure to Partisan News

Last registered on November 06, 2018

Pre-Trial

Trial Information

General Information

Title
Trust, Truth, and Selective Exposure to Partisan News
RCT ID
AEARCTR-0002894
Initial registration date
April 13, 2018

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
April 16, 2018, 1:17 PM EDT

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

Last updated
November 06, 2018, 9:51 AM EST

Last updated is the most recent time when changes to the trial's registration were published.

Locations

Region

Primary Investigator

Affiliation
Harvard University

Other Primary Investigator(s)

PI Affiliation
New York University
PI Affiliation
Stanford University

Additional Trial Information

Status
Completed
Start date
2018-04-15
End date
2018-08-31
Secondary IDs
Abstract
Americans are increasingly exposed to news and information that confirms instead of challenges their pre-existing viewpoints. Why do we tend to demand ideologically aligned news? Theory and laboratory evidence suggest two broad groups of mechanisms that may play a key role. The first is psychological mechanisms leading to a “taste for consistency," which entails that partisans will be seek ideologically aligned sources even if this means choosing sources which they know to be less accurate. The second mechanism is rational updating that leads partisans to infer that sources which agree with their prior beliefs are likely to be more accurate. Distinguishing the relative importance of these two mechanisms is crucial for understanding how media consumption will evolve and what changes in policy or platforms will be most likely to lead to more accurate beliefs. We aim to use a field experiment to precisely identify the importance of trust vs. taste for consistency in generating selective exposure.
External Link(s)

Registration Citation

Citation
Allcott, Hunt, Matthew Gentzkow and David Y. Yang. 2018. "Trust, Truth, and Selective Exposure to Partisan News." AEA RCT Registry. November 06. https://doi.org/10.1257/rct.2894-4.0
Former Citation
Allcott, Hunt, Matthew Gentzkow and David Y. Yang. 2018. "Trust, Truth, and Selective Exposure to Partisan News." AEA RCT Registry. November 06. https://www.socialscienceregistry.org/trials/2894/history/36840
Experimental Details

Interventions

Intervention(s)
There are two main interventions in the experiment:
1. Incentive treatment: subjects randomly receive different amount of bonus payment if they make correct predictions. The bonus payment ranges from 0, $5, and $25. There is both a within and across subject component of the treatment variation.
2. Information treatment on media outlets' trustworthiness: subjects randomly receive information regarding how trustworthy the media outlets of interest are. The format of the information treatment is the truthful information on the percentage of study participants who could make correct predictions if they were randomly assigned to read news articles from a given outlet. Each treated subjects are informed about two outlets.
Intervention Start Date
2018-04-15
Intervention End Date
2018-08-31

Primary Outcomes

Primary Outcomes (end points)
The media outlet choices subjects make before they make predictions; there are two such choices. Specifically, an important outcome of interest based on these media outlet choices is whether the outlets chosen are aligned with subjects' political party affiliation and/or political ideology. In addition, we are interested in the heterogeneity of treatment effects along subjects' political party affiliation and/or political ideology.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
1. Number of articles subjects read from the outlet news feed, and which articles are read.
2. Total time spent on reading the articles from the outlet of choice.
3. Whether subjects make the correct predictions.
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
There are 7 parts of the experiment:
1. Introducing subjects to the experiment, and consent
2. Collecting demographic information and subjects' political positions
3. Asking subjects general questions regarding trust and readership on the media outlets of interest
4. Introducing subjects to prediction questions, and randomly assign subjects to read articles from one of the media outlets of interest before they make the prediction
5. Eliciting subjects' prior beliefs on the percentage of study participants who can make correct predictions, if they are assigned to each of the news outlets of interest
6. Information treatment that provides truthful information on two of the outlets (or placebo if in the control group)
7. Eliciting subjects' posterior beliefs on the percentage of study participants who can make correct predictions
8. Subjects are asked to make two additional predictions, and they can read articles from the media outlets of their choice beforehand. Subjects are randomized into 3 groups: (1) no incentive in 1st question, $5 incentive in 2nd question; (2) no incentive in 1st question, $25 incentive in 2nd question; and (3) $5 incentives in both groups.
Experimental Design Details
Randomization Method
Randomization done by a computer, embedded in the experimental program.
Randomization Unit
Individual.
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
All randomization will be implemented at the individual level, and treatment will not be clustered.
Sample size: planned number of observations
We aim to recruit 5,000 individuals for the experiment. For the final analyses, we plan to exclude the experimental subjects based on the following criteria: 1. Exclude those who fail comprehension questions on incentives more than twice; 2. Exclude subjects whose response look either duplicative or inattentive (e.g. responses with same value for all belief questions).
Sample size (or number of clusters) by treatment arms
1. Incentive treatment: 2,800 subjects in the arm of no incentive in 1st question, $5 incentive in 2nd question; 600 subjects in the arm of no incentive in 1st question, $25 incentive in 2nd question; and 1,600 in the arm of $5 incentive for both questions.
2. Information treatment: 3,400 subjects receive information treatment, 1,600 subjects receive no information or a placebo information.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB

Institutional Review Boards (IRBs)

IRB Name
New York University Institutional Review Board
IRB Approval Date
2018-02-21
IRB Approval Number
FY2018-1762
IRB Name
Stanford University Institutional Review Board
IRB Approval Date
2017-09-26
IRB Approval Number
42883

Post-Trial

Post Trial Information

Study Withdrawal

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

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials