Effects of COVID information on Consumer Behavior and Policy Preferences

How to cite

Rothwell, Jonathan. 2022. "Effects of COVID information on Consumer Behavior and Policy Preferences." AEA RCT Registry. AEARCTR-0009070. https://doi.org/10.1257/rct.9070-1.0

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

Jonathan Rothwell*

* Primary investigator

Registry ID
AEARCTR-0009070
DOI
https://doi.org/10.1257/rct.9070-1.0
First registered
Last updated
Status
Completed
Trial dates
2020-12-01 – 2020-12-07
Countries
United States of America ()
JEL codes
C91, D83, E21, E71, I38
Registry record
http://www.socialscienceregistry.org/trials/9070

Abstract

We investigate the role of information exposure in shaping attitudes and behaviors related to the SARS-CoV-2 (COVID-19) pandemic and whether baseline political affiliation and news diet mediate effects. We randomly assigned 5,009 U.S. adults to nine brief text-based segments related to the dynamics of the pandemic and the safety of various behaviors, estimating the effects on 15 binary outcomes related to COVID-19 policy preferences, expected consumer behavior, and beliefs about safety. Average effects reach significance (95% CI) in 47 out of 120 models and equal 7.4 ppt. The baseline effects are large for all outcomes except beliefs. By contrast, interaction effects by political party and media diet are significant for beliefs but rarely significant for policy and behavioral attitudes. These findings suggest partisan policy and behavioral gaps are driven, at least in part, by exposure to different information and that equalizing information sources would lead to partisan convergence in policy preferences and aspects of consumer behavior.

Intervention

In December of 2020, we launched an experiment using the Franklin Templeton-Gallup Economics of Recovery Study containing nationally representative data from 5,009 U.S. adults. Respondents were randomly assigned to one of nine conditions. These conditions were akin to news or information that respondents may have seen as media headlines, shared social media content, or summary messages from radio or television segments. The research team purposely chose segments that conveyed more or less alarming information, while requiring the segment to be accurate, fact-based, and objectively stated. The goal of the experiment was to test the effects of high-quality information that could be readily sourced and confirmed as originating from credible public agencies or experts.

Experimental design

Using a nationally-representative sample of U.S. adults, we test whether information segments affect the following outcomes:
-Policy support for COVID restrictions
-Plans to engage in travel and in-person dining
-Confidence in safety of going out in public
-Belief that virus is under control
-Plans to take COVID vaccine

We also test whether these effects differ for Democrats relative to Republicans and consumers of a liberal media diet compared to a conservative media diet.

Design details

Randomization method
Computer
Randomization unit
Individual
Clusters
0
Observations
5009 adults living in the United States
Arms
556 across nine groups.
Primary outcomes
We have 15 outcomes. Eight are measures of policy preferences, defined as support or no support for various COVID restrictions. We have four measures of planned consumer behavior within the next 30 days, and 3 measures of other beliefs about the safety of being in public, the extent to which the virus is locally under control, and plans for the vaccine. We use the most salient and alarming news segment as our reference group. Salience was determined by text search of online news websites. Thus, causal effects are calculated with respect to groups that saw widely shared negative messages.
Secondary outcomes
We test for interaction effects between our treatments and political party affiliation and with news diet.
Sponsors
Name: Franklin Templeton Location: United States Url: https://www.franklintempleton.com/

Papers from this trial

  1. Rothwell, Jonathan T. and Makridis, Christos and Ramirez, Christina and Desai, Sonal, COVID-19 Information Treatments Shift Related Policy Preferences and Plans for Consumer Behavior (June 3, 2021). Available at SSRN: https://ssrn.com/abstract=3859751 or http://dx.doi.org/10.2139/ssrn.3859751 http://dx.doi.org/10.2139/ssrn.3859751

Cited by

No citation data available for this registration yet.

Keywords