A survey experiment on belief and information regarding COVID-19 in China

How to cite

Wang, Siyu; Jiang, Shuguang; We, Qian. 2022. "A survey experiment on belief and information regarding COVID-19 in China." AEA RCT Registry. AEARCTR-0010659. https://doi.org/10.1257/rct.10659-2.0

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

Siyu Wang* , Shuguang Jiang , Qian We

* Primary investigator

Registry ID
AEARCTR-0010659
DOI
https://doi.org/10.1257/rct.10659-2.0
First registered
Last updated
Status
Completed
Trial dates
2022-05-01 – 2023-04-30
Countries
China ()
JEL codes
I18
Secondary IDs
C93
Registry record
http://www.socialscienceregistry.org/trials/10659

Abstract

We conduct a field experiment to understand people's belief and opinions on COVID-19 risk and policies. With intervention, we investigate how people perceive and interpret COVID-19 related information and how they update their beliefs.

Intervention

We present the respondents with two news stories in random order. One of the stories used the historical data of death rate in Shanghai to argue that COVID didn't lead to additional death in March and April. The other news stated that 86.32% of death occurred with people who are older than 70 years old. Half of the respondents read the news before reporting their opinions on COVID policies, while the rest read the news stories at the end of the entire session.

Besides information intervention, the experiment also uses the natural intervention as China moved from "Zero-COVID policy" to "Co-existing with COVID policy" in December of 2022.

Experimental design

In May 2022, we conducted an anonymous online experiment through the largest Chinese message platform Wechat. In the experiment, we asked questions to understand the individual belief over pandemic situation and preferences over COVID policies. In addition, we included questions regarding the reasoning ability and demographic variables. In the experiment, respondents read two news stories which included data on mortality rate and other information. Half of the respondents read the news as the final section while the other half revealed their opinions on COVID after reading the news. Besides information intervention, the experiment also uses the natural intervention as China moved from "Zero-COVID policy" to "Co-existing with COVID policy" in December of 2022. Same survey with a few additional questions at the end regarding their opinion of more recent COVID news will be conducted in December and the months afterwards.

Design details

Randomization method
We adopt a device question in which we check whether the birth date of respondent is an odd or even number. With the odd number, respondents read the news before reporting their opinions on COVID policies, while the rest read the news stories at the end of the entire session. Other randomization is done by a computer.
Randomization unit
Randomization is conducted at individual level.
Clusters
The experiment is randomized at individual level. Provinces of the respondents are used as the control variable.
Observations
2500-4500 subjects before the policy change, 2500-4500 subjects after the policy change. Additional pool with highly educated individuals is collected to understand how the opinion is associated with education level.
Arms
1250-2250 subjects in each information intervention treatment.
Primary outcomes
How information and policy change people's belief and opinions on COVID-19.
Secondary outcomes
How demographic background and reasoning abilities are associated with people's belief and opinions on COVID-19.

Cited by

No citation data available for this registration yet.

Keywords