A Field Experiment on Douyin Usage

How to cite

Yu, Shizhe; Meng, Juanjuan; Gao, Yu; Gao, Qianyun. 2022. "A Field Experiment on Douyin Usage." AEA RCT Registry. AEARCTR-0008255. https://doi.org/10.1257/rct.8255-1.0

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

Shizhe Yu* , Juanjuan Meng , Yu Gao , Qianyun Gao

* Primary investigator

Registry ID
AEARCTR-0008255
DOI
https://doi.org/10.1257/rct.8255-1.0
First registered
Last updated
Status
Completed
Trial dates
2022-05-05 – 2022-07-25
Countries
China ()
JEL codes
D91
Registry record
http://www.socialscienceregistry.org/trials/8255

Abstract

This study aims to investigate the effect of Douyin and its algorithms on users’ subjective well-being, cognitive skills, and opinion polarization. We carry out a randomized experiment among both the general population and university students in China. Participants in the two treatment groups were asked to either log out of their Douyin accounts (so that the algorithm stops working) or refrain from using Douyin completely. Outcomes are measured through online surveys, direct observation of Douyin accounts, and the screen time records of their smartphones.

Intervention

"No Algorithm" group participants: log out of their Douyin accounts (so that the algorithm stops working);
"No Douyin" group participants: refrain from using Douyin completely.

Experimental design

This study aims to investigate the effect of Douyin and its algorithms on users’ subjective well-being, cognitive skills, and opinion polarization. We carry out a randomized experiment among both the general population and university students in China. Participants in the two treatment groups were asked to either log out of their Douyin accounts (so that the algorithm stops working) or refrain from using Douyin completely. Outcomes are measured through online surveys, direct observation of Douyin accounts, and the screen time records of their smartphones.

Design details

Randomization method
Randomization done by a computer.
Randomization unit
Individual
Clusters
220 Individuals
Observations
230 Individuals
Arms
75 participants control, 75 participants "No Algorithm" treatment, 75 participants "No Douyin" treatment.
Primary outcomes
Mental health, efficiency, creativity, reading ability, general intelligence, issue polarization, use of time on offline activities online activities
Secondary outcomes
Attitude towards Douyin's influence

Cited by

No citation data available for this registration yet.

Keywords