Find the Fake: Boosting Resistance to Tobacco and Smoking Misinformation with a Chatbot Game

How to cite

Lim, Rhys; Dugas, Michelle; Pinzon, Daniel; Afif, Zeina; Vakis, Renos; Mind, Behavior, and Development. 2024. "Find the Fake: Boosting Resistance to Tobacco and Smoking Misinformation with a Chatbot Game." AEA RCT Registry. AEARCTR-0012979. https://doi.org/10.1257/rct.12979-1.0

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

Rhys Lim* , Michelle Dugas , Daniel Pinzon , Zeina Afif , Renos Vakis , Mind, Behavior, and Development

* Primary investigator

Registry ID
AEARCTR-0012979
DOI
https://doi.org/10.1257/rct.12979-1.0
First registered
Last updated
Status
In development
Trial dates
2024-02-19 – 2024-05-31
Countries
Private
JEL codes
I12, I18, I28
Registry record
http://www.socialscienceregistry.org/trials/12979

Abstract

The increasing circulation of misinformation and disinformation poses a significant threat to the credibility of public institutions and their ability to enact policies that promote the well-being of society. Despite the urgency of addressing misinformation on a global scale, there has been limited research into effective solutions, particularly in low- and middle-income countries. One context where widespread misinformation proliferates is in the domain of smoking and tobacco. To tackle this issue, we conducted an experimental study in schools to assess the effectiveness of a unique intervention using a chatbot game on messaging services (e.g., WhatsApp, Viber) to prubunk misinformation by teaching commonly used misinformation tactics. Our goal was to enhance individuals' ability to identify deceptive misinformation techniques used in misleading headlines and reduce the intention to share such misinformation content with others. Essentially, we sought to inoculate and immunize individuals against misinformation through this intervention.

Intervention

Participants will complete three lessons on misinformation (pick and choose, fake experts, distraction, including examples of what misinformation looks like and what misinformation does not look like. Users received feedback on their performance in the form of scores and badges.

Experimental design

Participants will be randomly assigned to either the control group or the treatment group. Both groups will take part in intervention sessions aimed at educating them about three commonly used misinformation tactics. However, the treatment group will undergo the intervention sessions first, followed by outcome assessments. In contrast, the control group will undergo outcome assessments before participating in the intervention sessions.

Design details

Randomization method
randomization done by token within a chatbot
Randomization unit
individual
Clusters
minimum 620 individuals
Observations
minimum 620 individuals
Arms
minimum 310 individuals control, 310 individuals treatment
Primary outcomes
The primary outcomes include (i) rates of accurately discerning between headlines that use misinformation tactics and those that do not, and (ii) intentions to share misinformation headlines with others

Cited by

No citation data available for this registration yet.

Keywords