Jahanbakhsh, F., Zhang, A. X., Berinsky, A. J., Pennycook, G., Rand, D. G., & Karger, D. R. (2021). Exploring lightweight interventions at posting time to reduce the sharing of misinformation on social media. Proceedings of the ACM on Human-Computer Interaction, 5(CSCW1), 1-42. https://dl.acm.org/doi/abs/10.1145/3449092 

University of Michigan - Ann Arbor

Researchers looked at the effectiveness of lightweight interventions on preventing social media users from spreading and sharing misinformation. The first experiment (Taxonomy study) deals with why people believe claims to be true or false. The second experiment (nudge study) deals with 3 types of nudges (checkbox for headline accuracy, checklist of tagging reasons, and short response explanations). They found that administering an accuracy test and having participants explain their reasoning decreased the amount of misinformation content that was shared. 
- RQ1: “What are the reasons why people believe or disbelieve news claims?”
- RQ2: “How does asking people to provide accuracy assessments of news claims and their rationales affect their self-reported sharing intentions on social media?”

Exploring lightweight interventions at posting time to reduce the sharing of misinformation on social media

- Data collection
- Taxonomy study (RQ1)
- Nudge study (RQ2)
- Analysis

Exploring lightweight interventions at posting time to reduce the sharing of misinformation on social media - Methodology

- Asking people to assess the accuracy of news claims decreased sharing of both true and false claims, but significantly more for misinformative claims
- Asking people to explain their rationale through short responses led to similar results
- Asking people to go through a checklist for why a claim may be false did not have significant impacts on sharing behaviors

Exploring lightweight interventions at posting time to reduce the sharing of misinformation on social media - Findings

- Behavioral nudges (checkbox, checklist, and responses) were helpful in reducing the sharing of false headlines compared to true headlines
- Requiring these behavioral nudges can reduce the amount of misinformation circulating, especially information that has not been fact-checked or moderated by a platform yet
- Possibility of Hawthorne effect affecting participants’ answers and behaviors 
- Further research on how people interact with the nudges vs. posts without nudges and how that helps them to determine whether the information presented is accurate or not
- Study the option of nudges and whether users will take advantage of them

Learn Before

Related