- Influential politicians received more Birdwatch notes, most of which were reported for misleading tweets
- Accounts reported for misleading tweets are typically verified and have large followings but follow very few accounts 

University of Michigan - Ann Arbor

- Birdwatch notes (RQ 1 & RQ 2)
- Source tweets (RQ 3)
- Ratings (RQ 4)
- Helpfulness (RQ 4 & RQ 5)

Community-Based Fact-Checking on Twitter's Birdwatch Platform - Analysis

Pröllochs, N. (2021). Community-Based Fact-Checking on Twitter's Birdwatch Platform. arXiv preprint arXiv:2104.07175. https://arxiv.org/abs/2104.07175 

Community-Based Fact-Checking on Twitter's Birdwatch Platform

- 87.96% of Birdwatch notes were written for misleading tweets and only 12.04% were for non-misleading tweets 
- Misleading tweets are also a little more likely to include trustworthy sources (63.96%) than non-misleading tweets (62.92%)
- Misleading tweets reported due to factual mistakes (30%), missing contexts (28%), unverified facts (25%), outdated information (5.35%), satire (5%), manipulated information and media (3%), or other (3%)
- Not misleading tweets reported for containing factual information (62%), personal opinion (16%), or satire (12%)
- Notes for misleading tweets tend to contain longer explanations and less complexity with the use of negative sentiments 

Community-Based Fact-Checking on Twitter's Birdwatch Platform - Birdwatch Note Analysis

Community-Based Fact-Checking on Twitter's Birdwatch Platform - Source Tweet Analysis

- Birdwatch notes that report misleading tweets generally have a larger helpfulness ratio and more votes compared to non-misleading tweets with notes 
- Users found Birdwatch notes more helpful if trustworthy sources were included and if the reported tweet had more votes saying it is not misleading
- Birdwatch notes were considered helpful if they are informative (31%), clear (24%), or provide trustworthy sources of reference (19%)
- Birdwatch notes considered unhelpful due to opinions/bias (25%), unreliable sources (20%), or missing key ideas (15%)

Community-Based Fact-Checking on Twitter's Birdwatch Platform - Ratings Analysis

- Regression model used to determine helpfulness of Birdwatch notes
- Contents of notes, social influence of user accounts, and reference to trustworthy sources are all important for determining helpfulness

Learn Before

Related