In this study, online hate speech as a growing concern was examined, particularly during the COVID-19 pandemic, when anti-Asian sentiment increased across social media platforms.
While the prevalence of hateful content has been widely documented, the causal mechanisms underlying emotional and behavioral changes among users who post such content remain insufficiently explored.
The study addresses this gap by investigating the causal relationship between engaging in hateful content and subsequent changes in linguistic and emotional expression on social media.
Using a dataset of 6,002 Twitter/X users, the authors apply causal inference methods, including propensity score matching, alongside advanced topic modeling techniques. This approach allows for a comparison between users who post hateful content and a matched group of non-hateful users.
Findings show that users who engage in hateful posting display significantly higher levels of anger, anxiety, and negative emotions, as well as increased use of third-person pronouns. Moral outrage and profanity peak during hateful posts and decline over time, though they remain higher than in non-hateful content.
The analysis also reveals that hateful posts are more interconnected, address a wider range of topics, and are more similar to one another, indicating lower cohesion within individual posts but greater cohesion across posts overall.
Learn more about this study here: https://doi.org/10.1016/j.ipm.2025.104079
Reference
Ghenai, A., Noorian, Z., Moradisani, H., Abadeh, P., Erentzen, C., & Zarrinkalam, F. (2025). Exploring hate speech dynamics: The emotional, linguistic, and thematic impact on social media users. Information Processing & Management, 62(3), 104079