Misogynistic targeting of women in power predicts broader online harassment patterns

Abstract

Online harassment is a well-documented and studied problem on social media. Who does this harassing, how, and to what degree are important questions that can inform platform policies and automated controls as well as helping understand harassers more broadly. This study investigates users who were discovered because they created a post that harassed a women in power using misogynistic slurs. Do these tend to be isolated incidents, or do such users engage in higher rates of harassment more generally? Findings from Twitter, Parler, and Reddit suggest that this population uses offensive slurs at several times the rate of control groups. We break down these findings and discuss the implications for moderation, automation, user well-being, and platform success.