Misogynistic Language on Twitter

Abstract

Studies have demonstrated that social media may offer insights into social behaviors. Here we investigate the potential of social media in predicting criminal behavior, in particular rape and sexual abuse. Traditional approaches for studying sexual violence are effective but laborious, although often limited to small sample sizes and coarse temporal resolutions. Additionally, the sensitive nature of sexual violence and stigmas against victims result in serious under reporting of rape crime statistics. The factors contributing to rape are not fully agreed upon, but research shows that the acceptance of, and willingness to commit rape are highly correlated with sex­role stereotyping, rape myth beliefs, and misogyny. Here we explore whether social media can be used as an indicator of sexual violence in the US, by tracking misogynistic tweets. We compared the number of tweets and rape crime statistics for each state, and found a significant association. Our work paves the way to the design of a ‘social sensor’ system to detect rape and other abuse by monitoring social streams.

Publication
Proc. ACM Web Science Workshop on Computational Approaches to Social Modeling (ChASM), 2014
Date