Cross-Platform Spread of Online Polarization

• In Event: Conflict, Coercion, and Radicalization through the Lens of Text as Data

Abstract

How news spreads and is consumed on social media and what individual characteristics or behaviors drive migration to radical echo chambers is a topic of much debate in political science. In this paper I explore how and when individuals consume news online, whether partisan news behaviors are consistent across platforms, and whether certain individuals are consuming increasingly extreme content over time. Using a panel of mobile data from provider MFour, I analyze YouTube viewing behaviors using text classification of transcripts to label videos as political or non-political as well as by political leaning. I combine the panel information with data from Spotify and web browsing as well as additional information collected using the YouTube API. Using a Bayesian hierarchical model, I impute panelist partisanship based on affiliations from the Cooperative Congressional Election Survey (CCES), which includes similar demographic characteristics to the MFour data. Preliminary results indicate that though content viewership does not become more extreme over the duration of the panel, several areas outside of political news are highly polarized, including gaming channels and other entertainment.

Author

• Erin Elizabeth Walk, Massachusetts Institute of Technology