Prevalence and Targets of Anti-normative Content on Facebook and Instagram

• In Event: Political Parties and American Democracy Mini-Conference III: How the Public Used Facebook and Instagram in the 2020 U.S. Election

Abstract

This study analyzes the prevalence of content on Facebook and Instagram that deviates from conventional norms of civility, tolerance, and or accuracy in the United States. Specifically, we analyze English-language content that is classified as (a) containing hateful or intolerant slurs, (b) incitement to political violence, (c) likely to suppress voting, (d) false information, and/or (e) uncivil. This study also examines whether certain groups of Facebook and Instagram users are more likely to be exposed to, engage with, or produce each type of anti-normative content. Finally, we look more closely at one category in particular—content containing hateful or intolerant slurs—and examine what groups and communities are targeted by these posts and messages. The study examines each of these phenomena beginning just before the 2020 U.S. federal elections and ending in mid-February 2021, allowing us to examine data from the elections, the January 6th Capitol insurrection, and the presidential inauguration.

Authors

• Rebekah Tromble, George Washington University

• Deen Freelon, University of Pennsylvania

• Stefan McCabe, Northeastern University

• Taylor W Brown

• Winter Mason, Facebook Research