Examining factors associated with Twitter account suspension following the 2020 U.S. presidential election

Abstract

Online social media enables mass-level, transparent, and democratized discussion on numerous socio-political issues. Due to such openness, these platforms often endure manipulation and misinformation - leading to negative impacts. To prevent such harmful activities, platform moderators employ countermeasures to safeguard against actors violating their rules. However, the correlation between publicly outlined policies and employed action is less clear to general people. In this work, we examine violations and subsequent moderations related to the 2020 U.S. President Election discussion on Twitter. We focus on quantifying plausible reasons for the suspension, drawing on Twitter's rules and policies by identifying suspended users (Case) and comparing their activities and properties with (yet) non-suspended (Control) users. Using a dataset of 240M election-related tweets made by 21M unique users, we observe that Suspended users violate Twitter's rules at a higher rate (statistically significant) than Control users across all the considered aspects - hate speech, offensiveness, spamming, and civic integrity. Moreover, through the lens of Twitter's suspension mechanism, we qualitatively examine the targeted topics for manipulation.