The Emotional Rift: Evaluating Affective Polarization on TikTok during the 2024 U.S. Presidential Election

TikTok has become one of the most popular social media platforms. By focusing on short videos which users can comment on, emotionally charged communication is both encouraged and rewarded. However, it is unclear how affective polarization manifests within the TikTok infra-structure. Drawing on Social Identity Theory to explain intergroup dynamics, this mixed methods study applies a systematic, large-scale corpus analysis to comments (n=10,000) collected from the TikTok accounts @KamalaHarris and @DonaldTrump shortly before the 2024 U.S. presidential election. Overall, findings indicate that affective polarization TikTok-specific communication can be sorted into the following categories: 1) (explicit) affective polarization, 2) positive ingroup affirmation, and 3) inferred affective polarization. This study contributes to a toolkit for automated prediction models, offering the base for mitigating negative effects of affective polarization and fostering cohesive engagement on social media.