Sympathy over Polarization: A Computational Discourse Analysis of Social Media Posts about the July 2024 Trump Assassination Attempt

TL;DR

The paper examines whether a high-profile political shock—the July 2024 Trump assassination attempt—fuels sympathy or polarization in online discourse. It combines aspect-based sentiment analysis via large language models, a Difference-in-Differences causal framework across state groups, and BERTopic-based topic modeling to analyze a two-week window of X posts from the United States. The findings indicate a broad, non-polarizing improvement in sentiment toward Trump after the event, with significant overall sympathy and no reliable Group × Event interaction; topic shifts focus from controversies to assassination-related and supportive content, and discourse patterns are broadly consistent across regions. These results contribute to understanding crisis communication dynamics on social media, suggesting that such shocks can generate nationwide sympathetic responses rather than deepening partisan divides, while also highlighting how discourse realigns around emotional and symbolic themes in the short term.

Abstract

On July 13, 2024, at the Trump rally in Pennsylvania, someone attempted to assassinate Republican Presidential Candidate Donald Trump. This attempt sparked a large-scale discussion on social media. We collected posts from X (formerly known as Twitter) one week before and after the assassination attempt and aimed to model the short-term effects of such a ``shock'' on public opinions and discussion topics. Specifically, our study addresses three key questions: first, we investigate how public sentiment toward Donald Trump shifts over time and across regions (RQ1) and examine whether the assassination attempt itself significantly affects public attitudes, independent of the existing political alignments (RQ2). Finally, we explore the major themes in online conversations before and after the crisis, illustrating how discussion topics evolved in response to this politically charged event (RQ3). By integrating large language model-based sentiment analysis, difference-in-differences modeling, and topic modeling techniques, we find that following the attempt the public response was broadly sympathetic to Trump rather than polarizing, despite baseline ideological and regional disparities.

Figures (6)

• Figure 1: Performance evaluation of stance detection across different models.
• Figure 2: Temporal evolution of public sentiment following the attempted assassination.
• Figure 3: Changes in the frequency of broad topic categories before vs. after the assassination attempt
• Figure 4: Sentiment scores by topics and types of states along time
• Figure 5: Result of the Elbow method to determine the optimal K clusters in BERTopic modeling.