Analyzing Political Discourse on Discord during the 2024 U.S. Presidential Election

TL;DR

This paper pioneers a large-scale analysis of political discourse on Discord during the 2024 U.S. presidential election by collecting over 30 million messages from 81 public servers, categorized as Republican-, Democratic-, or unaligned. It combines polarity-based term valence analysis, temporally segmented Word2Vec embeddings with WEAT bias testing, and a RoBERTa-based multi-class hate-speech classifier to characterize ideological differences, implicit biases, and toxicity dynamics across campaign moments. Key findings include Republican-focused economic discourse, Democratic emphasis on equality issues, higher toxicity in Republican and unaligned spaces, and a surge in sexist hate speech following Kamala Harris’s nomination, alongside moderation challenges inherent to Discord’s decentralized structure. The work provides a foundational, public dataset and methodological framework for studying political engagement on Discord, with implications for understanding online mobilization, polarization, and platform moderation in emergent social networks.

Abstract

Social media networks have amplified the reach of social and political movements, but most research focuses on mainstream platforms such as X, Reddit, and Facebook, overlooking Discord. As a rapidly growing, community-driven platform with optional decentralized moderation, Discord offers unique opportunities to study political discourse. This study analyzes over 30 million messages from political servers on Discord discussing the 2024 U.S. elections. Servers were classified as Republican-aligned, Democratic-aligned, or unaligned based on their descriptions. We tracked changes in political conversation during key campaign events and identified distinct political valence and implicit biases in semantic association through embedding analysis. We observed that Republican servers emphasized economic policies, while Democratic servers focused on equality-related and progressive causes. Furthermore, we detected an increase in toxic language, such as sexism, in Republican-aligned servers after Kamala Harris's nomination. These findings provide a first look at political behavior on Discord, highlighting its growing role in shaping and understanding online political engagement.

Figures (6)

• Figure 1: Volume of messages shared on Discord political servers, categorized by metadata. The first plot shows total daily messages across all servers. The second plot shows daily mentions of "Trump", "Biden", and "Kamala".
• Figure 2: Political valence of top terms and URLs shared and discussed in Democratic and Republican-aligned Discord servers across the studied time period.
• Figure 3: Visualization of S-values computed using models trained on servers with Democratic and Republican alignment for each period of the electoral campaign. The baseline model corresponds to the used pre-trained model. A positive S-value indicates closer alignment with positive terms, while a negative S-value indicates closer alignment with negative terms.
• Figure 4: Bar plot of Hate Speech for Democratic Aligned, Republican Aligned and Unaligned in each period.
• Figure 5: Radar charts displaying the percentage of hate speech messages by period. The highest value, 0.55%, occurred in the 'Sexism' category during the Kamala vs. Trump race.