Disproportionate Voices: Participation Inequality and Hostile Engagement in News Comments

TL;DR

This study analyzes a vast dataset of news comments to reveal that online participation is highly unequal and that a small, active subset of users disproportionately influences discourse, often engaging in hostility, especially during elections.

Contribution

It provides the first large-scale empirical analysis linking participation inequality with hostile engagement in news comment sections using advanced classification models.

Findings

Participation is highly skewed with few users dominating discussions.

Participation inequality increases during elections.

Frequent commenters are more likely to post hostile content.

Abstract

Digital platforms were expected to foster broad participation in public discourse, yet online engagement remains highly unequal and underexplored. This study examines the digital participation divide and its link to hostile engagement in news comment sections. Analyzing 260 million comments from 6.2 million users over 13 years on Naver News, South Korea's largest news aggregation platform, we quantify participation inequality using the Gini and Palma indexes and estimate hostility levels with a KC-Electra model, which outperformed other Korean pre-trained transformers in multi-label classification tasks. The findings reveal a highly skewed participation structure, with a small number of frequent users dominating discussions, particularly in the Politics and Society domains and popular news stories. Participation inequality spikes during presidential elections, and frequent commenters are significantly more likely to post hostile content, suggesting that online discourse is shaped disproportionately by a highly active and often hostile subset of users. Using individual-level digital trace data, this study provides empirical insights into the behavioral dynamics of online participation inequality and its broader implications for public digital discourse.