Public Discourse in the Web Does Not Exhibit Group Polarization

TL;DR

This study analyzes online public discourse across multiple websites and finds that, unlike offline group polarization, online discussions tend to become more moderate over time, possibly due to self-selection bias.

Contribution

It provides large-scale empirical evidence that online group deliberation does not lead to polarization but rather moderates opinions, challenging common offline assumptions.

Findings

Online discussions tend to become more moderate over time.

Self-selection bias may cause contrarian views to soften opinions.

Contrasts offline group polarization phenomena.

Abstract

We performed a massive study of the dynamics of group deliberation among several websites containing millions of opinions on topics ranging from books to media. Contrary to the common phenomenon of group polarization observed offline, we measured a strong tendency towards moderate views in the course of time. This phenomenon possibly operates through a self-selection bias whereby previous comments and ratings elicit contrarian views that soften the previous opinions.