Identifying Competition and Mutualism Between Online Groups

TL;DR

This paper introduces a time-series method inspired by ecology to distinguish between competition and mutualism among online groups, revealing that mutualism is more prevalent than competition in overlapping subreddit clusters.

Contribution

It proposes a novel framework for analyzing intergroup relationships over time, addressing previous inconsistencies in understanding competition and mutualism.

Findings

Mutualism is more common than competition among overlapping groups.

The method effectively distinguishes between competition and mutualism.

Large-scale analysis confirms the prevalence of mutualism in online communities.

Abstract

Platforms often host multiple online groups with overlapping topics and members. How can researchers and designers understand how related groups affect each other? Inspired by population ecology, prior research in social computing and human-computer interaction has studied related groups by correlating group size with degrees of overlap in content and membership, but has produced puzzling results: overlap is associated with competition in some contexts but with mutualism in others. We suggest that this inconsistency results from aggregating intergroup relationships into an overall environmental effect that obscures the diversity of competition and mutualism among related groups. Drawing on the framework of community ecology, we introduce a time-series method for inferring competition and mutualism. We then use this framework to inform a large-scale analysis of clusters of subreddits that all have high user overlap. We find that mutualism is more common than competition.