The "Unfriending" Problem: The Consequences of Homophily in Friendship Retention for Causal Estimates of Social Influence

TL;DR

This paper highlights how homophily in friendship retention can bias social influence estimates in longitudinal network studies, especially when friendship attrition occurs, challenging previous contagion findings.

Contribution

It demonstrates that homophily in friendship retention can cause significant bias in social influence estimates, a factor often overlooked in longitudinal social network analysis.

Findings

Homophily in friendship retention induces upward bias in social influence estimates.

Friendship attrition exacerbates bias and reduces coverage in the Christakis-Fowler model.

Monte Carlo simulations show the impact of friendship retention homophily on causal inference.

Abstract

An increasing number of scholars are using longitudinal social network data to try to obtain estimates of peer or social influence effects. These data may provide additional statistical leverage, but they can introduce new inferential problems. In particular, while the confounding effects of homophily in friendship formation are widely appreciated, homophily in friendship retention may also confound causal estimates of social influence in longitudinal network data. We provide evidence for this claim in a Monte Carlo analysis of the statistical model used by Christakis, Fowler, and their colleagues in numerous articles estimating "contagion" effects in social networks. Our results indicate that homophily in friendship retention induces significant upward bias and decreased coverage levels in the Christakis and Fowler model if there is non-negligible friendship attrition over time.