Plasmode simulation for the evaluation of causal inference methods in homophilous social networks

TL;DR

This paper develops a statistical framework for creating plasmode simulations from private social network data, enabling evaluation of causal inference methods while preserving data privacy and network features.

Contribution

It introduces a novel approach to generate realistic simulated social network data from private datasets using exponential family models.

Findings

Validated the simulation framework on real social network data.

Demonstrated the method's ability to preserve network and variable associations.

Provided insights into causal inference method performance in realistic settings.

Abstract

Typical simulation approaches for evaluating the performance of statistical methods on populations embedded in social networks may fail to capture important features of real-world networks. It can therefore be unclear whether inference methods for causal effects due to interference that have been shown to perform well in such synthetic networks are applicable to social networks which arise in the real world. Plasmode simulation studies use a real dataset created from natural processes, but with part of the data-generation mechanism known. However, given the sensitivity of relational data, many network data are protected from unauthorized access or disclosure. In such case, plasmode simulations cannot use released versions of real datasets which often omit the network links, and instead can only rely on parameters estimated from them. A statistical framework for creating replicated simulation datasets from private social network data is developed and validated. The approach consists of simulating from a parametric exponential family random graph model fitted to the network data and resampling from the observed exposure and covariate distributions to preserve the associations among these variables.