This short course provides a tutorial on statistical methods for causal inference for networks with applications to public health. Students will learn cutting-edge research for causal inference in networks and understand how public health practitioners and other researchers can implement these methods. Part 1 covers causal inference methods under the Stable Unit Treatment Value Assumption (SUTVA). Part 2 covers network science visualizations, descriptions, and modeling approaches. In the last session, we will learn about methods that synthesize the two fields to assess causal effects in networks.

This course uses a simulated data set based on the Transmission Reduction Intervention Project Athens site to practice applying the statistical methodology. This study sought to develop effective intervention techniques against HIV transmission during the recent infection period using a combination of injection-, sexual- and social-network-based contact tracing methods. Community alerts were also distributed in the networks and venues of recent infectees.