This project replicates three prior studies of how alliance participation affects the likelihood of militarized conflict. I assess the robustness of each paper's results to corrections for dyadic clustering of observations, using cluster-robust variance estimation, and varying intercept models.
The three studies are:
- Benson 2011: Unpacking Alliances: Deterrent and Compellent Alliances and Their Relationship with Conflict, 1816--2000
- Johnson and Leeds 2011: Defense Pacts: A Prescription for Peace?
- Kenwick and Vasquez 2017: Defense Pacts and Deterrence: Caveat Emptor
Each paper has a corresponding directory with the data and R scripts. VI STAN Model.stan provides the varying intercepts model for all three replications- all three re-analysis scripts will compile this model.