I am a Postdoctoral Fellow at the Identity and Conflict Lab at Yale University and a Research Associate at the Center of the Politics of Development. I received my PhD in political science from the University of California, Berkeley. Two interconnected themes run through my work: how civilians experience violence, and how they navigate loyalties between state and non-state actors in their search for protection.

My dissertation and book project focus on an understudied phenomenon in the civil war literature: legal militias. I examine the conditions under which civilians form legal militias during civil wars and how legalization shapes interactions between civilians and armed groups. In particular, I investigate how legal militias affect civilian collaboration and violence against civilians.

Methodologically, my work combines quantitative and qualitative approaches to study the politics of violence. I assemble novel datasets and employ research designs that enable me to identify potential causal effects using observational data. I then draw on fieldwork-based evidence to shed light on the mechanisms that may drive those effects. In my dissertation, for example, I assembled a new municipality–year panel dataset of legal militias in Colombia and linked it with data on violence and political institutions, which I analyze using recent advances in panel data methods to identify the causal effects of legalization. I also conducted extensive fieldwork, interviewing national-level decisionmakers such as ministers and peace commissioners, as well as local elites, including large landowners and former militia members.

Alongside this substantive research, I contribute to political methodology through the development of the CausalQueries R package. This software enables researchers to specify and update causal models with Bayesian inference in order to answer counterfactual questions within a potential outcomes framework. The package is particularly useful in contexts where causal identification through standard research designs is difficult or impossible. To document and extend this tool, I co-authored the paper Making, Updating, and Querying Causal Models with CausalQueries (with Till Tietz, Georgiy Syunyaev, and Macartan Humphreys), which was recently conditionally accepted at the Journal of Statistical Software.