Statistical Biases, Measurement Challenges, and Recommendations for Studying Patterns of Femicide in Conflict

Abstract

Collecting data on conflict mortality—including data on femicide—is difficult and can be dangerous. The resulting data is often incomplete and not statistically representative of the victim population as a whole. Data on femicide in conflict suffers from additional complications due to measurement challenges stemming from definitional and operational ambiguities. Despite these difficulties, as more and higher quality data on femicide becomes available, there are new opportunities to use statistical methods to study patterns of violence, which can help inform policy and accountability efforts. However, this data needs to be used carefully: drawing population level inferences from incomplete datasets risks misunderstanding the true underlying dynamics of the violence. This article explores the challenges and opportunities of collecting and analyzing data on femicide and offers four recommendations for data collectors and data analysts.