Trials and triangulations: Aggregation problems in event data on political violence

Triangulation constitutes a key approach to analyzing data on political violence, a phenomenon that is both sensitive and suffers from multiple reporting biases. Nevertheless, the triangulation process whereby researchers move from the underlying sources to produce event datasets on political violence remains under-documented and under-developed. Building on recent calls to improve data transparency by collecting political violence data at the event report level and relying on systematic and replicable aggregation procedures, Professor van Baalen will discuss both benefits and challenges of this approach. He will argue that aggregation procedures are more complicated than often assumed, and examine how both theoretical, empirical, and pragmatic considerations affect the choice of aggregation approach. Drawing on novel event report level data from the Masters and Agents of Election-Related Violence in Côte d'Ivoire and Kenya (MAVERICK) dataset, he'll further present descriptive evidence and two replication studies that illustrate how different aggregation models can yield different descriptive and correlational statistical results. In doing so, the new study contributes new insights, best practices, and lessons learned on how to make event datasets on political violence more reliable, transparent, and replicable.