Meta-Analysis and the Integration of Terrorism Event Databases

Abstract

Why do terrorist attacks occur in certain places and times but not others? Despite advances in collection and empirical methods, the literature has produced divergent results and reached little consensus for common hypotheses about the economic, political, and social causes of terrorism. It is hard to know what to make disagreements as studies adopt disparate research designs using different datasets covering different locations and times. This article applies the xSub data protocol to conduct a meta-analysis of terrorism event datasets and isolate explanations for variations in findings. Although the datasets are constructed for different purposes by different research teams, with different inclusion standards, processing data onto a common event typology, and conducting analysis across common coverage reduces heterogeneity in findings. This protocol also facilitates comparisons with general conflict event datasets, providing researchers, policymakers, and practitioners with a broader context for understanding terrorism in relation to other forms of violence.