Quantifying Triggers With Event Coincidence Analysis: An Application to Mass Civilian Killings in Civil War, 1989-2017

Angela Chesler (University of Notre Dame)

Abstract: Scholars of international relations, comparative politics, and peace and conflict studies are often interested in questions concerning the triggers of extreme political outcomes such as war, coups, and genocide. In a causal chain, a trigger is an immediate cause that can be distinguished from the underlying or permissive causes in that the latter allow for the possibility of a given outcome, while the trigger serves as the decisive driver that instigates an outcome’s materialization (e.g. Waltz, 2001). Existing research has largely used qualitative methods, particularly process tracing, to evaluate the causal impact of triggers of momentous political events. However, scholars have lacked the quantitative tools to analyze the extent to which certain types of events may systematically trigger outcomes across a wide range of cases. In this paper, I introduce a method called event coincidence analysis (ECA), a statistical technique that originates in climatology and has the power to quantify the strength and directionality of the proximate drivers of rare events. ECA offers unique benefits over conventional econometric approaches for testing trigger-response dynamics and is well-suited for analyzing relationships in which the outcome of interest is extremely rare, the driver-outcome relationship may be non-linear, and the researcher is uncertain about the role of timing. After surveying the role of triggers in theoretical models for various political outcomes, I explain the mechanics and implementation of event coincidence analysis and demonstrate the utility of ECA with an application to the triggers of state-led mass killings during civil wars.

View Poster in a New Tab