Working Papers / Reports

This report critically evaluates the Early Warning Project (EWP) Model, a prominent quantitative framework designed to predict and prevent genocide. As a predictive tool widely referenced in genocide prevention efforts, the EWP Model’s methodology, transparency, and adherence to academic standards are thoroughly analyzed. The report examines key components, including variable selection, statistical underpinnings, and the model’s evolution, to assess its reliability and practical utility. Complementing the model analysis, an extensive literature review contextualizes the EWP Model within the broader field of genocide research. It highlights significant methodological flaws, including reliance on correlational rather than causal data, use of static and redundant variables, and selective adaptation of foundational studies without adequate justification. The report critiques the lack of transparency in variable selection, methodological updates, and scholarly attribution, which compromises the model’s credibility. Key issues identified include logical inconsistencies, such as equating extremes (e.g., full civil liberties with complete repression), reliance on outdated or inconsistent data, and counterintuitive conclusions, such as associating multi-party democratic system, civil liberties with increased risk of mass killings. Comparisons with foundational studies reveal substantial methodological divergence. The report calls for greater transparency, theoretical rigor, and adherence to best practices to improve the model’s utility as a predictive tool for mass atrocity prevention.