Agents of Electoral Anarchy: Introducing the MAVERICK Dataset

Abstract

Electoral violence is a serious threat to free and fair elections. Theoretical explanations of electoral violence increasingly move beyond the simplified incumbent-opposition dichotomy, and yet, there is a lack of nuanced actor data needed to fully test such explanations. This article introduces MAVERICK, an actor-centered geo-referenced event report dataset on electoral violence in Côte d’Ivoire and Kenya. MAVERICK covers the period 1995–2022, and includes two of Africa’s deadliest election crises: the Ivorian 2010–2011 post-election crisis and the Kenyan 2007–2008 post-election crisis. Given the dataset’s more granular information on actor characteristics, affiliations, and roles, as well as added indicators of violence severity, MAVERICK is ideal for testing nuanced theories on the causes, dynamics, and consequences of electoral violence. The article introduces the conceptual and operational definitions used, outlines the data structure and data collection approach, and provides a first analysis of the agents of electoral violence in the two cases. Moreover, we demonstrate the dataset’s added value by probing how security force involvement, and involvement by different security force branches, shapes electoral violence severity.