Spatial and temporal analysis of political violence in the United States

Abstract

Acts of political violence in the continental United States have increased dramatically in the last decade. For this rise in political violence, we are interested in where and when such incidents occur: how are the locations and times of incidents of political violence distributed across the continental United States, and what can we learn from a detailed examination of these distributions? We find the distribution of locations of political violence is neither uniform nor Poisson random, and that such locations cluster into well-defined geographic regions. Focusing on the county level we find a markedly skewed distribution of county counts of incidents of political violence. Examination of news reports and commentaries provided by the Armed Conflict Location & Event Data Project for the extreme outlier counties reveals compelling political and social background to the reported incidents of political violence. This, together with credible information on the role of social media in fomenting political violence leads us to postulate a field notion of upsetness as a major background to political violence. Using the time stamp on incidents of political violence we constructed a nearest neighbor model to predict future incidents of political violence at specific locations that involved a fatality.