Righteous Shoot or Racial Injustice? What Crowdsourced Data Can(not) Tell Us About Police-Caused Homicide

Abstract

Social scientists have developed various theoretical perspectives to explain the disproportionate incidence of police-caused homicide involving Black citizens in the United States. A common approach focuses on structural characteristics (e.g., percent Black) of cities. Such research relies primarily on Uniform Crime Reports’ Supplemental Homicide Reports, which poses two problems for researchers. Undercounting raises concerns about the reliability of findings, and the data are not amenable to testing influential alternative hypotheses. Recently, efforts to more accurately count police-caused homicides have been undertaken, with these new crowdsourced databases increasingly being used in research. When merged with structural-level data, they may allow the estimation of multilevel statistical models that include city-level and event-level predictors of police-caused homicide. These databases also pose methodological challenges, but they hold out the promise of providing a more reliable answer to a fundamental question that has yet to be adequately addressed. Is the racial disparity in police-caused homicide primarily attributable to the objective threats posed by Black citizens or the subjective biases of police officers? This article evaluates the degree to which the new crowdsourced databases can help resolve this conundrum.