The Limitations of Government Databases for Analyzing Fatal Officer-Involved Shootings in the United States

Abstract

Federal government databases recording officer-involved shooting fatalities are incomplete and unreliable. Voluntary reporting to the Supplementary Homicide Report (SHR), the National Vital Statistics System (NVSS), and the Arrest-Related Death Program (ARDP) are subject to underreporting and classification errors. The same shortcomings apply to statewide reporting in California and Texas, the only states with mandatory reporting requirements. Content analysis of open source records identified officer-involved shooting fatalities that occurred in the United States from January 1, 2006, through December 31, 2015. Those data were compared with data from the government databases. Analysis revealed 7,869 officer-involved shooting fatalities, an average increase of 51.8 incidents per year. Fatalities increased from 594 in 2006 to 1,007 in 2015—an increase of 69.5% in 10 years. Government data sources reported a low of 46.0% of incidents to a high of 75.3%, depending on the reporting year. Open source research reveals 30% to 45% more cases than official federal or state databases and can reveal much more data about other critical questions. The history of federal program efforts suggests it is unlikely that government recording of data on officer-involved shooting fatalities will improve. Government reporting programs have produced decreasingly effective results. Current web-based data collection efforts suffer from many of the same limitations exhibited in the federal programs. One promising option for improved data collection includes funding an independent party, such as a university, to collect data from open sources and supplement that data with public records requests and the currently collected official government data.