SPOTLITE Anomaly Detector Dataset

The SPOTLITE Anomaly Detector was designed to identify U.S. counties with unusually high or low numbers of law enforcement uses of lethal force relative to national benchmarks. Because counties differ substantially in population size, population density, and levels of violent crime, simple incident counts can be misleading when comparing jurisdictions. The anomaly detector provides a standardized approach for identifying counties where observed incident levels are substantially higher or lower than predicted. The anomaly detector is not intended to explain why anomalies occur or to evaluate whether a county’s performance is good or bad. Instead, it serves as a screening tool that highlights counties deserving closer examination. By providing a consistent method for comparing every county across the United States, the SPOTLITE Anomaly Detector offers a useful starting point for research, policy analysis, and efforts to better understand patterns of police use of lethal force. **CITATION NOTE To cite the white paper (or any other documentation associated with the SPOTLITE Anomaly Detector) please use the following citation: Jennings, Jay (2026) "Understanding the SPOTLITE Anomaly Detector". University of Illinois at Urbana-Champaign. To cite data from the SPOTLITE Anomaly Detector Dataset please use the following citation: Jennings, Jay; Singh, Ajay; Althaus, Scott; Martin, Michael; Bajjalieh, Joseph; Robbennolt, Jennifer (2026): SPOTLITE Anomaly Detector Dataset. University of Illinois Urbana-Champaign. https://doi.org/10.13012/B2IDB-2880543_V1