Incident data encompass time of occurrence (year, month, day, and time of day), geographic locations (State and region), injuries sustained, incident circumstances, and a comparison with crime rates for the general population. The study was based on a sample of 1,103 incidents. There was little long-term correlation between any given months and the number of officer deaths. Fewer deaths occurred on Sunday, and most deaths occurred between 10:01 P.M. and midnight. Most officers were killed in the Southern region, and the fewest were killed in the Northeast. Only 7 percent of the victims were not killed by some type of firearm. Most officers were killed while attempting an arrest (491). Significant relationships existed between the police homicide rate and the general rates of homicide, rape, and aggravated assault, suggesting the hypothesis that the police homicide rate is a function of the prevalence of cultural violence rather than specific types of criminal activity. An analysis of particular incident circumstances, such as the wearing of body armor and whether the officer was shot with his/her own gun, has implications for training and safety procedures. 11 references and 13 data tables.