The study found that large differences exist between the SHR and the NVSS estimates expressed both as counts and as rates per 100,000 residents. Of the 167 sample cities, only 5 reported the same number of victims for both systems. The study also found that in15 of the cities, the NVSS estimate exceeded the SHR estimate by 100 or more and in 5 of the cities the SHR estimates exceeded the NVSS estimates by 100 or more. Three major reasons were identified for these differences: 1) many cities failed to report the SHR data in some years during the study; 2) in some places, the police jurisdiction used in the SHR did not correspond to the boundaries of the city used in the NVSS; and 3) the SHR associated victims with the place where the assault occurred, whereas the NVSS associated victims with their place of residence. The findings suggest that NVSS data have several advantages over the SHR data for cities Victim-level homicide data were obtained from the SHR and the NVSS for the period 1987 through 1991. Population estimates from the 1990 Census of Population were used in the analysis. Figures, tables, notes, and references