The original study included homicide incidents with multiple victims. As a result, Baldus, Woodworth, and Pulaski (1990) operationalized the victim race variable as ―one or more white victims.‖ In the current study, we limit the analysis to homicides that involve single victims in order to be certain of the gender and race of the victim. For instance, using the variables provided by Baldus, Woodworth, and Pulaski (1990), in a multiple victim case where one or more victims is white and a female victim is involved, it is unclear whether it is the female victim who is white. To resolve this complication, we delete the few multiple victim cases (63 out of 2,484 weighted cases) from the sample for the remaining analyses.11 The regression results that take into account these modifications are presented in table 1, model 2. Limiting the sample to single-victim cases (table 1, model 2) produces results that are comparable with those obtained in the original model (table 1, model 1).