The study began with the hypothesis that, although the elderly's overall risk of victimization is relatively low because of their private lifestyle and routine activity patterns, their risk of theft-related homicide victimization may be relatively high because they are more likely than younger persons to lack capable guardianship and to be perceived as suitable targets. Multinomial logistic regression analysis of Chicago homicide data (1975-1981) largely supported the hypothesis by indicating that the risk of theft-related homicide victimization increased with advanced age and was more likely among socially distant victims and offenders. Although the elderly have a relatively low risk of homicide victimization, they are disproportionately more likely to die in theft-related situations when they are victimized. In this study, 53 percent of elderly victims were slain under theft-related circumstances, whereas this was true for only 16.2 percent of victims under the age of 65. Tables, appendix, notes, references