In the first part of the study, an exploratory analysis of homicide at the individual level was conducted using national vital statistics mortality data. Cases defined as homicides were selected for further analysis. In the second part of the study, community level information was introduced to the individual level analysis by focusing on multi-community data originally collected by other researchers. Data in this part involved 1,748 cases of homicide recorded by police departments and medical examiners in several U.S. cities. The study assumed patterns of homicide occurrence would reflect both general patterns of behavior between individuals and the larger social structure within communities. Findings revealed a persistent pattern in homicide data was that homicides were concentrated on weekends. The degree of concentration, however, was not the same for all groups in the U.S. population. There was a large and significant interaction between race and age; age made little difference among blacks but it had a large impact on non-blacks. Different mechanisms appeared to affect the distribution of homicides across the week for blacks and others. For those who were not black, being young had an added impact. With respect to indicators of behavior that should be considered when homicides occur, there were large and statistically significant effects for alcohol consumption and homicide motive. The author indicates the findings are suggestive and point to the need to collect more comprehensive multi-community data. 15 references, 4 tables, and 5 figures