Previous research on criminal behavior has focused on the links between job markets and crime rates. This research generally involved large metropolitan areas and large spans of time. The current analysis expands this research literature by focusing on whether the variation of job markets within the city of Chicago is related to spatial patterns of crime, in this case, homicide. Data were drawn from the Chicago homicide dataset maintained by the National Archive of Criminal Justice Data, police files on homicides from the Chicago Police Department for the years 1965 through 1995, and job accessibility information from the 1990 Census Transportation Planning Package. Results of covariate analysis, spatial clustering, and multivariate regressions indicated that after controlling for socioeconomic covariates, there was an inverse relationship between job accessibility and homicide rates across geographic areas of Chicago. The author offers a discussion of the complexities involved with analyzing rare events in small populations and how these problems were addressed using constructs of various levels of geographic areas from census tracts and through the use of the spatial clustering method based on scale-space theory. The results hold implications for public policy insomuch as only jobs that are accessible have the potential to alter crime rates in geographic areas. Future research should examine the same constructs using different methodologies, such as Poisson-based regressions. Tables, figures, references