Neighborhood Effects on Crime: The Concentration of Racial/Ethnic Groups and the Heterogeneity Among Such Groups.

It is not entirely clear what attribute of neighborhoods' race/ethnicity affects crime rates over time. Drawing on a sample of block groups in Baltimore (MD) we therefore estimate negative binomial regression models to test the concentration of specific ethnic groups in neighborhoods (% Black, % White, % Latino, % Asian, and % other race) as well as the level of heterogeneity among these groups. The results show that ethnic heterogeneity is associated with higher rates of both violent and property crime years later, whereas the measures of nonwhite ethnic groups are not significant. We also determine that concentrated disadvantage moderates the effect of ethnic heterogeneity on property crime. The implications of these findings for criminology and urban sociology are discussed. [ABSTRACT FROM AUTHOR]