U.S. flag

An official website of the United States government, Department of Justice.

Investigating the Simultaneous Effects of Individual, Program and Neighborhood Attributes on Juvenile Recidivism Using GIS and Spatial Data Mining

NCJ Number
Date Published
March 2009
253 pages
This study developed, applied, and evaluated improved techniques for investigating the simultaneous effects of neighborhood and program forces in preventing juvenile recidivism.
Five broad conclusions are drawn from the data. First, delinquent reoffending is spatially dependent rather than spatially diverse; this finding is strongest for drug offending. Second, recidivism among delinquent adolescents spreads through a process of peer contagion and is specific to offense type. Third, for some types of offending, especially drug selling, juveniles are likely to specialize; this specialization is likely to be influenced by opportunities, constraints, and pressures in the youth's neighborhood. Fourth, different spatial units and their unique problems require different causal models; social disorganization is not always a useful explanation of recidivism. Fifth, drug selling is more easily predicted than other types of offending and is more logically related to opportunities presented in the neighborhood. A general conclusion regarding the impact of treatment programs for juvenile offenders is that large community-based programs that serve diverse populations of juveniles are less effective than small, more specialized programs that are tailored to the needs of youths being served and the characteristics of the neighborhoods where the youths live. A primary implication of this research is that future research on causes of delinquency should include an examination of the neighborhoods in which youth live, particularly regarding peer contagion, which in turn determines offense specialization. Study methodology involved the use of linear modeling, geographic information systems (GIS), and spatial data mining. GIS enables the integration of diverse spatial data sets, the quantification of spatial relationships, and visualization of the results of spatial analysis. This approach facilitates the investigation of how and why recidivism rates vary from place to place, through different programs, and among individuals. Tables, figures, references, and appendixes

Date Published: March 1, 2009