Childhood obesity is a significant health problem potentially influenced crime. This study employed secondary analysis of cross-sectional data using Geographic Information Systems for spatial comparison and logistic regression for statistical analysis. This was done to investigate the association between obesity, as measured by Body Mass Index of preschool children, and distance to the nearest instance of different categories of crime. Three covariates were included in the study: median family income, proportion of the population who self-identified as visible minorities, and straight-line distance from the child’s postal code to the closest park or green space. Of the eight categories of crime studied, three categories of person crime (commercial robbery, street robbery and other violence) and one category of property crime (theft of vehicle) predicted childhood obesity. This research is unique, as it separated crime into eight categories for analysis, and measured the straight-line distance to crime, rather than using neighbourhood boundaries.