An Analysis of the Patterns of Crime and Socioeconomic Status Visualized Through Self-Organized Maps

This work is research to explore the association of spatial patterns between crime and socioeconomic status (SES) through the use of self-organized maps (SOM). It had been found that the spatial patterns of crime could be associated with those of socioeconomic, and this work sought to further these analyses in order to better understand how crime patterns and SES were related. To explore this association, patterns of crime and SES were examined in three cities: Nashville, TN; Portland, OR; and Tucson, AZ. Three SOMs were used in each city: one to analyze the patterns of crime, a second to analyze the patterns of SES, and a third to analyze the patterns of crime and SES. Nodes from each of these SOMs were also mapped to analyze the geographic distribution of their associated tracts. The results found an association between the patterns of crime and SES. In the Nashville Case Study, the patterns of high crime and low SES were not clearly associated in the combined Crime-SES SOM, but a stronger association was found in the geographic analysis. In the Portland Case Study, high crime and low SES patterns were found to be associated in the SOM. In the Tucson Case Study, high crime was found to be associated with low SES, but low SES was not always found to be associated with high crime. In each case study, the spatial patterns of low crime and high SES were found to be strongly associated. The spatial patterns of high crime were found to be associated with those of low SES, but the spatial patterns of low SES were not always found to be the same as those of high crime.