All Times EDT
Virtual
Exploratory Spatial Data Analysis of Rape in Philadelphia, 2015 (309600)
*Michelle Sarah Livings, University of Southern CaliforniaKeywords: Exploratory Spatial Data Analysis, Spatial Statistics, Spatial Regression, Crime Data, Rape
Spatial analyses of crime data are common in the literature, but no studies have focused on rape. Two questions arise from reading non-spatial studies about rape: Are rapes committed in spatial clusters? Do neighborhood demographics impact where rape is committed? An exploratory spatial data analysis was conducted on reported rapes in Philadelphia, PA in 2015. In 2016, Philadelphia had a rate of rape (80 rapes per 100,000 population) double the national average (41 per 100,000).
Data were obtained from the Philadelphia Police Department Crime Incidents dataset, American Community Survey 2013-2017 5-year estimates, and U.S. Census Bureau TIGER/Line Shapefiles database. Spatial patterns of reported rape were analyzed using point pattern (Ripley's K), areal (Global Moran's I, Getis-Ord Gi*, Bivariate Local Moran's I), and regression analyses (OLS, spatial lag).
Spatial clustering was significant in point data and aggregated data. Rate of rape was significantly related to neighborhood poverty and proportion of single mother households. A spatial lag model provided the best fit for the data with R^2=0.35. Additional analytic details, limitations, and next steps will be discussed.