A Spatial Analysis of Predictors of Different
Types of Crime in Chicago Community
Brett Beardsley
Pennsylvania State University MGIS Candidate
Geog 596A
Stephen A. Matthews
Faculty Advisor
• Background
• Goals and Objectives
• Proposed Methodology
• Work Completed
• Hypothesis
• Timeline
• Chicago is the 3rd
Largest City in the
United States with
2.7 million people
• Much higher rates of
crime than New York
City and Los Angeles
Literature Review
• Spatial crime studies increasingly popular
• Origins date back to 1820s(France)
• Data and methods have evolved
Chicago Studies
• Focused on 5 studies from 1990-2009
• All regression or modeling techniques
• Numerous standard outcome and predictor
Previous Studies’ Conclusions
• Surrounding areas have an effect on one
another (i.e., Spatial dependence matters)
• Traditional indicators of crime ring true (e.g.
unemployment, poverty, population density)
• Not every variation can be explained
Goals and Objectives
• Analyze homicide, aggravated assault, criminal sexual
assault, robbery, burglary, motor vehicle theft,
larceny/theft, and arson within the 77 Community Areas in
Chicago from 2007 to 2011
• Identify most influential factors of crime in Chicago
Community Areas
• Identify common themes in high crime areas
• Identify how strong of an affect surrounding community
areas have on one another
• Outcome Variables
*rate per 100,000 people
• Predictor Variables
Time Frame and Unit of
• 2007-2011
• 77 Chicago Community
• Step 1: Collect the data
• Step 2: Manipulate data
• Step 3: Analyze manipulated data
Data Collection
• Crime data came from the
Chicago Police Department
• Retrieved some ready to use
predictor variable data from
the Chicago Data Portal
• Most of the predictor
variables came from the 5
year (2007-2011) American
Community Survey (ACS)
Data Manipulation
• Combined all crime data over the 5 year span
• Determined which attributes I needed from ACS data
• Create centroids for each Community Area
• Assigned 805 modified ACS tracts a Community Area name and
number based on location to centroids
• Dissolve ACS tracts by Community Area name and number and
compiled statistics for each
• Spatially joined 805 modified ACS tracts to 77 Community Area
Centroids based on Community Area name and number
• Finally did field calculations for percentages and means
Preliminary Analysis
• Map each outcome and predictor variable by
Community Area
• Visually identify patterns and irregularities
• Descriptive analysis-mean, standard deviation,
min, and max
Analysis Outcome
Variable Maps
Analysis Outcome
Variable Maps
Analysis Predictor
Variable Maps
Analysis Predictor
Variable Maps
Further Analysis
• As shown there is likely spatial autocorrelation within both the
outcome and predictor variables and correlation between them.
Calculate Moran's I (global) and LISA (local) spatial
autocorrelation/dependence measures
• Create spatial weights matrices in GeoDa
• Run Ordinary Least Squares (OLS) regression models using spatial
weights matrices on all crimes, violent crimes, property crimes, and
finally each individual crime.
• Check model assumptions and regression diagnostics
As necessary run spatial lag or spatial error models .
• Affect of surrounding neighborhoods will be
• Percent of vacant housing will have the most
influence on total crime rate
• Small number of observations for the unit of
analysis (77)
• ACS is an estimate
• Winter 2014-Perform more
advanced analysis on data and finish
• Spring-2014-Present at ILGISA
Regional Conference
• Advisor-Stephen A. Matthews
• Geography 586 Instructor-David O’Sullivan
• Capstone Workshop-Pat Kennelly
• Overall Guidance-Doug Miller and Beth King
Arnio, A. N. & Baumer, E. P. (2012). Demography, foreclosure, and crime: Assessing spatial heterogeneity in contemporary models of neighborhood
crime rates. Demographic Research 26:18, 449-488.
Berg, M.T., Brunson, R.K., Stewart, E.A., & Simons, R.L (2011). Neighborhood Cultural Heterogeneity and Adolescent Violence. Journal of
Quantitative Criminology 28, 411-435.
Boggs, S. (1965). Urban Crime Patterns. American Sociological Review 30:6. 899-908.
Bowers, K. & Hirschfield, A. (1999). Exploring links between crime and disadvantage in north-west England: an analysis using geographical
information systems. International Journal of Geographical Information Science 13:2. 159-184.
Ceccato, V. (2005). Homicide in Sao Paulo, Brazil: Assessing spatial-temporal and weather variations. Journal of Environmental Psychology 25:3, 307321.
Earls, F., Morenoff, J.D, & Sampson, R.J. (1999). Beyond Social Capital: Spatial Dynamics of Collective Efficacy for Children. American Sociological
Review 64:5, 633-660.
Graif, C. & Sampson, R. J. (2009). Spatial Heterogeneity in the Effects of Immigration and Diversity on Neighborhood Homicide Rates. Homicide
Studies 13:3, 242-260.
Matthews, S.A., Yang T-C., Hayslett, K.L., & Ruback, R.B. (2010). Built environment and property crime in Seattle, 1998-2000: a Bayesian analysis.
Environment and Planning 42:6, 1403-1420.
Morenoff, J.D. (2003). Neighborhood Mechanisms and the Spatial Dynamics of Birth Weight. American Journal of Sociology 108:5, 976-1017.
Raudenbush, S.W., Sampson, R.J., & Sharkey, P. (2008). Durable effects of concentrated disadvantage on verbal ability among African-American
children. Proceedings of the National Academy of Sciences 105:3, 845-852.
Shaw, C.R. (1929). Delinquency Areas. Chicago: University of Chicago Press.
White, R.C. (1932). The Relation to Felonies to Environmental Factors in Indianapolis. Social Forces 10:4, 498-509.
Brett Beardsley
[email protected]

similar documents