Am I Safe in My Home? Fear of Crime Analyzed with Spatial Statistics Methods in a Central European City Daniel Lederer - KFV (Austrian Road Safety Board), Research and Knowledge Management

Am I Safe in My Home? Fear of Crime Analyzed with Spatial Statistics Methods in a Central European CityDaniel Lederer | 19.6.2012 | ICCSA 2012, Salvador de Bahia, Brazil

2

Presentation Overview

• Introduction• Methods and Techniques Used• Analysis and Results• Fear of Residential Burglary• Vulnerability to Residential Burglary

• Conclusion and Future Research

19.06.2012 Urban Crime Analysis and Mapping

Urban Crime Analysis and Mapping 3

Main Project: Urban Crime Analysis and Mapping

19.06.2012

comprehensive report on the urban crime

situation

citizen‘s personal perception of

crime

police-reported crime

Urban Crime Analysis and Mapping 4

Introduction

Research Questions of the Present Study:• Are there differences in the level of fear of becoming a victim of a

residential burglary between the districts in the city?• Within the city, are there certain areas with a lack of technical safety

measures, which may lead to an increased vulnerability to burglary?

19.06.2012

Urban Crime Analysis and Mapping 5

Methods and Techniques Used

Quantitative Survey in a Central European City• Computer Assisted Telephone Interviews of 1,505 randomly selected

citizens• respondents were asked about different topics to personal security• special selection in the present study:• fear of residential burglary• anti-victimizations strategies to protect personal property

19.06.2012

Urban Crime Analysis and Mapping 6

Methods and Techniques Used

Quantitative Survey in a Central European City • dataset includes two important characteristics for spatial analysis:

1. 35 inhabitants were selected in a disproportional stratified random sampling from every district

2. the use of personal addresses

• important for measuring local differences in personal security

19.06.2012

Urban Crime Analysis and Mapping 7

Methods and Techniques Used

Spatial-based Information is Available on 2 Levels:• level of polygon data (districts)• level of point data (addresses)Advantages:• possibility to analyze the data with different spatial statistics methods• reduces certain sources of errors (e.g. Modifiable Areal Unit Problem)

19.06.2012

Urban Crime Analysis and Mapping 8

Methods and Techniques Used

Descriptive and Exploratory Spatial Data Analysis:• Spatial Autocorrelation• Kernel Density Estimation (KDE)• Nearest Neighbor Hierarchical Clustering (NNHC)

19.06.2012

Urban Crime Analysis and Mapping 9

Methods and Techniques Used

Spatial Autocorrelation• Global Moran’s I• useful to understand general spatial patterns• measures the deviation from spatial randomness by comparing the value at any

one location with the value at all other locations• Moran’s I statistic varies from -1 to +1

• Local Indicator of Spatial Association (LISA)• useful to identify statistically significant local spatial clusters• e.g. hot or cold spots• compares local averages to global averages and assesses the association of

certain events

19.06.2012

Urban Crime Analysis and Mapping 10

Methods and Techniques Used

Kernel Density Estimation (KDE)• interpolation method, which creates a smooth surface of the point

data with a variation in the density of enclosed points• areas with a high quantity of points result in a high density• based on two parameters:• grid cell size• bandwidth (search radius)

19.06.2012

Urban Crime Analysis and Mapping 11

Methods and Techniques Used

Nearest Neighbor Hierarchical Clustering (NNHC)• grouping spatially close points into hierarchical clusters• depends on the Nearest Neighbor Index test, which compares the

distances between the points of the actual distribution against a random distributed data set of the same sample size

• depending on two parameters:• threshold distance• minimum number of points for each cluster

19.06.2012

Urban Crime Analysis and Mapping 1219.06.2012

Urban Crime Analysis and Mapping 1319.06.2012

Urban Crime Analysis and Mapping 1419.06.2012

Urban Crime Analysis and Mapping 1519.06.2012

Urban Crime Analysis and Mapping 16

Conclusion

• spatial analysis methods help to better understand special topics in fear of crime in the selected European city

• by using different aggregation levels and techniques in clustering spatial data, a large amount of complex information could be compressed in thematic maps

• identifying of important clusters:• fear-of-residential-burglary hot spot in the Westside• vulnerability-to-residential-burglary hot spot in downtown• combination of hot spots matches in the Westside an overlapping result

19.06.2012

Urban Crime Analysis and Mapping 17

Future Research

• Why are the hot spots located in these specific areas?• enlarging the spatial analysis by using confirmatory spatial statistical

methods• investigating links between fear of crime, vulnerability to crime and

the actual occurrence of crime

19.06.2012

THANK YOU FOR YOUR ATTENTION!KFV (Austrian Road Safety Board)Mag. Daniel LedererResearch & Knowledge ManagementAustria | 1100 Vienna | Schleiergasse 18 Tel: +43-(0)5 77 0 77-1405 | Fax: +43-(0)5 77 0 77-1186E-Mail: [email protected] | www.kvf.at

Urban Crime Analysis and Mapping 1919.06.2012

• Anselin, L.: Local Indicators of Spatial Association-LISA. Geographical Analysis, vol. 27 (2), pp. 93—115

(1995)

• Anselin, L., Cohen, J., Cook, D., Gorr, W., Tita, G.: Spatial Analyses of Crime. Criminal Justice 2000, vol. 4,

pp. 213—262 (2000)

• Bailey, T.C., Gatrell, A.C.: Interactive Spatial Data Analysis. Longman (1995)

• Eck, J., Chainey, S.P., Cameron, J., Leitner, M., Wilson, R. (eds.): Mapping Crime: Understanding

Hotspots. National Institute of Justice, Washington DC (2005)

• Getis, A., Ord, J.K.: Local Spatial Statistics: An Overview. In: Longley, P., Batty, M. (eds.) Spatial Analysis:

Modelling in a GIS Environment. John Wiley & Sons (1996)

• Levine, N.: CrimeStat 3.0. A Spatial Statistics Program for the Analysis of Crime Incident Locations. Ned

Levine & Associates, Houston and U.S. Department of Justice, Washington DC.

http://www.icpsr.umich.edu/CrimeStat/download.html (2004)

• Openshaw, S.: The Modifiable Areal Unit Problem. Concepts and Techniques in Modern Geography, vol.

38 (1984)