Neighborhood Boundaries and Arson: Examining the Modifiable Areal Unit Problem

• In Event: Spatial and Temporal Analysis

Abstract

The concept of neighborhood remains important in criminology but there is an increasing academic interest in the potential impact of the Modifiable Areal Unit Problem (MAUP) on neighborhood based studies. In the present study data over arson from the Swedish rescue services 2007-2012 have been employed to analyze MAUP in the city of Malmö, Sweden. The city has been divided into 50*50 meter pixels as micro-places (n=64540) which have been assigned a value for arson from frequency of arson within the pixel. The analysis is based on a comparison of two types of administrative geographical units alongside 40 randomly generated sets of thiessen polygon geographical units. Empty two-level hierarchical regression models with the micro-places as level 1 unit have been used to calculate Intra-Class Correlations (ICC) separately with each of the 42 different geographical units of analysis as level 2 units. The analysis is repeated with two alternative methods, kernel density and euclidian distance, to calculate a value for each micro-place. Results show that administrative geographical units of analysis in some cases just are marginally better than geographical units with random boundaries if the basic urban structure is taken into account.

Author

• Manne Gerell, Malmö University