Citation

Schnell C, Braga AA, Piza EL. J. Quant. Criminol. 2017; 33(3): 469-496.

Copyright

(Copyright © 2017, Holtzbrinck Springer Nature Publishing Group)

DOI

10.1007/s10940-016-9313-x

PMID

unavailable

Abstract

OBJECTIVEsThe influence of three hierarchical units of analysis on the total spatial variability of violent crime incidents in Chicago is assessed. This analysis seeks to replicate a recent study that found street segments, rather than neighborhood units of analysis, accounted for the largest share of the total spatial variability of crime in The Hague, Netherlands (see Steenbeek and Weisburd J Quant Criminol. doi:10.1007/s10940-015-9276-3, 2015).

METHODSWe analyze violent crime incidents reported to the police between 2001 and 2014. 359,786 incidents were geocoded to 41,926 street segments nested within 342 neighborhood clusters, in turn nested within 76 community areas in Chicago. Linear mixed models with random slopes of time were estimated to observe the variance uniquely attributed to each unit of analysis.

RESULTSSimilar to Steenbeek and Weisburd, we find 56-65 % of the total variability in violent crime incidents can be attributed to street segments in Chicago. City-wide reductions in violence over the observation period coincide with increases in the spatial variability attributed to street segments and decreases in the variability attributed to both neighborhood units.

CONCLUSIONSOur results suggest that scholars interested in understanding the spatial variation of crime across urban landscapes should be focused on the small places that comprise larger geographic areas. The next wave of "neighborhood-effects" research should explore the role of hierarchical processes in understanding crime variation within larger areas.

Language: en