Crime and Place: A Longitudinal Examination of Street Segment Patterns in Vancouver, BC

Objectives To test the generalizability of previous crime and place trajectory analysis research on a different geographic location, Vancouver BC, and using alternative methods. Methods A longitudinal analysis of a 16-year data set using the street segment as the unit of analysis. We use both the gr...

Full description

Saved in:  
Bibliographic Details
Main Author: Curman, Andrea S. N. (Author)
Contributors: Andresen, Martin A. ; Brantingham, Paul J.
Format: Electronic Article
Language:English
Published: 2015
In: Journal of quantitative criminology
Year: 2015, Volume: 31, Issue: 1, Pages: 127-147
Online Access: Presumably Free Access
Volltext (lizenzpflichtig)
Volltext (lizenzpflichtig)
Journals Online & Print:
Drawer...
Check availability: HBZ Gateway
Keywords:
Crime and place
Spatial criminology
Trajectory analysis
Description
Summary:Objectives To test the generalizability of previous crime and place trajectory analysis research on a different geographic location, Vancouver BC, and using alternative methods. Methods A longitudinal analysis of a 16-year data set using the street segment as the unit of analysis. We use both the group-based trajectory model and a non-parametric cluster analysis technique termed k-means that does not require the same degree of assumptions as the group-based trajectory model. Results The majority of street blocks in Vancouver evidence stable crime trends with a minority that reveal decreasing crime trends. The use of the k-means has a significant impact on the results of the analysis through a reduction in the number of classes, but the qualitative results are similar. Conclusions The qualitative results of previous crime and place trajectory analyses are confirmed. Though the different trajectory analysis methods generate similar results, the non-parametric k-means model does significantly change the results. As such, any data set that does not satisfy the assumptions of the group-based trajectory model should use an alternative such as k-means.
ISSN:1573-7799
DOI:10.1007/s10940-014-9228-3