Use this URL to cite or link to this record in EThOS:
Title: Examining the extent to which hotspot analysis can support spatial predictions of crime
Author: Chainey, S.
ISNI:       0000 0004 5352 3323
Awarding Body: University College London (University of London)
Current Institution: University College London (University of London)
Date of Award: 2014
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
The premise that where crime has occurred previously, informs where crime is likely to occur in the future has long been used for geographically targeting police and public safety services. Hotspot analysis is the most applied technique that is based on this premise – using crime data to identify areas of crime concentration, and in turn predict where crime is likely to occur. However, the extent to which hotspot analysis can accurately predict spatial patterns of crime has not been comprehensively examined. The current research involves an examination of hotspot analysis techniques, measuring the extent to which these techniques accurately predict spatial patterns of crime. The research includes comparing the prediction performance of hotspot analysis techniques that are commonly used in policing and public safety, such as kernel density estimation, to spatial significance mapping techniques such as the Gi* statistic. The research also considers how different retrospective periods of crime data influence the accuracy of the predictions made by spatial analysis techniques, for different periods of the future. In addition to considering the sole use of recorded crime data for informing spatial predictions of crime, the research examines the use of geographically weighted regression for determining variables that statistically correlate with crime, and how these variables can be used to inform spatial crime prediction. The findings from the research result in introducing the crime prediction framework for aiding spatial crime prediction. The crime prediction framework illustrates the importance of aligning predictions for different periods of the future to different police and prevention response activities, with each future time period informed by different spatial analysis techniques and different retrospective crime data, underpinned with different theoretical explanations for predicting where crime is likely to occur.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available