Journal Article

Towards the Modest Predictability of Daily Burglary Counts

Shane D. Johnson, Kate J. Bowers and Ken Pease

in Policing: A Journal of Policy and Practice

Volume 6, issue 2, pages 167-176
Published in print June 2012 | ISSN: 1752-4512
Published online May 2012 | e-ISSN: 1752-4520 | DOI: http://dx.doi.org/10.1093/police/pas013
Towards the Modest Predictability of Daily Burglary Counts

Show Summary Details

Preview

Police decision making involves matching available resources to calls for service, which vary widely. In this article, daily counts of residential burglary, which reflect this variability, are analysed. Even modest levels of predictability in this variability would prove extremely useful. The article demonstrates the ‘burstiness’ of burglary counts, and the time in days over which useful levels of predictability may be gleaned. It is found that high burglary counts on certain days of the week presage high counts on subsequent days; that a degree of predictability is also achievable by knowing that a day or days 6–8 days before also had a high burglary count. Analyses suggest that the levels of prediction gained by the approach are quite substantial. The next step in research involves disaggregating a larger data set to further explore observed patterns and to compare predictive accuracy with the levels of performance that operational police officers can achieve on the basis of local knowledge.

Journal Article.  5044 words.  Illustrated.

Subjects: Policing

Full text: subscription required

How to subscribe Recommend to my Librarian

Users without a subscription are not able to see the full content. Please, subscribe or login to access all content.