Overview

This one-day hands on GIS-based training course develops the existing skills that an experienced intelligence/crime/community safety analyst has for understanding hotspots.

It does this by introducing you to several advanced techniques that can help you explore spatial patterns in crime data.

In particular, the course aims to develop your skills in spatial statistical techniques that provide a means of testing for significance (i.e. testing and observing in statistical terms whether a geographic pattern is particularly unusual). 

The course may introduce new techniques and demystify others, with the aim of exploring their practical application for understanding hotspots.

Who this course is for

This GIS-based course is for users of MapInfo, ArcGIS (ArcMAP) and CadCorp SIS.

This course is suitable for:

• police
• community safety partnership (CSP) analysts
• researchers
• information officers

It may also be relevant to those in the following interest groups: hotspots, spatial statistics, spatial significance, spatial autocorrelation, spatial association, local indicators of spatial association and dual kerne density estimation.

Course content

The methods and techniques you'll learn about on the course include the following:

Risk-based clustering techniques

Techniques such as dual kernel density estimation allow for a secondary variable (e.g. population) to be considered in determining hotspots. That is, these techniques can be used to identify hotspots based on not only the distribution of crime, but how the underlying population influences this spatial distribution. We examine the use of dual KDE and its practical application.

Identifying emerging problem areas

A common analytical requirement is to determine how patterns of crime have changed over time. One approach for observing changes in crime is with map subtraction, where hotspot maps are created for two time periods, and a change map is produced showing those areas where crime has reduced or has increased. A problem with the map subtraction approach is that the change map can result in identifying a large number of areas where crime has increased, restricting the ability to be selective in the targeting of resources. On this course we introduce a more useful analytical approach which involves identifying areas that have contributed most to an increase. By doing this we can help to target response resources more specifically.

Local indicators of spatial association: local Moran's I, local Geary's C, and the Getis- and Ord-Gi and Gi* statistics

This group of spatial statistics provides a means of extending beyond methods such as kernel density estimation by identifying those areas where the clustering of crime points is significant. That is, they can determine areas that can be statistically defined as hot from those that are not, plus rank each hotspot according to significance thresholds (i.e. 95%, 99%, 99.9%). This part of the course guides you through these techniques and discusses their practical application. Particular emphasis is placed on using the Gi* statistic.

Software used

Some of the functionality we teach on the course is unavailable in standard GIS packages. ArcGIS v9.3 and above has the functionality for most of what we teach on this course, whereas MapInfo is still rather limited. We therefore make use of freeware software that includes the Dispersion Calculator, CrimeStat and RooksCase to perform some of the analytical techniques.

Entry requirements

To take this course you'll need:

• at least a foundation in MapInfo, ArcGIS or Cadcorp SIS software
• to be experienced in using KDE

Cost and concessions

There's a 10% reduction for bookings of two or more people - all group delegates must be booked at the same time.

Course team

Spencer Chainey

Spencer is the Principal Research Associate at the UCL Department of Security and Crime Science. His particular research interests are in developing geographical crime analysis and crime mapping. He carries out most of his day-to-day work on developing the use of data, information sharing and analysis to aid intelligence development and decision-making by police forces, community safety partnerships, and national crime reduction and policing agencies.

His work has influenced national (UK) policy, and has contributed to policing and crime reduction developments in the USA, Canada, Brazil, China, Germany, Northern Ireland, Australia, New Zealand and South Africa. His work is also used in examples of good practice by the UK Cabinet Office (Social Exclusion Unit), Local Government Improvement and Development, The Home Office, the Audit Commission, The Housing Corporation and the United States National Institute of Justice.

Course information last modified: 30 Nov 2022, 15:59