Defining hot spot thresholds

Hotspot maps should be designed to help direct and prioritise the focus of crime and disorder reduction resources to particular areas. Hotspot maps that are designed should be practical in identifying the areas of highest crime and disorder concentration against areas of less concentration.

Maps showing the distribution of crime or disorder as a continuous density surface are increasingly replacing point maps and thematic boundary maps as ways to visualise and understand patterns of criminal and anti-social behaviour activity. An approach that helps to separate and define thresholds of crime concentration in the map uses incremental values of the mean. The approach is useful as it is,

• practical to apply

• the thresholds generated are meaningful and can be linked to a value that can be easily understood as a unit describing crime concentration

• the separation of thematic thresholds follow a consistent methodology and where the upper most categories consistently define when a crime concentration can be defined as reaching hotspot status

• the method is more robust by taking into account the statistical spatial distribution of the point data

• yet retains flexibility in map design, appropriate to the output required at different scales and for different applications.

A method that detects crime and disorder hotspots on a relative scale removes the restricted application of a static setting that defines levels in crime concentration (e.g. a hotspot is a hotspot when there are more than 25 crimes per square kilometre). Implementing a relative approach also enables community safety partnerships operating at different geographic scales (e.g. ward, local, or regional levels) to apply the same methodology to identify and prioritise the tackling of crime hotspots.

When is a hot, hot? An incremental mean approach.

Incremental multiples of the grid cells’ mean are used to define crime hotspot map thresholds. The first step requires selecting only those cells generated by a kernel density estimation method which have a value greater than 0. From this grid cell set, the mean cell value can be calculated and used to set thematic thresholds at:

• 0 to mean

• Mean to 2 mean

• 2 mean to 3 mean

• 3 mean to 4 mean

• 4 mean to 5 mean

• Greater than 5 mean

As a statistic, the mean is an easy value for the novice map reader to grasp. Increments of the mean are more obviously linked to increasing values, and their relative significance. This makes this approach appealing as a method to define crime hotspot legend thresholds.

The map below shows the application of this incremental mean threshold approach for robbery.

Click here for an enlarged version