Section 6.1 Temporal data in crime analysis
In this book so far we have really emphasised the role of place in presenting opportunities for crimes to occur. However, we cannot consider space without also considering time. An area might look very different during the day and during the night, on a weekend or on a weekday, and in the summer or in the winter.
On a macro-scale, the relationship between crime and the seasons is something that had been a topic of concern for researchers as long as space has ([Baumer and Wright, 1996]). Zooming into a micro-scale, changes in routine activities with time of day or day of week will affect the profile of a place, and linking back to the idea of crime places such as generators or attractors, will have significant effect on crime rates. In conceptualising these crime places, [Newton (2018)] emphasised the importance to consider the measure of busyness of a place by time of day in order to understand its role as an attractor, generator, or other crime place. For example, at transit stations during school days, there is a morning peak time of work and school users combined, an afternoon school closing peak, and a secondary and slightly later end-of-workday peak time. Any calculation of crime rates needs to account for these micro-level temporal variations. At whatever unit of analysis, time is a vital variable to include in our crime analysis and criminological research. To quote [Tompson and Bowers (2013)]:
[I]t is important to disaggregate data into sensible temporal categories to have a real understanding of the relationship between the variables under scrutiny (p.627).
Returning to the importance of the role of place, we can introduce spatio-temporal data analysis. Spatio-temporal analysis is the process of utilising geo and time-referenced data in order to extract meaning and patterns from our data. In the earlier days, crime pattern analysis has tended to focus on identifying areas with higher densities of criminal activity, but not so much the monitoring of change in crime patterns over time ([Ratcliffe and McCullagh, 1998]). However, crime hot spots display significant spatio-temporal variance, and the identification of spatio-temporal patterns of hot streets provides significant ’actionable intelligence’ for police departments ([Herrmann, 2013]). Evidently, we cannot ignore time as a variable in our analyses. And while the main focus of this book is space, we must take at least one chapter to introduce some key concepts, and provide additional resources for readers to follow up with.
