Crime Mapping and Spatial Data Analysis using R

In this chapter we had a play around with some regular old crime data and discovered how we can use the sf package in R to assign it a geometry (both at point and polygon level), and how that can help us visualise our results. If you want to explore other spatial data, you can explore [231] for short tutorials in how to get and visualise different kinds of spatial data.

We covered some important concepts such as projections and coordinate reference systems, and we had a go at acquiring shapefiles which can help us visualise our data. We had a think about units of analysis, and how that will affect how we visualise our data. In the next chapter we will spend a bit of more time discussing how to make good choices when producing maps.

There are a number of introductory texts to geographic information systems and analysis — eg., [232], [124] — that provide adequate background to some of the key concepts we introduce in this chapter. The report by [2] produced for the National Institute of Justice still provides a good general introduction for basic ideas around crime mapping and can be accessed for free online. Chapter 3 of [178] offers invaluable observations on the nature of spatial and spatial-temporal attribute data, often using examples from crime research. The chapter by [236] on the spatial analysis of crime offers a very succinct review of what will be the focus of most of this book. From a more domain-knowledge point of view, the early chapters of [222] and [168] set the stage for the use of GIS as part of the crime analysis process, whereas [224], edited handbook provides an excellent introduction to environmental criminology (that provides the theoretical and empirical backbone to spatial analysis of crime). [163] is a more recent addition, but focuses on ArcGIS. Finally, [154] provides a general introduction to data visualisation with R using the ggplot2 package. Although Healy’s text is not just about mapping, it offers a very practical and helpful introduction to using ggplot2 where you can learn how to further customise your maps and other charts. For specifics of ggplot2 refer to [153].

This is a book about maps and spatial analysis, but clearly one of the first questions you need to ask yourself is whether producing a map is the right answer to your question. Just because your data is spatial doesn’t mean you need a map for every question you pose to this data. There may be other forms of data visualisation that are more appropriate for exploring and summarising the story you want to tell with your data. If you are uncertain about whether you need a map or other kind of plot, books such as [149], [225], or [237] provide useful guidance.