Geodemographics has been a rich topic of interest for geographers and regional scientists over a long period of time. In recent years there has been substantial interest in the development of commercial systems, but the use of these methods to support public policy, and their pedagogic value in teaching and research remains significant. The roots of neighbourhood classification lie in attempts from academic research to investigate and demonstrate the basis of urban spatial structure, and can be traced back clearly to ecological studies from the early to middle part of the last century. One of the significant features of the theoretical work is that it considers spatial structure as the outcome of an interaction between various land use and activity types: for example, the Burgess model features a Central Business District, zones of transition and blight, a Factory Zone and so on. In contrast, applied classification systems are strongly residential in their focus. In part, this surely relates to the fact that appropriate data is readily available and has been for some time (initially from censuses of population, and lately augmented in commercial classifications at least by data from lifestyle surveys and other registers).
However many uses of geodemographics imply a much broader interest in land use and activity patterns, just as in the original ecological studies. For example, in trying to understand spatial distributions of crime, particularly offences that occur outside of residential areas (alcohol-related violence, shop theft, autocrime), then alongside the residential population one might be interested in daytime populations, open space, vacant dwellings, transport infrastructure and building types. High quality information about such factors is now more prevalent than ever before: for example, in the UK a National Land Use and Property Gazetteer is now emerging for the first time. This provides an opportunity to significantly extend earlier work on the influence of ‘supply-side’ factors into geodemographic systems.
In this paper, we will review some of the extended sources from which it is possible to consider a wide range of land use indicators alongside traditional residential metrics of neighbourhood structure. We will experiment with the incorporation of land use data into geodemographic classifications within a specific local context, and evaluate the utility of alternative data sources. As well as reporting on the results, the outcomes will be benchmarked in the context of real applications. We will consider whether a fully specified national classification including land use indicators would add significant value to the systems which are already available.
For many years, geographers and regional scientists have generated and refined classifications which describe the spatial structure of cities and regions. Under the shorthand label of “geodemographics” these methods have found widespread application in commercial organisations, but also in the analysis of public service provision, resource allocation and questions relating to equity and social deprivation.
Typically, these classification systems are strongly demographic in their focus. For example, the Output Area Classification for the UK is built entirely on small area data from the census. While other sources are introduced in commercial systems like ACORN and MOSAIC, these are still almost exclusively about people and their behaviour. In this paper, we wish to explore the relevance of major new land use data sets for small area classification. In the UK this could include sources such as the National Land and Property Gazetteer, street data, or satellite imagery for example.
We will begin with a review of alternative sources of supply-side data before considering potential integration with conventional geodemographic perspectives. The need for these new types of measurement and representation will be considered in the context of our specific interest in the spatial distribution of criminal incidents. We will evaluate the extent to which revisions to both data and methodology might be significant in both an academic and an applied policy context.
Title and abstract TBC
There are a large number of publicly available Geo-spatial datasets (for example land use, street networks, transport routes, and neighbourhood geodemographics) that can potentially be very useful in helping academics and practitioners understand crime patterns and in explaining why crime hotspots occur in particular places. However, the extent to which potential users are aware of the existence of such data is largely unknown. Moreover, the added value that such data might bring to a wider audience is largely unexplored.
This paper describes the initial results of a project funded by JISC (an organisation with the primary aim to disseminate knowledge about Geospatial data with a view to encouraging its use) to explore how geospatial data could be better exploited in the analysis of crime patterns, particularly, in understanding the distribution of domestic burglary. It focuses on an exploration of data available, the preliminary findings of an online survey of potential users of Geo-spatial data, and some of the theoretical considerations as to the potential uses and limitations of this approach.