Interest in analysing spatial data has grown considerably in the
scientific research community. This reflects the existence of well-formulated
questions or hypothesis in which location plays a role, of spatial data of sufficient quality, of appropriate statistical methodology.
In writing this book I have drawn on a number of scientific and also policyrelated fields to illustrate the scale of interest –actual and potential–in
analysing spatial data. In seeking to provide this overview of the field I have
given a prominent place to two fields of research: Geographic Information Science (GISc) and applied spatial statistics.
It is important as part of the process of understanding the results of spatial data analysis to define the relationship between geographic reality and how
that reality is captured in a digital database in the formof a data matrix containing both attribute data and data on locations. The usefulness of operations on
that data matrix–revising or improving an initial representation (e.g. spatial
smoothing), testing hypotheses (e.g. does this map pattern contain spatial clusters of events?) or fitting models (e.g. to explain offence patterns or health
outcomes in terms of socio-economic covariates)–will depend on how well
the reality that is being represented has been captured in the data matrix.
Awareness of this link is important and insights can be drawn from the GISc
literature.