Introduction

In this chapter we give an introduction to spatial data analysis, and distinguish it from other forms of data analysis. By spatial data we mean data that contain locational as well as attribute information. We focus on two broad types of spatial data: area data and origin–destination flow data. Area data relate to a situation where the variable of interest—at least as our book is concerned—does not vary continuously, but has values only within a fixed set of areas or zones covering the study area. These fixed sites may either constitute a regular lattice (such as pixels in remote sensing) or they may consist of irregular areal units (such as, for example, census tracts). Origin–destination flow (also called spatial interaction) data are related instead to pairs of points, or pairs of areas in geographic space. Such data—that represent flows of people, commodities, capital, information or knowledge, from a set of origins to a set of destinations—are relevant in studies of transport planning, population migration, journey-to work, shopping behaviour, freight flows, and the transmission of information and knowledge across space. We consider the issue of spatial autocorrelation in the data, rendering conventional statistical analysis unsafe and requiring spatial analytical tools. This issue refers to situations where the observations are non-independent over space. And we conclude with a brief discussion of some practical problems which confront the spatial analyst.