chapter  4
Geographic Information Systems in Spatial Epidemiology and Public Health
ByRobert Haining, Ravi Maheswaran
Pages 30

Definitions of geographical information systems (GIS) usually fall into one of three categories: toolbox-, database-, and organization-based definitions (Burrough and McDonnell, 2000). Geographical or spatial data are data about entities in the real world (physically represented as point, line, or area objects) that define their location and the attributes recorded at each location. Knowing where entities are located allows spatial relationships between them to be defined, such as distances between points, adjacency or otherwise of areas, and the proximity of one object to another. A GIS database is distinguished from most other databases by knowing where in geographical space objects (points, lines, and polygons) are located in relation to one another since most other kinds of databases capture only entities and their attributes. Database definitions of GIS emphasize this difference: “any . . . computer based set of procedures used to store and manipulate geographically referenced data” (quoted in Burrough and McDonnell, 2000, p. 11, italics added). Geography is stored in the form of a series of discrete layers in the database (e.g., a layer for the roads, a layer for the waste sites, a layer for the forested areas, and a layer for the deprivation scores by census area), enabling different features to be switched on or off when visualizing or mapping an area.