Comparing the spatial content of thematic maps
Abstract:Measures of categorical association can be used to assess the degree of similarity between two thematic maps, having regard to the spatial distribution of map classes but not to the identity of the classes. Such measures can be used, for example, to compare maps that use different classification schemes, or to identify misregistration between maps. The theory of these measures and their relationship to measures of categorical agreement are discussed. Three representative measures of categorical association are investigated quantitatively using various categorical maps, including maps generated by classification of satellite images, of a study area in the Russian north. On the basis of this investigation, it is suggested that Cramer's V statistic provides a simple but effective measure of the similarity in the spatial content of maps.
Document Type: Research Article
Affiliations: Scott Polar Research Institute, University of Cambridge, Cambridge CB2 1ER, UK
Publication date: 2008-01-01