Scale-independent spatial analysis

Abstract

It seems likely that all spatial data and all types of spatial analysis contain some type of error. Therefore it is thus impossible to perform error-free spatial analysis and it becomes the task of the spatial analyst to reduce error to the point at which it does not interfere which the conclusions drawn from a particular analysis. Hence, any type of spatial analysis can be criticized in some way; what becomes important is to differentiate between substantive error, the presence of which is likely to affect the conclusions reached and trivial error, whose presence is unlikely to affect the conclusions of the study. In this paper several examples of substantive error are discussed and ways in which these errors can be minimized to the point where they are trivial are described.