USING SPATIAL SEGREGATION MEASURES IN GIS AND STATISTICAL MODELING PACKAGES

It is generally known that many traditional measures of segregation cannot distinguish different spatial population patterns. Several spatial measures of segregation have been proposed to overcome this problem, but these spatial measures are difficult to use because they require explicit spatial information and complex computation. This paper shows that Geographic Information Systems (GIS) can provide the spatial information required by these indices and that statistical packages can offer the complex computation functions needed to calculate the indices. We use ARC/INFO as the GIS package and S-Plus as the statistical package to demonstrate how spatial segregation indices can be calculated by combining the capabilities of these two types of systems.