Conventional methods and models for spatial data analysis are based on 'hard' (quantitative, cardinally-measured) information. Approaches such as location-allocation models, optimization models, entropy models, spatial assignment models and regional growth models all re flect the past trend to cast complex and multidimensional spatial interaction patterns in the framework of a cardinal metric system. In recent years, significant progress has been made in the analysis of 'soft', qualitative or categoricaily-measured data. In the fields of both parametric and non-parametric statistics and...