Spatial Data Mining is the process of discovering interesting and previously unknown, but potentially useful patterns from large spatial databases. Most relationships in spatial datasets are regional and there is a great need for regional regression methods that derive regional reflects different spatial characteristics of different regions. A central challenge in spatial data mining is the efficiency of spatial data mining algorithms, due to the often huge amount of spatial data and the complexity of spatial data types and spatial accessing methods. This paper proposes a regional regression technique for regions that are defined by a categorical attribute, in particular soil type. The result is a series of hierarchically grouped regions according to their similarity.