Classification for Market Segmentation: An Improved Linear Model for Solving Problems of Arbitrary Origin

A current problem in marketing research involves the use of scale measures of consumer characteristics to develop homogeneous classifications of consumers. The quality of these classification efforts is influenced by the nature of the scales, which are frequently characterized by arbitrary origin and variance. This paper draws on recent developments in classification methodology to propose and illustrate an integrative classification methodology which overcomes some of the current problems of classification approaches.