The Influence of Involvement on Disaggregate Attribute Choice Models

Abstract

The state of the art in applying the various disaggregate choice algorithms is to assume homogeneity of choice process within the population being analyzed. This article uses the involvement variable a priori to categorize the (sample) population into a high-involvement segment using a model that assumes a simultaneous or alternative processing approach and a low-involvement segment using a model that assumes a hierarchical or an attribute processing approach. Empirical results indicate increased predictive accuracy and diagnostic information that is more closely related to established consumer behavior theory.