The Sequence of Factor Analysis and Cluster Analysis: Differences in Segmentation and Dimensionality Through the Use of Raw and Factor Scores

Abstract

Factor analysis and cluster analysis are both common techniques within this field. Cluster analysis often follows factor analysis, with clustering based on the resulting factor scores. This article compares the dimensionality and segmentation that result from this approach with the results that are obtained by first clustering all subjects and then factor analyzing within each of the resulting clusters. Problems with the ability to discriminate cases using factor scores and after discarding items are also discussed. An illustrative example, using artificial data, is used to demonstrate the differences that are inherent in the two methods. Suggestions for determining the applicability of the two methods are given.