Tree representations of similarity data are investigated. Hierarchical clustering is critically examined, and a more general procedure, called the additive tree, is presented. The additive tree representation is then compared to multidimensional scaling. (Author/JKS)

Describes the American mathematical psychologists' computer-based method of constructing representations of the psychological structure of a set of stimuli on the basis of pairwise measures of similarity on confusability. Psychological structure is represented utilizing multidimensional spatial configurations and nondimensional tree-structures or…

A three-way multidimensional scaling model is presented as a method for identifying classroom cliques, by simultaneous analysis of three variables (for example, chooser/choosen/criteria). Two scaling models--Carroll and Chang's INDSCAL and Lingoes' PINDIS--are presented and applied to two sets of empirical data. (CP)