Abstract
The present report describes a procedure of setting up category systems in which the classes are defined by a group of Ss, rather than by E. The procedure has the advantage over intuitively derived systems, of supplying a measure of the reliability of category definition. Essentially, the procedure consists of separating descriptions of interpersonal behavior according to similarity of meaning by a modification of an associative overlap measure and classifying the overlap measures into categories by a clustering technique.
Get full access to this article
View all access options for this article.
