Dimensionality of Hierarchical and Proximal Data Structures
David J. Krus, Patricia H. Krus · Applied Psychological Measurement · 1980
The coefficient of correlation is a fairly general measure which subsumes other, more primitive re-lationships. At the fundamental classification level, similarities among objects and cladistic relation-ships were conceptualized as generic concepts un-derlying formation of proximal and hierarchical structures. Examples of these structures were iso-lated from data obtained by replicating Thurstone’s classical study of nationality preferences and were subsequently interpreted. The search for structure among the elements of data matrices is dependent on often tacitly as-sumed classifications of relationships to be ana-lyzed. One of the more frequently assumed rela-tionships is that of similarity and dissimilarity, as indexed by the (positive and negative) coeffi-cient of correlation. The classifactory principles inherent in the concept of correlation can be illustrated by con-sidering Equation 1 I derived around the turn of the century by Pear-son (1902, p. 292). Here, the product-moment coefficient of correlation is conceptualized in terms of added variances of the variables X and