Canonical correlation analysis in counseling psychology research

Discusses several approaches to analyzing data on 2 sets of variables and the advantages of canonical correlation analysis. Previous applications of the method are summarized and a number of suggestions are made for additional uses in counseling psychology. Interpretations of information derived from a complete canonical correlation analysis are presented. Concepts discussed include (a) the canonical variates, (b) BETA weights, (c) canonical correlations, (d) maximum and successive canonical correlations, (e) significance tests, (f) correlations of variables and variates, (g) redundancy, and (h) cross-validation and replication. The presentation assumes, in part, a basic understanding of product-moment correlation. (29 ref.) (PsycINFO Database Record (c) 2006 APA, all rights reserved).