Original Publication Date
Date of Submission
This paper describes and demonstrates Canonical Correlation Analysis (CCA) with orthogonal rotation to facilitate interpretation. The purpose of CCA is to explain the relationship between two or more sets of variables. CCA can be thought of as a kind of principal components analysis on two set of variables, except that the criteria for the pairs of linear combinations is that they have the highest possible correlation while being orthogonal to “earlier” pairs. Social work researchers should rarely be satisfied with a strategy that determines which sets of variables to model on purely statistical grounds. However, there are times when there is a need to measure of association between two or more sets of variables that cannot be ordered theoretically. In this situation, CCA may be appropriate. The strengths and limitations of CCA are discussed.
Dattalo, P. (2014). A demonstration of canonical correlation analysis with orthogonal rotation to facilitate interpretation. Unpublished manuscript, School of Social Work, Virginia Commonwealth University, Richmond, Virginia.
Is Part Of
VCU Social Work Publications