Authors : Omer C. Bilgin , Ebru Emsen and Michael E. Davis
Abstract: Canonical Correlation Analysis (CCA), as a multivariate technique, has the ability to deal with two sets of multivariate variables simultaneously and to produce both structural and spatial meanings. In the present study, six head and two scrotal measurements from Awassi male lambs at seven months of age were analyzed using CCA to examine the interdependent relationships between multiple head and scrotum variables. Although the analysis derived two statistically significant (p<0.01) canonical variate pairs, only the first pair was considered for interpretation. The canonical correlation between the variates in the first pair was 0.89, exceeding all bivariate correlations between the head and scrotum variables. Canonical weights, canonical loadings and cross loadings were examined to interpret the meanings of the canonical variates in terms of the original variables.
Omer C. Bilgin , Ebru Emsen and Michael E. Davis , 2003. An Application of Canonical Correlation Analysis to Relationships Between the Head and Scrotum Measurements in Awassi Fat Tailed Lambs . Journal of Animal and Veterinary Advances, 2: 343-349.