Monte Carlo methods were employed to investigate the effect of nonnormality on the bias associated with the squared canonical correlation coefficient (Rc2). The majority of Rc2 estimates were found to be extremely biased, but the magnitude of bias was impacted little by the degree of nonnormality.
Leach, Lesley F. and Henson, Robin K.
"Bias and Precision of the Squared Canonical Correlation Coefficient Under Nonnormal Data Condition,"
Journal of Modern Applied Statistical Methods:
1, Article 8.
Available at: http://digitalcommons.wayne.edu/jmasm/vol13/iss1/8