Issues in the Application of Covariance Structure Analysis: A Comment