Researchers have been limited in their ability to examine multiple constructs simultaneously due to the constraints imposed by traditional statistical methods. The most notable limitations include the need for a relatively large sample size while restricting the variables to a relatively small number. The application of a newly discovered statistical technique to overcome these limitations, namely, the use of weighted structural regression techniques to generate an empirical Bayes estimate of the covariance matrix as the starting point for a canonical correlation/interbattery factor analysis, is described. How availability of the new method affects research design and analysis is illustrated through a study of adolescent alcohol use. High school students (n=338) completed self-report measures of alcohol use and psychosocial mediators of alcohol use. The canonical correlation/interbattery factor analysis produced five factors, four of which were used for interpretation. The fifth factor did not contain any variable with a sufficient proportion of reliable variance within the criterion set. Analytic results provide extensive information about the reliability of individual items and the proportion of reliable variance of each item within the factor structure. The investigator was able to describe multiple relationships that emerged within the overall factor structure as well as the strength of those relationships. Two figures illustrate the discussion, and five tables show analysis data. (SLD)