Depression study with canonical correlation analysis

Abstract

The purpose of this study is to reduce potential statistical barriers and open doors to canonical correlation analysis (CCA) for applied behavioral scientists and personality researchers. CCA was selected for discussion, as it represents the highest level of the general linear model (GLM) and can be quite easily conceptualized as a method associated with the more widely understood Pearson r correlation coefficient. An understanding of CCA can lead to a more global appreciation of other univariate and multivariate methods in the GLM. We investigate the relationships between the demographic variables and general health variables and want to know (in terms of degree and directionality) what demographic variables were related to what general health variables in this multivariate analysis. The data presented here are from a subset of 294 observations randomly selected from the original 1000 respondents sampled in Los Angeles. In the present study, no very high correlations within a set existed and we found that variable age and education greatly influenced the dependent variables compared to the other. We seek to demonstrate CCA with basic language, using technical terminology only when necessary for understanding and use of the method. We present an entire example of a CCA analysis using SPSS(Version 20.0)