ABSTRACT
Discriminant Analysis (Discrim) provides a statistical approach to investigate the extent to which a set of measured variables can distinguish-discriminate-between members of different groups or distinct levels of another, nominal or possibly ordinal, variable. Discrim is also similar to logistic regression (LR) and cluster analysis (CA). Discrim is a unique application of the multivariate ANOVA family, falling within the general linear model approach to inferential statistics. There are, of course, many other available statistical packages that facilitates Discrim; however, for the sake of providing concrete and consistent guidance, this chapter focuses largely on the conduct of Discrim within SPSS version 21. There are several additional options in Discrim, but the intent is to offer a straightforward approach for getting started toward understanding how to run the analysis and interpret the results. Predictor variables consisting of interval-scale measures of whatever phenomena can be used in attempting to classify individual cases according to their membership in the levels of the grouping variable.
