Discriminant analysis is a statistical technique used to compare effectiveness, understand the profits and losses, and set the best practices and recommendations. The target of this analysis is to determine the detectable difference between products. The Discriminant command in SPSS performs the canonical linear Discriminant analysis. It is based on linear combinations of the predictor variables. The Discriminant function analysis, linear function analysis are all generalised terms of Fisher's linear Discriminant. LDA is very closely related to ANOVA and regression analysis. In ANOVA there is the use of categorical independent variables and a continuous dependent variable, whereas Discriminant analysis has continuous independent variables and a categorical dependent variable. The analysis is done by creating one or more linear combinations of predictors. The purpose of Discriminant analysis is to maximally separate the groups and predict a classification based on the independent variables.
A Discriminant score can be calculated based on the weighted combination of the independent variables Di = a + b1x1 + b2x2 +…+ bnxn
Here, Di is the predicted score (Discriminant score)
x is predictor and b is the Discriminant coefficient
The assumptions of Discriminant analysis are very same as those for MANOVA.
Discriminant analysis is to predict group membership from a set of predictors. Thus, DA is MANOVA turned around.
Research question: Do PERF, INFO, VERBEXP permit the classification of students by types of learning disabilities?
Open SPSS then: