Help System (web edition)

The effect of grouping variables on several dependent variables may
be examined by the multivariate analysis of variance. The dependent
variales are given by the parameter DEPENDENT. The grouping
variables are defined in the same way as in the univariate case, e.g.
    ANOVA IEADATA,20
    DEPENDENT=Knowldge,Comprehn,Applicat,Mentalpr
    GROUPING=Attitude,Sex  Sex=1(Boys),2(Girls)
The interactions in the model may be defined by parameters
INCLUDED or EXCLUDED as in univariate case. There are two ways
in which hypotheses may formed depending on parameter HFORM:

HFORM=0: main effect A is tested after the main effect S
         is included in the model, interaction AS is ignored
HFORM=1: instead of testing main effect A alone, the interaction
         AS is included in the hypothesis as in the univariate case
         (DEFAULT).

Predicted values of dependent variables or residuals cannot be
computed or saved yet!

Further information:
    1 = More on definitions of grouping variables 
    2 = More on main effects and interactions in the model 
    M = Multiple comparisons of mean vectors 
    G = Multivariate analysis of repeated measures 
    A = More on ANOVA operations