General ANOVA/MANOVA and GLM Notes - Multiple Dependent Measures (MANOVA)

ANOVA/MANOVA and GLM will automatically perform a multivariate analysis of variance when multiple dependent variables are specified (i.e., dependent variables that are not levels of repeated measures factors). For example, if you specify a between-groups design with three dependent variables, a multivariate analysis of variance (MANOVA) with three dependent variables is automatically performed. However, if in addition a within-subjects factor with three levels is specified, then a univariate analysis of variance is performed because the three dependent variables are interpreted as representing the levels of the repeated measures factor. If six dependent variables are specified, then once again a multivariate analysis of variance is performed, with 2 dependent variables for each level of the repeated measures factor.

In general, ANOVA/MANOVA and GLM will divide the number of dependent variables that are specified by the product of numbers of levels of all repeated measures factors. If the result of this division is greater than 1, then a multivariate analysis of variance is performed (if the result of the division is not an integer value, then an error message will be displayed). For example, if a design is specified with 2 repeated measures factors, the first with 3 levels and the second with 2 levels, then at least 6 dependent variables must be specified (since there is a total of 2 x 3 = 6 levels). However, if you specify 12 dependent variables, then STATISTICA will perform a multivariate analysis of variance with 2 dependent variables (since 12 divided by 6 is equal to 2). Note that if you have multiple within-subjects (repeated measures) factors, you will need to use the GLM module.

In all other respects, the analysis of designs with multiple dependent variables (i.e., calculation of major effects and planned comparisons) proceeds in exactly the same manner as in the univariate case. However, when entering contrast coefficients for a planned comparison, you also have the option of entering contrasts for the dependent variables. For example, you may choose to ignore some of the dependent variables in the comparison by assigning 0s (zeros) to them when specifying the contrast. In order to specify contrasts for the dependent variables, select the Yes option button in the Contrasts for dependent variables group box on the GLM, GRM, and ANOVA Results - Comps tab.

Refer to Notes (repeated measures) for more detailed examples of how to specify multivariate repeated measures designs.

See also, ANOVA/MANOVA Introductory Overview, General ANOVA/MANOVA and GLM - Notes, Methods for Analysis of Variance, General Linear Model (GLM), General Regression Models (GRM), Variance Components and Mixed Model ANOVA/ANCOVA, and Experimental Design (DOE); to analyze nonlinear models, see Generalized Linear/Nonlinear Model (GLZ).