How to do a multiple ANOVA procedure in STATISTICA:

Ó 2003, Right Brain, Ink.

First, open your data set in STATISTICA. Then under Statistics, highlight ANOVA and then select Main Effects ANOVA.

The following example is of a measure of evaluation of effectiveness of a course of study across group (whether it was an instructor or a student) and across the various courses.

The independent variables are group and course

The dependent variable is the total score.

After clicking OK, you get the following table:

The “error” term is the “within” term.

Group and Course are both the “between” terms.

Ignore the intercept information.

Note, that you get graphs of the two variables:

This shows the significant difference between the two groups of people. The students rated the courses higher than the instructors did.

This graph shows how the courses were evaluated by the instructors and the students put together. There was no difference in the courses:

Now, if one were interested in seeing the INTERACTION (if there was some difference in courses by instructor versus student – say, course one was evaluated higher by students and course two evaluated higher by instructors, etc., then one should do a “Factorial ANOVA” shown below.

First, under Statistics, choose ANOVA and the Factorial ANOVA:

Again, choose the appropriate dependent and independent variables:

Now when you click OK you will get the following table:

The one that says, “Group*Course” is your interaction term.

In this case, the interaction is not significant and the only difference appears to be that between the students and the instructors. Here’s a graph of the interaction (which shows up below the table in STATISTICA).

If there were a significant interaction, you would expect to see more of a “cross-over” effect. These lines all appear to be fairly parallel.