Dear Joelle,
This is an effects of interest contrast, you would go with an F contrast with as many rows as you have predictors (ignoring the constant, so eight), and for each of the rows you would weight one of the columns with "1" and the rest with "0", resulting in
[1 0 0 0 0 0 0 0 0;
0 1 0 0 0 0 0 0 0;
0 0 1 0 0 0 0 0 0;
0 0 0 1 0 0 0 0 0;
0 0 0 0 1 0 0 0 0;
0 0 0 0 0 1 0 0 0;
0 0 0 0 0 0 1 0 0;
0 0 0 0 0 0 0 1 0]
I think there was also a simpler way to create effects of interest contrast, but I don't remember. Usually the contrast should be useless though.
Best
Helmut
|