Dear all,

I am comparing 2 groups (patients [n=14] and controls [n=30]), in 2 conditions (congruent and incongruent), i.e a 2x2 mixed-model ANOVA, and I'd like to make sure if I am doing it correctly. 

In my design matrix (as attached):
- columns 1 to 44 represent each subject (1 to 14=patients; 15 to 44=controls)
- column 45 represents condition (congruent=1; incongruent= -1)
- column 46 represents the group x condition interaction
- lines 1 to 28 represent the patients group (1 to 14=patients-congruent condition; 15 to 28=patients-incongruent condition)
- lines 29 to 88 represent the control group (29 to 58=controls-congruent condition; 59 to 88=controls-incongruent condition)

Additionally, the contrast for main effect of condition would be either:
[0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0] (for congruent > incongruent condition); or
[0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 -1 0] (for congruent < incongruent condition)

Are the design matrix and contrasts correct?

Any help would be appreciated,

Carolina

Hi Carolina, 

the design matrix and contrasts all look correct to me. 

You may also want to add an additional contrast to test for an interaction effect, but that's up to you. 

best,
Andrew
Originally posted by Carolina Ogawa:

Hi Andrew

Thank you very much for your reply and for making such a wonderful tool available!


Best,
Carolina