Dear Users,
I am sure you are growing tired of clarifying mixed model designs, but having gone through the listserv I am still unsure of the correct way to specify my 2(session; within) x 2(group; between) design. I would like to examine the group*session interaction, and so I believe that I will be able to use randomize for this purpose?
I have attached screenshots of what I think may be a suitable design for this, with a dummy situation in which I have 2 subjects in each group, each with two sessions of data. Again, I'd like to look at the group*session interaction, and so I've followed the framework outlined on the GLM FslWiki. I've then specified my contrasts as F-tests as I want them to be two-tailed.
My concerns stem largely from the various approaches I've seen people take within this forum...I've attached another screenshot of what appears to be an approach for my situation yet features a quite different design matrix.
I would appreciate any feedback regarding the correct design specification and contrasts. If my example is correct, terrific, but otherwise any advice on the ideal way to perform this assessment would be great.
Thank you,
Ben
|