We are struggling with designing the GLM for a GLM for 3 (between subjects) x 2 (within subjects) repeated measures ANOVA.
We have 3 groups of patients with different diseases (between subject group factor) each of whom attend for 2 visits - once to receive placebo and once to receive drug (within subject drug factor).
We are interested in finding significant whole brain results for drug x group interaction, and overall effect of drug independent of group.
Looking at GLM FSL webpage we can see how to do a GLM for 2 (between subjects) x 2 (within subjects) repeated measures ANOVA i.e. with only 2 groups, where one has to omit the group EV to avoid rank deficiency, and so just include 1 EV for drug, 1 EV for drug x group interaction, with additional EVs for each subject to do the pairing.
But how do we do this for 3 groups?
We could include 1 EV for drug and then have 2 EVs for drug x group interaction (calculated by using the 2 EVs from for a triple t-test design multipled by the drug EV), but then how do we do the contrasts to find overall drug x group interaction (does this need to use F-tests somehow)?
Thanks to community for help with this.
Tony Goldstone
Division of Brain Sciences
Imperial College London
[log in to unmask]
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
|