Hello,
I would like to set up a higher-level analysis in FEAT, a repeated measures 2 X 2 ANOVA.
My study contains 2 patient groups with 2 different conditions for each participant. The
aim is to analyse the group X task interaction.
I have set up the analysis as recommended in an earlier post:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind04&L=FSL&P=R72427
group EV1 EV2 EV3 EV4
group_1_task_A_subj1 1 1 1 1 1
group_1_task_A_subj2 1 1 1 1 1
group_1_task_A_subj3 1 1 1 1 1
group_1_task_A_subj4 1 1 1 1 1
group_2_task_A_subj1 2 1 -1 -1 1
group_2_task_A_subj2 2 1 -1 -1 1
group_2_task_A_subj3 2 1 -1 -1 1
group_1_task_B_subj1 1 -1 1 -1 1
group_1_task_B_subj2 1 -1 1 -1 1
group_1_task_B_subj3 1 -1 1 -1 1
group_1_task_B_subj4 1 -1 1 -1 1
group_2_task_B_subj1 2 -1 -1 1 1
group_2_task_B_subj2 2 -1 -1 1 1
group_2_task_B_subj3 2 -1 -1 1 1
Running this in FEAT, I get the error message: "design matrix uses different groups (or
different variances), but these do not contain "separable" EVs for the different groups".
Replacing EV4 with separate EVs for each subject results in the same error message.
What is wrong with the design matrix? Your suggestions are very much appreciated.
Peter
|