Hi - it's probably just that with this design FEAT can't estimate
different variance groups - so just set the "group" to 1 for all
datapoints.
Cheers.
On 27 May 2009, at 23:30, Peter Soros wrote:
> 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
>
---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre
FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
---------------------------------------------------------------------------
|