Dear Helen,
probably everything is ok with your data. SPM estimates error covariance
based on your design and factor levels. In your case, you decided that
error variance should be unequal between group 1 and 2. After model
estimation, SPM will display the design matrix whitened with the error
covariance filter.
Volkmar
On Mon, 15 Oct 2007, Dr H.C. Slawik wrote:
> Dear all, sorry to bother you again about this. Is there really no one out
> there who has an answer to this?:
>
>
> I have a probably very basic question, though I could not find any advice in
> the Manual. I did a two sample t-test on my structural MRI data (n(1,2) = 11
> and 12, respectively) and seem to get some significances above FDR 0.05.
> However, the display of the design matrix in the results window looks very
> strange (please, see attachment). Instead of the usual arrangement of a white
> and a black box and in the next row a black and a white box, the design
> matrix figure displays a white and a black box and in the next row a black
> and a GREY box. I have never seen this before and wonder if this is an
> indication that something in the test design is not right? Any help would be
> very much appreciated.
>
> Helen
>
>
--
Volkmar Glauche
-
Department of Neurology [log in to unmask]
Universitaetsklinikum Freiburg Phone 49(0)761-270-5331
Breisacher Str. 64 Fax 49(0)761-270-5416
79106 Freiburg http://fbi.uniklinik-freiburg.de/
|