Hello all,
I am trying to choose the best conjunction analysis method to implement on my data. While I am using multiple regression on the second level I faced with an error while trying to set t-test contrast[1 0 ] and [0 1].What I have done(SPM5) is exactly like this:
Under the multiple regression design I choose my scans which are(2 group of 14 Con* ) and for covariates, I defined two covariate under the name of regression1 and regression 2 with the following vectors respectively: [1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0]' and [0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1]' .I also set the centering as overall mean. everything goes well until I suppose to define the contrast. While I define the contrast(t-contrast) like [1 0] or even [1 0 0] it gives me an error like invalid contrast (my model consist of constant term so it has three column on design matrix).Can you tell me if I did anything wrong? Is it the problem of constant term? if yes, how can I get rid of it in SPM5?
the second question is related to the method which has been implemented by Nichols, As it was stated in his paper he used Annova method.Is that the one that can be done one second level using full-factorial design?any clue for implementation in SPM5 will be appreciated.
Thanks in advance.
Alireza
---------------------------------
Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now.
