I have been getting an error when trying to run a flexible factorial
analysis with SPM5.
My design is a 2x4 factorial: group (patients vs controls) x the 4 within
subjects conditions of the experiment.
In the design specification I have gone for the subject option, so the first
factor specified is 'subject', the second 'group' and the third 'condition'.
There are 13 subjects in the first group, and 9 in the second.
For each of the subjects in the first group, I specified the vector [1,1;
1,2; 1,3; 1,4], and for the subjects in the second group [2,1; 2,2; 2,3;
2,4], as well as the 4 con images that match the levels of the factors.
I have specified two main effects: factor 2 and factor 3, as well as the
interaction [2 3].
When I run the design, spm returns the following error:
Design configuration :
Error running job: Error using ==> spm_meanby at 31
row dimension mismatch between inputs
Can anyone tell me what is wrong here?
Running a similar analysis (without the subject factor) using the full
factorial design specification option seems to work fine.
Many thanks,
Ans Vercammen
Prince of Wales Medical Research Institute
Barker&Easy St
Randwick, NSW 2031
Australia
+61 02 9399 1258
|