Hello there:
I've run a higher-level analysis in which I fed in 3 cope images for each
of 16 subjects. I'm exactly duplicating the tripled paired t-test as
described at
http://www.fmrib.ox.ac.uk/fsl/feat5/detail.html#TripledTwoGroupDifference
except that I asked for an F test across contrasts 2 and 3.
The dfs reported for the F test are correct-- 2 df for the main effect (3
conditions - 1) in fdof1_f1.nii.gz, and 30 error df (2*(n-1) = 2*15 = 30)
in fdof2_f1.nii.gz.
However, the tdof1_t1.nii.gz files indicate df = 30. Shouldn't that be n-
1 thus df = 15 for my 16 subjects? I assume I'm missing something obvious
and would really appreciate it if someone could point it out!
Thanks so much,
Heather
|