Dear all,
I've been digging in the forums but not been able to find an answer
to what is probably a relatively simple question:
Say one has a second level repeated measures ANOVA model where one
can specify a contrast that, for statistical reasons, can only be
specified here. The objective is to go one step above and make a
contrast of these second level contrasts. As there's no straight forward
way of doing this with ConMan weightings, it seems necessary to create
an SPM_T by manually contrasting the second level con*.img files using
ImCalc and calculating the T map.
Now, I've figured out that this should involve dividing the
con*.img with sqrt(ResMS.img), but my dummy examples give different
values than spmT's specified by ConMan, leading me to suspect I'm
missing something. Could someone please supply the procedure by which
this should be done?
Best,
Haakon
|