Dear SPMlers,
supposse the following study:
Group A: 12 patients
Group B: 12 controls.
Each groups performs Task I and II.
We are interested in task specific effects withingroups and group
differences between tasks.
To analyse within group effects I perform two fixed effects analysis,
for each group seperately with 12 subjects, compute subjects specific
con-images and feed them into a second level, Rx-analysis. So far so
clear.
But what if I want to know something about the group differences for
Task A? I see two possibilities: Either I take the con-images of each
group specific Fx-effects analysis. OR I do a third Fx-Effects analysis
with all 24 subjects in one group. The main difference is that global
normalisation in the first case is group specific in the second case
not. But isnt the second case, i.e. to make a new Fx-effects analysis
not the correct way because I want to compare both groups?
Thanks for an advice
H. Walter
--
Dr. Dr. Henrik Walter
Abteilung Psychiatrie III
Universitätsklinikum Ulm
Leimgrubenweg 12
89075 Ulm
fon: 0731/502 1489
fax: 0731/502 1549
e-mail: [log in to unmask]
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|