Dear Selene,
what do you mean with "different"? The effect sizes and T values should
be the same, but your p values will be different due to the different
number of degrees of freedom in your analysis.
With regard to the replication options, SPM does not make a difference
between subjects and sessions. If your design matrix is huge, it may be a
display problem, but you should check once by defining a contrast
over all subjects without automatic replication (i.e. do the repmat
yourself). If this gives a different result, I would like to ask you for
a copy of your xml batch files to setup the stats and contrasts via
private email.
Volkmar
On Wed, 4 Apr 2007, Selene Cansino wrote:
> Dear SPMers,
>
> Modeling one single design for each subject should be equivalent to
> modeling several subjects in one single model. However, the results for the
> same subject are different between both procedures. Which one is right?
> This occurs with SPM5 last update and Matlab R2006b.
>
> Besides, the display above the design matrix indicating which conditions
> were calculated in each contrasts are displayed only over the first three
> subjects and not throughout the whole design over all subjects, as
> expected. This occurred when I performed the contrasts in the multiple-
> subject model, employing the repeat contrast in all sessions/subjects
> option. Should I ignore this display and believe that the right conditions
> of each session are taking into account or there is really a problem?
>
> I appreciate your help
>
> Selene Cansino
>
>
> *****************
> Professor Selene Cansino, Ph.D.
> Laboratory of NeuroCognition
> Faculty of Psychology
> National Autonomous University of Mexico
> Av. Universidad 3004, Col. Copilco Universidad
> Building D, 2º floor, Room 12
> México D.F., 04510
>
> Mailing address:
> Apartado postal 70-390
> México D.F., 04511
> México
>
> Phone: 52 55 56 22 23 39
> Fax: 52 55 56 16 07 78
> Email: [log in to unmask]
>
>
--
Volkmar Glauche
-
Department of Neurology [log in to unmask]
Universitaetsklinikum Freiburg Phone 49(0)761-270-5331
Breisacher Str. 64 Fax 49(0)761-270-5416
79106 Freiburg http://fbi.uniklinik-freiburg.de/
|