Dear Karsten,
Thank you for your reply.
That's what I thought first because Global Normalization is helpful
between subjects.
However, the spm_RandFX.man page describes the following at page 2:
"Global normalization and grand mean scaling should have been handled at
the within-subject level, so there is no need for such things at the
between subject level. Similarly, the contrast image produced are
implicitly masked, so there is no need to redefine the analysis volume
in the second level analysis."
It does not seem to be consistent with my thought.
Any clarification please?
Hyo Jong
-----Original Message-----
From: [log in to unmask] [mailto:[log in to unmask]]
Sent: Tuesday, February 08, 2005 10:20 AM
To: HyoJongLee; [log in to unmask]
Subject: Re: [SPM] Global scaling and grand mean
Dear Hyo Jong,
> I have a basic question about global scaling and grand mean.
> I searched archives and found similar questions asked without
> clear answers to them.
>
> Should I set 'Global scaling' option during the first level
> analysis, if I plan to apply results to second level analysis
> for group comparison later?
> The scale of baseline from each subject are not same. Thus,
> I am confused whether the first level analyses should have
> the global scaling option or spm2 handles this matter by default.
> Also, what is purpose of setting 'grand mean' to a specific value?
I would recommend to skip the scaling on the first level, and skip it
definitely on the second level.
Your session/subject regressor, assuming that you don’t have scaling
differences within one session, captures the different baselines on the
first level.
The scaling on the second level is only useful, if your are going to
analyses other images than con_images; MRI-Perfusion images or PET images,
for example. Here, it is necessary, that the global mean is the same across
subjects.
In case of a 'classical' second level analysis, you are analysing the
con-images, which are already appropriately scaled, i.e. positive values are
reflecting a signal increase, negative values a decrease and the value
itself represents somewhat like the strength of the effect.
Good luck,
Karsten
--
-------------------------------------------------------------
Karsten Specht, PhD
Department of Biological and Medical Psychology
University of Bergen
Jonas Lies vei 91
5009 Bergen
Norway
Tel.: +47-555-86279
Fax: +47-555-89872
[log in to unmask]
http://fmri.uib.no/
|