Dear Matthew,
Thanks for your suggestion. How do you think about changing the Degree of
Freedom for T test to the equivalent higher value in SPM.mat? The DOF should
be much bigger after pooling the variance. Could you give some suggestion on
how to calculate an equivalent DOF? Thanks a lot.
Sincerely,
Yanjun
-----Original Message-----
From: Matthew Brett [mailto:[log in to unmask]]
Sent: Monday, September 11, 2000 10:02 AM
To: Xu, Yanjun
Cc: [log in to unmask]
Subject: Re: Pooled variance III
Dear Yanjun,
> We're really interested in the pooled variance as we're doing PET
analysis.
> My quesion is:
>
> As mentioned by Matthew, the corrected statistics within SPM interface are
> no long valid after using the Worsleyized image, how to get the valid
> statistics? Does someone have some code for it?
I am afraid I don't. The SPM routines will give corrected p values for Z
fields (see spm_P.m). What you need therefore are the appropriate values
for
the smoothness, which I think will need some hacking to the spm_spm.m file,
and I have never looked into this. Maybe Andrew has some thoughts?
Best wishes,
Matthew
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|