Hi Douglas,
I checked but as far as I see it I used the same models. I attached the SPSS
model and output as wel as the FSGD file and contrast file (to test a diagnosis
x age interaction). Now that I come to think of it, I think the reason for the
discrepant findings is that I used a t-contrast instead of an F-contrast for
the interaction in Freesurfer. Apologies I did not think about this earlier.
But maybe there is something else I did completely wrong.
What do you think?
If it is indeed the t vs F contrast, than how should I specify that the
contrast is an F-contrast? Is it ok to just add another line to the
contrastfile (so0 0 0 0 0 0 0 0 1 1 1 1 -1 -1 -1 -10 0 0 0 0 0 0 0 -1 -1 -1 -1
1 1 1 1)?
Thanks again,
Maaike
> Date: Mon, 26 Jan 2015 11:23:59 -0500
> From: gr...@nmr.mgh.harvard.edu
> To: freesurfer@nmr.mgh.harvard.edu
> Subject: Re: [Freesurfer] average values per cluster
>
>
> We get these kind of reports occasionally. When I ask people to confirm
> that they use exactly the same design matrix in SPSS, I never hear back,
> so I assume that it gets resolved. So please check. The other thing you
> can do is to run in matlab, something like
>
> cd glmdir/contrast
> X = load('Xg.dat');
> y = load('ocn.dat');
> C = load('C.dat');
> [beta rvar] = fast_glmfit(y,X);
> [F p] = fast_fratio(beta,X,rvar,C);
> p will be the p-value
>
> If you used an unsigned cluster-forming threshold (ie, abs), then it is
> possible that some of the voxels are pos and some are neg so that they
> average out
>
> doug
>
>
>
>
> On 01/26/2015 09:03 AM, maaike rive wrote:
> > Dear Freesurfer experts,
> >
> > Sorry to bother you again, but I have two more questions about
> > extracting (thickness/surface/GI) values from a certain cluster.
> >
> > As I understood, the abs.y.ocn.dat file gives the average values for a
> > given significant cluster (e.g. a cluster where there is a significant
> > AxB interaction).
> >
> > I may be completely misunderstanding things, but if I use these values
> > in SPSS for further statistics and test the same interaction (AxB),
> > than according to SPSS this interaction is /not /significant
> > (corrected for the same covariates as in the FSGD file).
> >
> > Could you tell me what is going wrong here? I do not trust my results now.
> >
> > Furthermore, is it possible (and if so, how?) to extract the average
> > values of exactly the same cluster, but in an independent group not
> > used in the analysis, for post-hoc comparisons in SPSS?
> >
> > Thanks,
> >
> > Maaike
> >
> >
> > _______________________________________________
> > Freesurfer mailing list
> > Freesurfer@nmr.mgh.harvard.edu
> > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
>
> --
> Douglas N. Greve, Ph.D.
> MGH-NMR Center
> gr...@nmr.mgh.harvard.edu
> Phone Number: 617-724-2358
> Fax: 617-726-7422
>
> Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
> FileDrop: https://gate.nmr.mgh.harvard.edu/filedrop2
> www.nmr.mgh.harvard.edu/facility/filedrop/index.html
> Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/
>
> _______________________________________________
> Freesurfer mailing list
> Freesurfer@nmr.mgh.harvard.edu
> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
>
>
> The information in this e-mail is intended only for the person to whom it is
> addressed. If you believe this e-mail was sent to you in error and the e-mail
> contains patient information, please contact the Partners Compliance HelpLine
> at
> http://www.partners.org/complianceline . If the e-mail was sent to you in
> error
> but does not contain patient information, please contact the sender and
> properly
> dispose of the e-mail.
>
FSGD_diagnbystate_ex.txt~ copy
Description: Binary data
0 0 0 0 0 0 0 0 1 1 1 1 -1 -1 -1 -1
SPSS model and output.docx
Description: MS-Word 2007 document
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.
