How are you grouping the Marsbar output to calculate the means in your external program?
If you used Marsbar to extract from the SPM.mat file, then the order of the output is the same as the order of scans in the file, which is:
Subject1Scan1
Subject1Scan2
Subject2Scan1
Subject2Scan2
.
SubjectNScan1
SubjectNScan2
Not (as you might be thinking)
Subject1Scan1
Subject2Scan1
.
SubjectNScan1
Subject1Scan2
Subject2Scan2
.
SubjectNScan2
-Mike
From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]] On Behalf Of Thomas Ditye
Sent: Monday, June 06, 2011 10:34 AM
To: [log in to unmask]
Subject: [SPM] paired ttest - stats
dear spm list.
in a second level paired ttest analysis of my fmri data, spm gives me a significant cluster of voxels. in order to look at the real values per subject I extracted the data from the significant peak voxel using marsbar. now looking at these data it turns out that the means of the two conditions are almost identical and running a paired ttest in matlab/spss comes out nowhere near significant. I am wondering how exactly spm is calculating it's stats. I also noticed that the paired ttest con_???? image does not correspond to the difference in means. however my input con images from the first level analysis do correspond to the means of the extracted data.
I would really like to understand how spm calculates the paired ttest. if someone could point me in the right direction, that would be great.
thank you very much, I appreciate your help.
thomas
**********************************************************
Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues
|