Hi,
>obtaining a r value and a scatter plot
Yes, you can do that (i.e. assess the temporal correlation of the Tmode for a given IC with your time series model in the GLM matrix).
>Smodes
...quantify the response strength for a given IC for each subject/session. This is different from the above.
Cheers,
Andreas
________________________________
Von: FSL - FMRIB's Software Library [[log in to unmask]] im Auftrag von Roser Sala [[log in to unmask]]
Gesendet: Mittwoch, 13. Juli 2011 20:37
An: [log in to unmask]
Betreff: [FSL] melodic group analysis
Hello,
we are performing a group analysis using MELODIC and we introduced a GLM matrix with an external variable (task performance) in order to perform post-hoc statistics.
We found that our IC of interest is highly correlated with task performance. Thus, which is the best way to quantify the strength of this correlation by obtaining a r value and a scatter plot?
We've thought about putting the vector into SPSS, but we still don't know if we are really imitating the statistics done within melodic.
Many thanks
Roser
|