Hi Mayuresh,
you might want to use the "threshold and transform spm-t/F-maps" option
as implemented in the VBM8 toolbox
(http://dbm.neuro.uni-jena.de/software/) which allows you to calculate
effect size measure 'd' for the respective spm-volume. I hope this helps.
Best regards
Raphael
Am 27.05.2011 08:21, schrieb Mayuresh K:
> Hello spm experts,
>
> Could anyone guide me how to do a power analysis using fMRI or VBM
> data? I have performed a random effects analysis comparing two groups
> of subjects using their fMRI and VBM data. I have found a number of
> regions with significant differences for both data sets.
> I can probably calculate the mean and standard deviation of the
> %signal change (or volume) for each group for the peak voxels say in a
> particular anatomical region. Then use these to calculate effect sizes
> and power - Is this a valid approach?
> Any other suggestions welcome.
>
> Thanks,
> Mayuresh
>
|