> I have analyzed my data ( used with both 3dANOVA3 and GroupAna). The
> results form these two programs are different when i look at maps: the value of
> threshold is much much higher with GroupAna, although the intensities are the
> same. So i don't know which one should be used.
Yes, this can happen especially when sphericity violation is severe as explained here:
<[afni.nimh.nih.gov];
So in such a situation 3dANOVA/2/3 are definitely superior to GroupAna.
> Here is number for Min and Max threshold:
> the result from GroupAna: -12717.84 25105.98
> the result form 3dANOVA3: -6.218 3.589
>
> the intensities are the almost same: -0.868122(-0.868122) 0.435057(0.435055)
Don't worry about the range of t or F in GroupAna since dividing by 0 is handled differently in the two programs. It is the specific t or F at each voxel and the activation maps that matter here.
Gang