On Mon, Apr 26, 2010 at 9:12 AM, Gabor Perlaki
<[log in to unmask]> wrote:
> Hi all,
>
> What is the lowest Z for a group analysis if I use cluster-like thresholding? I've some results for a group with Z=1.8, P=0.05. The probability for Z=1.8 is P=0.0359, it is less than 5%, but I don't know I can believe it or not.
If the z refers to the voxelwise threshold used to create clusters,
then there's no set minimum as long as the clusters subsequently pass
the required size (or mass) threshold required to ensure a familywise
error of 0.05, as computed using random field theory or permutations.
This should be clearly stated in the method and/or results section of
any publication.
That said, using an initial voxelwise threshold that is very low can
often reveal very large clusters of activation across wide brain
regions that are probably not very informative.
-Tom
|