Hi,
When you use cluster-based thresholding in randomise, your final
cluster corrected p-values ('maxc') are definitely valid regardless of
what cluster-forming threshold you used. Setting the cluster-forming
threshold is just a matter of preference whether you are weighting
your clustering for smaller, stronger clusters or lower, wider
clusters. This is the main reason why we developed TFCE, which tries
to avoid this arbitrariness - I would recommend trying that (see
recent emails).
Cheers.
On 13 Mar 2008, at 01:26, Jiansong Xu wrote:
> Dear Friends:
>
> I read a DTI paper using TBSS analysis. I noticed that the authors
> used t=1 as voxel level threshold and corrected at cluster level
> (p< .05). If I understood it correctly, t = 1 means that the voxel
> level threshold p > .05, and possibly p > .1. Is it Ok to use such
> voxel level threshold? In fMRI study, my impression is that no one
> use voxel level threshold p > .05, and most authors use p < .01.
>
>
>
> Best
>
> Jiansong Xu, M.D., Ph.D.
> Associate Research Scientist
> Dept. of Psychiatry
> Yale School of Medicine
>
> Mailing Address:
>
> Connecticut Mental Health Center, Room S-100
> 34 Park St.
> New Haven, CT 06519
> Fax: 203-974-7366
>
---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre
FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
---------------------------------------------------------------------------
|