Welcome to the AFNI/NIfTI Server - AFNI Message Board

History of AFNI updates Suggested reading for beginners Add yourself to the AFNI map!

New Topic | Go to Top | Go to Topic | Search

Re: about '-contr' in 3dANOVA

Author: rick reynolds (---.nimh.nih.gov)
Date: 10-28-09 09:47

Hi Pengmin,

A mask in 3dANOVA is optional. But the fact that you did not use
one and still had such a high q-value worries me. You need the
q-value down at least near 0.05 (you don't want most of the results
expected to be false). What is the p-value when you set the q to
0.05 (by raising the threshold)?

The p-value is not corrected by FDR, as far as I know. The p-value
is the voxelwise probability of a false positive. The q-value is
the expected fraction of false positives (among those voxels that
survive the threshold). So you would not state a corrected p, but
probably both the p and the q values.

- rick

Reply To This Message | Flat View

Newer Topic | Older Topic

Topics