Hi,

We've noticed that FDR correction produces different p-values for the left and right hemispheres. We found that this issue was observed previously and an email was sent to the mailing list (see below), however, no answer was received. We wondered if anybody has noticed this problem before and if there are any suggestions on how best to deal with it.

Thanks
Manja

Hi,
FDR correction for multiple comparisons gives you different p-values  for the left and right hemispheres.  It seems really weird to report  in a paper:
a) that you've corrected for the right and left hemispheres  separately and
b) that the p-value that gave rise to a corrected probability of .05  was different for each.
How do people deal with this issue?  Take the most stringent p-value  and use it for both hemispheres?  Just report that FDR adjusted the p- value to .05?

Also, correcting for right and left hemispheres separately assumes  that they are independent observations.  Unless you're studying  callosotomy patients (and maybe even if you are) this doesn't seem  justified.  Are we supposed to be adjusting p-values accordingly?

Thanks,
Dara


--

Manja Lehmann

Research Assistant

Dementia Research Centre

University College London

Institute of Neurology

Queen Square

London WC1N 3BG



lehmann@drc.ion.ucl.ac.uk

Tel: 020 7837 3611 ext. 3962



P Please consider the environment before printing this e-mail



_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer