Print

Print
If you only have 5 ROI, then you should only correct for 5 tests; although you might be to argue that the regions are independent. Comparing controls to patients is independent of comparing patients to their behavior.

As for the behavioral comparisons. It depends on the goal of the study; additionally, taking the global peak is probably a bad idea as you are comparing different regions between individuals potentially and a single voxel is a poor estimate of the activity.

Best Regards, Donald McLaren
=================
D.G. McLaren, Ph.D.
Postdoctoral Research Fellow, GRECC, Bedford VA
Research Fellow, Department of Neurology, Massachusetts General Hospital and Harvard Medical School
Office: (773) 406-2464
=====================
This e-mail contains CONFIDENTIAL INFORMATION which may contain PROTECTED HEALTHCARE INFORMATION and may also be LEGALLY PRIVILEGED and which is intended only for the use of the individual or entity named above. If the reader of the e-mail is not the intended recipient or the employee or agent responsible for delivering it to the intended recipient, you are hereby notified that you are in possession of confidential and privileged information. Any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited and may be unlawful. If you have received this e-mail unintentionally, please immediately notify the sender via telephone at (773) 406-2464 or email.


On Fri, Mar 4, 2011 at 5:05 PM, James Lee <[log in to unmask]> wrote:
Statistical Savants,

I am unsure about the following multiple comparisons question:

Let's say I've acquired whole-brain fMRI data on patients and
controls. I have behavioral variables on patients only, and run
correlations of five behavioral variables with global measures
extracted from each patient's whole-brain statistical results, such as
peak activation in the entire brain, or volume of activation in some
specified area etc.

In a separate ROI analysis (not a voxel analysis) on the same fmri
data I compare patients and controls for their activation in 5
previously chosen anatomic ROIs.

Should my multiple comparisons correction extend to all 10 p-values of
the five correlations and the five t-tests? I tend to think it should.
Can you mention a reference that will help clarify this situation?

Many Thanks in advance!

Jim

James N. Lee, Ph.D. (Biomed. Eng.), M.S. (Psychology)
Imaging and Neurosciences Center
729 Arapeen Drive
Salt Lake City, Utah  84108
801 581 4228