Print

Print
Your F-contrast and t-contrasts are measuring different things.

Your F-contrast is two rows and is evaluating whether age has a significant effect on either group. Your T-Contrasts are asking if the effect of age (slope of age on Y) is different between the two groups.

The F-contrast that would give you effects in both directions is 0 0 1 -1.

Even then the T- and F- maps will be slightly different. The reason is that in SPM, T-contrasts are one-sided tests meaning the p-value for the T will be half the p-value of the F. The relationship between T and F for single row F-contrasts is F=T^2.

Hope this helps.


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 Wed, Nov 9, 2011 at 8:50 AM, Vitria Adisetiyo <[log in to unmask]> wrote:
Hello SPM experts,

I have a basic question about how to interpret the output of the F vs. T contrast of a 2nd level analysis two-sample t-test. My design is straight forward. I have 2 groups (1st 2 columns of matrix), I included age a covariate (last 2 columns of matrix) and my contrasts are set up as follows:

F contrast:
0 0 1 0
0 0 0 1

T contrasts:
0 0 1 -1 (group 1>group2)
0 0 -1 1 (group 2>group1)

My output always has many voxel clusters that have significant F values even with FDR correction. Since the F contrast is detecting bi-directional group differences in the voxels, I assumed that when I apply the T contrast, the specific directional differences would be extracted out in the relevant t-contrast set up. However, no signficant voxels are detected with either of the t contrasts.

Is my assumption incorrect? Does the F contrast significant results actually telling me anything meaningful? Thanks so much for any input!!!!

-Vee