Thanks. Is the df dependent on the number of subjects? For example, I
have 16 subjects in one group, I did the analysis of correlation
between FA and RT, is the dr = 15?
Best
Jiansong Xu, M.D., Ph.D.
Associate Research Scientist
Dept. of Psychiatry
Yale School of Medicine
Connecticut Mental Health Center, Room S-100
34 Park St.
New Haven, CT 06519
On Feb 12, 2008, at 12:45 PM, Gwenaëlle DOUAUD wrote:
> Hi again!
>
> --- Jiansong Xu <[log in to unmask]> a écrit :
>
>> Dear Friends:
>>
>> In fMRI, when we state stats threshold, we usually
>> say; voxel level
>> threshold < .01 (or .05), not corrected, extent
>> threshold p < .05
>> corrected. In tbss, if I use randomise -c to correct
>> multiple
>> comparisons, the voxel level threshold is determined
>> by the t value
>> after the -c option (my understanding). My question
>> is how to
>> calculate the t value corresponding to voxel level p
>> < .05 or p < .01?
> http://faculty.vassar.edu/lowry/tabs.html#t
> (one-tailed P-value). Then all you need to know are
> the degrees of freedom in your statistical model...
> Alternatively, you can just report in your manuscript
> the t-value you have chosen (it's always arbitrary
> anyway, even if 0.05 or 0.01 are somehow more
> evocative than 2.4 for instance) and the DOF.
>
> Cheers,
> Gwenaelle
>
>
>
> _____________________________________________________________________________
> Ne gardez plus qu'une seule adresse mail ! Copiez vos mails vers
> Yahoo! Mail http://mail.yahoo.fr
|