9 Nov
2015
9 Nov
'15
6:46 p.m.
Hi Doug,
Got it! Sorry that I keep forgetting this. So it’s an exponent statistic.
The reviewers asked for a statistic and df. I think I can convert it to a Z-statistic so it’s more “conventional”.
I also noticed that qdec provides a df when I set up the model. It is valid to use that df to derive the t-statistic?
Best, Daniel
--
Daniel Yang, Ph.D.
Yale Child Study Center
E-mail: daniel.yj.yang@yale.edu
On 11/9/15, 6:02 PM, "freesurfer-bounces@nmr.mgh.harvard.edu on behalf of
Douglas N Greve" <freesurfer-bounces@nmr.mgh.harvard.edu on behalf of
greve@nmr.mgh.harvard.edu> wrote:
>
>
>On 11/09/2015 10:51 AM, Yang, Daniel wrote:
>> Dear FreeSurfer experts,
>>
>> In the output mri_surfcluster (e.g.,
>>
>>https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/GroupAnalysisCluster
>>Summary),
>> what is the statistics behind the Max?
>>
>> Is it either t-statistic (if only 1 line of contrast) or F-statistic
>> (if two lines of contrast), but not Z-statistics?
>it is the maximum of the abs(-log10(p))