Hi,
Randomise should output at the start of each contrast the number of permutation required for an exhaustive test - eg.
"70 permutations required for exhaustive test of f-test 1"
If this number doesn't match equal 840 for your contrast then we can investigate further..
Many Regards
Matthew
>
> I think this is simply because you've got a small sample of 8 subjects and, depending on which contrast you're looking at, the number of permutations required for exhaustive test are different.
>
> I'm not sure though how 840 permutations are exhaustive for correlations in 8 subjects (cf end of randomise manual for counting the number of permutations). Matt, Tom, how is this calculated if not n! ?
>
> Hope this helps,
> Gwenaelle
>
>
>> Hi
>> I have been trying to run tbss randomise (for correlation
>> between a group of 8 subject's DTI data and their
>> behavioral scores) for some reason the TBSS stops after 840
>> permutation on a certain t -test (I have no idea which one
>> but it looks like it's the same one every time) the stats
>> directory then contains all the required tbss files
>> but i don't know if it means anything. the same thing
>> happened for the same Glm with deifferent dffusion devices
>> (FA and L23) however it didn't haqppen when i run the
>> randomise with a different Glm containing different test
>> scores of the same 8 subjects
>>
>> what can be the reason and how can i fix it?
>>
>> attached is the Glm used when the permutaions stopped in
>> case it is the problem.
>> thanks
>> reut
>
> --------------------------------------------------------------------
>
> Gwenaëlle Douaud, PhD
>
> FMRIB Centre, University of Oxford
> John Radcliffe Hospital, Headington OX3 9DU Oxford UK
>
> Tel: +44 (0) 1865 222 523 Fax: +44 (0) 1865 222 717
>
> www.fmrib.ox.ac.uk/~douaud
>
> --------------------------------------------------------------------
>
>
>
>
|