Hi,
On 8 Nov 2006, at 08:42, Soohyun Cho wrote:
> Thank you always for your help, Steve.
>
> Can I ask one more follow-up question?
> I need to make sure I understood your answer to my question #2.
>
> For example, if I set the cluster p threshold to 0.05, and I had 3
> contra
> sts in my design (without
> any F test), would fsl automatically adjust the threshold to 0.05/3
> in or
> der to compensate for the
> number of contrasts run on each cluster? or is it just using 0.05
> as a th
> reshold?
FEAT does not correct for multiple contrasts; each is treated as if
it were run in isolation.
Cheers.
>
> Thank you so much!
>
> Soohyun.
>
>>> 2. When correcting for multiple comparisons, does feat take into
>>> consideration the # of contrasts run as well as the # of
>>> independent voxels (or blobs)?
>>> If yes, does feat take into the # of contrasts that were turned off
>>> along with the ones that were turned on?
>>
>> Good question - no, correcting for multiple contrasts is up to the
>> user. So for example if you feed 5 contrasts into one f-test and
>> don't look at the individual contrasts, then there's no need to
>> correct for multiple comparisons.
>>
>> Cheers, Steve.
>>
>>
>>>
>>> Thank you for your help,
>>> Soohyun.
>>
>>
>> ---------------------------------------------------------------------
>> ---
>
>> ---
>> Stephen M. Smith, Professor of Biomedical Engineering
>> Associate Director, Oxford University FMRIB Centre
>>
>> FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
>> +44 (0) 1865 222726 (fax 222717)
>> [log in to unmask] http://www.fmrib.ox.ac.uk/~steve
>> ---------------------------------------------------------------------
>> ---
>
>> ---
------------------------------------------------------------------------
---
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre
FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
------------------------------------------------------------------------
---
|