Hi, the degrees-of-freedom are in general the number of timepoints
minus the number of EVs (minus 1 more if you demean the data with -
D). This is constant across all voxels, hence the mapping from t to z
is the same for all voxels so it doesn't make any difference to the
randomise inference (p-values) whether you use the raw t or convert
to z first. You can convert the raw t to z outside of randomise if
you need to. For your case dof=14.
Cheers.
On 31 Mar 2007, at 14:59, Vitaly Napadow wrote:
> steve,
>
> why does randomise output a t-stat and not a z-stat? if we were to
> convert
> these to z-stats, what would be the appropriate dof if we were doing a
> paired t-test type analysis with 2 groups of 15 scans each? dof=28?
> or for
> randomise, is dof calculated differently?
>
> thanks
>
> vitaly
>
>
>> Both options take the same kind of argument; it is the cluster-
>> forming threshold and is in the units of the raw t-statistic image
>> formed by randomise. So for example if you enter 2.5 then clusters
>> above t=2.5 will be formed and tested for significance on the basis
>> of their extent (-c option) or "mass (-C option).
>>
>> Cheers.
>>
>>
>>
>> On 30 Mar 2007, at 20:26, Polly Dhond wrote:
>>
>>> Hi,
>>>
>>> Can you please tell me the format(s) of the input for the -c and
>>> -C flags in randomize are? i.e. example value. Thanks.
>>>
>>> Regards
>>> Polly
>>>
>>>
>>>
>>> --
>>> Rupali P. Dhond, Ph.D.
>>> Dept. of Radiology, MGH
>>> Martinos Center for Biomedical Imaging
>>> Rm 2301, Bldg 149, 13th Street
>>> Charlestown, MA 02129
>>> e-mail: [log in to unmask]
>>
>>
>> ---------------------------------------------------------------------
>> ---
>> ---
>> 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
>> ---------------------------------------------------------------------
>> ---
>> ---
>>
>>
>>
>
>
> |`~~~~~~~~~~~~~\|/~~~~~~~~~~~~~'|
>
> Vitaly Napadow, Ph.D., Lic.Ac.
> Dept. of Radiology
> Massachusetts General Hospital
> CNY 149-2301, 13th St.
> Charlestown, MA 02129
> phone: (617) 724-3402
> fax: (617) 726-7422
> email: [log in to unmask]
>
> |`~~~~~~~~~~~~~/|\~~~~~~~~~~~~~'|
------------------------------------------------------------------------
---
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
------------------------------------------------------------------------
---
|