Hi Lin,
Presumably you are referring to the (variance) Group column on the
setup GUI.
The answer to how many subjects you need in a variance group in order
to be able to estimate that variance is clearly a similar question as
how many subjects you need to do a mixed effect analysis - in other
words, there is no straightforward answer (although some people might
say you need at least 12!). Here it is a murky trade-off between how
many subjects you have in each group (to get good variance estimation)
and how different you think the group variances are.
Here is the relevant bit from the feat web help page:
If you ask for more than one group, each group will end up with a
separate estimate of variance for the higher-level parameter
estimates; for example, if the first 10 inputs are first-level FEAT
outputs from control subjects and the next 10 inputs are first-level
FEAT outputs from patients, you can setup two different groups and
each will end up with its own variance estimates, possibly improving
the final modelling and estimation quality (see examples below for
further clarification). If you setup different groups for different
variances, you will get fewer data-points to estimate each variance
(than if only one variance was estimated). Therefore, you only want to
use this option if you do believe that the groups possibly do have
different variances.
Cheers, Mark.
----
Dr Mark Woolrich
EPSRC Advanced Research Fellow University Research Lecturer
Oxford University Centre for Functional MRI of the Brain (FMRIB),
John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK.
Tel: (+44)1865-222782 Homepage: http://www.fmrib.ox.ac.uk/~woolrich
On 13 Jan 2009, at 23:14, Lin Nga wrote:
> Hi Dr. Woolrich,
>
> Thank you for the response. I've recently realized that I don't need
> the
> second-level and have started running FSL under a similar set-up to
> what
> you've suggested. Although I have one more area of confusion. For my
> inputs,
> should I indicate that they are 4 separate groups (this is what I have
> set-up) or should they be considered one group?
>
> Many thanks,
> Lin
>
|