Hi Sophie, yes you're probably right that 2x3 isn't quite the same as
3x2, although presumably the difference really only comes down to
which set of options you want to describe different factors and
levels as fixed or random (see for example the table at the bottom of
http://www.fmrib.ox.ac.uk/fsl/feat5/detail.html#ANOVA3factors2levels ).
I'm afraid that the number of possibilities for ANOVA designs is
endless so we can't provide examples for all of them! But yes it is
always possible to implement any ANOVA design as a GLM (which both
FSL and SPM require) - though often it is easier and more direct to
model what you want explicitly as a GLM rather than thinking via an
ANOVA design.
Good luck!
Steve.
On 24 Jul 2007, at 06:16, Sophie Anisa wrote:
> On 7/23/07, Christian Beckmann <[log in to unmask]> wrote: Hi
>
> have a look at http://www.fmrib.ox.ac.uk/fsl/feat5/
> detail.html#ANOVA3factors2levels
>
> Hi Christian
>
> Having thought about it some more and tried to run some
> models I realised that a 2-factors-at-3 (2x3) levels design
> is completely different to the 3-factors-at-2-levels (3x2)
> example you identified. In my case I need to identify the
> significance of one factor with three levels. This, by itself,
> is like the example of:
>
> http://www.fmrib.ox.ac.uk/fsl/feat5/
> detail.html#ANOVA1factor4levelsRepeatedMeasures
>
> (since I have repeated measures on each subject) but with
> only 3 levels.
>
> I need to combine this with another factor at 2 levels.
> This, by itself, is like the example at:
>
> http://www.fmrib.ox.ac.uk/fsl/feat5/detail.html#ANOVA2factors2levels
>
> except there is only one factor. (So that part of the analysis
> alone would
> effectively be a paired t-test).
>
> But I do not see how to put the two together, and more particularly
> how look at the interaction between the two factors?
>
> The 3x2 example you have pointed out is really a 2x2x2 (ie 3^2)
> case as there are 8 separable conditions. Hence I can't just
> transpose
> my 2x3 design (with 6 conditions) into your example 3x2 case.
> And even if I could it would not answer my questions of interest,
> because I'm really concerned to investigate interactions between
> the 3 stimuli types I'm using.
>
> So I guess my original question still stands - is it possible to
> set up
> a 2x3 model in the FEAT design matrix. And if so, how? Or if not,
> is there an alternative analysis route?
>
> kind regards
>
> Sophie
------------------------------------------------------------------------
---
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
------------------------------------------------------------------------
---
|