On 3 January 2011 02:48, Klara Mareckova <[log in to unmask]> wrote:
> Hello,
>
> I have a two questions about 2by2 ANOVA in FSL
> (http://www.fmrib.ox.ac.uk/fsl/feat5/detail.html)
>
> 1) could somebody please clarify when to use fixed effects and when
> random effects for a 2by2ANOVA?
> Am I correct that fixed effects will be till used only for analysis
> of a single subject whereas random effects are used for a group
> analysis?
>
No, here, fixed and random effects reflects the assumption of whether
the factors you are modelling are fixed (e.g. two treatments/two
diseases) or randomly varying (e.g. disease scores). In a basic 2x2
case you'll be more than likely to be interested in fixed effects, I
think.
> 2) if I would run the random effects
> (fslmaths fstat1 -div fstat3 fstata
> fslmaths fstat2 -div fstat3 fstatb
> ftoz -zout zfstata fstata 1 1
> ftoz -zout zfstatb fstatb 1 1
> You could then do thresholding on zfstata and zfstatb with easythresh.)
>
> I get new outputs: zfstata, zfstatb, fstata, fstatb, and redered_grot.
> But I'm not getting any new cluster zstat reports or thresholded
> zstats. Where I can see what are the significant clusters? (I guess
> the cluster zstat and tresholded zstat are for the fixed effects part
> that is done before - viz " In order to carry this out, first run FEAT
> using the above - FE - design.")
>
You're doing this in the stats directory? Easythresh does not produce
a report. There should be a number of
_grot files (e.g. rendered_grot, thresh_grot.. grot should just be
whatever name you want) from the easythresh output? You can then use
fslview to visualised these images, which are the thresholded images..
Eugene
> Many thanks for your explanation,
>
> Klara
>
>
|