Hi Steve,
I'm confused about the 2x2 ANOVA setup. I have two factors with two levels
each (both are within-subject manipulations). Each subject has 3
runs/sessions of data, and these have been combined in a separate mid-level
FE analysis for each subject. I thought the example on the webpage was what
I needed to do
(http://www.fmrib.ox.ac.uk/fsl/feat5/detail.html#ANOVA2factors2levels);
however, in this thread, it seems you suggest removing the subject means?
Also concerning the webpage example, if there are 8 subjects and 4 different
copes for input (A1, A2, B1, B2), should there be 32 inputs instead of 8?
With this example (and some others), it's unclear how the different subject
inputs are corresponding to different copes/conditions that I thought had to
be entered separately into the model. This is especially true when there are
multiple conditions in an ANOVA and the number of inputs equals the number
of subjects like this example.
I must be missing something...
Thanks for clarifying,
David
-----Original Message-----
From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf
Of Steve Smith
Sent: Wednesday, September 02, 2009 2:16 AM
To: [log in to unmask]
Subject: Re: [FSL] ANOVA results
Hi - this does look really weird - never seen anything like that. I
don't think it's related to the model, which looks fine, but more
likely to something odd in the data.
Do you want to upload the whole .gfeat output and we'll take a look,
Please upload the files in a single compressed tarfile to
http://www.fmrib.ox.ac.uk/cgi-bin/upload.cgi
And then email me the upload ID.
Cheers.
On 1 Sep 2009, at 20:40, Jonathan Hakun wrote:
> Hi all -- I was wondering if someone could help me interpret the
> results from
> a recent 2 x 2 repeated measures ANOVA I ran...
>
> Attached is the design image and an example of a resultant fzstat
> overlay
> that I'm a little confused by. (I apologize for the quality of the
> fzstat,
> had to reduce quality to fit in this message)
>
> Does anyone know why I might be getting this type of "outlining" in
> the
> fzstat image? The contrasts of each level and the interaction look
> fine
> (normal clusters) but
> all 3 of my fstats look like outlines of clusters...
>
> I opened the fzstat map in fslview and played with the
> thresholds...it fills
> the clusters in a little to drop the minimum threshold and increase
> the
> maximum, but some voxels have odd values, like '-inf' for example.
>
> Thanks!
> ~Jon
> <design.png><rendered_thresh_zfstat1.jpg>
---------------------------------------------------------------------------
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
---------------------------------------------------------------------------
|