In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (Donald Burrill) wrote:
<snip>
Thanks for the response. Would only that Research Scientists here had time
to take
statistics courses, nothing would please me more. The experiment and primary
analyses are indeed cast in the form of a statistical model, in this case,
a within-subject voxel-wise 2-way ANOVA, with two two-level factors. Main
effects,
interactions, and post-tests have been performed and yielded good results. The
issue under examination is an interesting phenomenon I observed in that data
subsequently, and would like to describe statistically, rather than just
rhetorically. However, as I've said, I've had some trouble with that, and turned
to this news group where I have often had success in obtaining statistical
advise.
Even if that advise is to visit the campus consultant, which I have little
doubt,
will require a charge to one of our grants. If you are indeed familiar with the
consulting service here at Princeton, perhaps you could tell me if this is so?
I realize the problem as presented is rather vague, which owes to my uncertainty
in how to frame the problem in a statistical model to begin with (as you note).
Hopefully, somebody will find the problem description straight forward
enough to
suggest an approach.
Thanks
Jim
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
