Hi,
I get the "invalid contrast" message when trying to look at the group
effect on a 2-group 2-condition PET study.
The design matrix looks like:
1 0 0 0 1
0 1 0 0 0 1
1 0 0 0 0 0 1
0 1 0 0 0 0 0 1
: : : : : : : : : : : :
0 0 1 0 0 0 0 ... 1
0 0 0 1 0 0 0 ... 0 1
0 0 1 0 0 0 0 ... 0 0 1
0 0 0 1 0 0 0 ... 0 0 0 1
... etc....
So the contrast I think I should be entering is [ 1 1 -1 -1 ].
My understanding of this is an analogy with two-way repeated measures
anova (for a single variable) so I can look at Main Effect of Condition
[ 1 -1 1 -1], of Group [ 1 1 -1 -1] ("invalid contrast"), or interaction
of group*condition [ 1 -1 -1 1].
I can set up a "Compare Populations" design with 2 groups and ignore any
conditions but it seems a bit of a fudge.
How do I look at Group Effect properly? Have I got the wrong end of the
stick? Should I set up the design differently? (I used Full Monty, 2
groups with subjects scans entered with conditions).
Joel
--
-----------------------------------------------
Joel Dunn
Research Assistant
Clinical PET Centre
Guy's, King's and St Thomas' School of Medicine
Lower Ground Floor, Lambeth Wing,
St Thomas' Hospital
Lambeth Palace Road
London SE1 7EH UK
Email: [log in to unmask]
Tel: 020 7188 7445
Fax: 020 7620 0790
-----------------------------------------------
|