We are having a bit of trouble (actually lots) with defining F contrasts
using the Full Monty--hope someone can offer thoughts here or back channel
some life saving advise.
Our model has three groups (A, B, C) and each group completed one task
having three levels (1, 2, 3). So, in the design matrix col 1 thru 3 are
group A level 1, level 2, level 3; group B col 4-6, level 1, 2, 3; group C
col 7-9, level 1, 2, 3.
We'd like to contrast group A with group B to localize regional activation
that differed across levels 1, 2, 3 (essentially determine where the slope
across levels 1-3 is greater for group A relative to B). We'd also like to
do the same analysis comparing group A with group C.
Any comments/suggestions would be really appreciated.
Mike Schlund, Ph.D.
Johns Hopkins University School of Medicine,
University of Pittsburgh & University of North Texas