Hi All
I'm trying to prepare contrasts for a 3x3x2 anova, i know!
anyway,I have entered the 1st level contrast to the second level according
to rik's technical reports Factor A(2levels) B(3 levels) c(3 levels)
with A rotating slowest and c rotating fastest.
so i have tried to set up contrasts for a 3x3 first and while the contrasts
for the main effect of factor B and C work, the interaction does not. I then
used the formula from rik's paper to try to work out what was going on but
ran into trouble, probably due to my matlab skills)
so simply for a 3x3
a = [1 1 1]'
a =
1
1
1
b= diff(eye(3)')
ans =
-1 1 0
0 -1 1
c = orth(diff(eye(3)')) (from rik's paper, i hope that was correct)
ans =
-0.7071 0.7071
0.7071 0.7071
not [082 -0.41 -0.41; 0 0.71 -0.71]' as in rik's paper
obviously then the kronecker product resulting from their multiplication
isn't correct
for reference i used the following command k=kron(a,c)
i know i must have gone wrong somewhere, I'd really really appreciate any
help with this.
any advice on moving to a 2x3x3
cheers
O
|