Hi,
Sorry, you are right. Those contrasts do represent the interactions.
cond 1 A 1 1 1
cond 1 A 1 1 1
.. .
cond 2 A 1 -1 -1
cond 2 A 1 -1 -1
..
cond 1 B -1 1 -1
cond 1 B -1 1 -1
...
cond 2 B -1 -1 1
cond 2 B -1 -1 1
X Y Z
Z would be [(A1-A2)-(B1-B2)] (interactions, one sided?).
[(B1-B2)-(A1-A2)] would be (-Z)?
Roland Marcus Rutschmann wrote:
> now I am confused because I think Karli is right.
>
> I hope I read this correctly as 2 Factors (let's call them F1, F2) with 2
> levels each (F1(1) F1(2) F2(1) F2(2)) and you build contrasts
> [F1(1)-F1(2)]-[F2(1)-F2(2)]?
>
> Why don't you take the F contrast over that contrast instead of 2 T-contrasts
> (the "direction" of an interaction term is pretty hard to interpret, isn't
> it?). But other than that I don't see the incorrectness.
>
> This is interaction: If the difference between cond1 and cond2 is different in
> condA then condB, the maineffect(1-2) depends on the other effect (A-B). This
> is exactly the 3rd coloumn in fig 7 of Rick Hanson's excellent technical
> guide to Anova. http://www.fil.ion.ucl.ac.uk/~wpenny/publications/rik_anova.pdf
What I'm not sure about is: if you're interested in the interactions,
can you use a design matrix just containing those contrasts? I guess you
need to make a model containing at least the two main effects and an
interaction (column 1, 2 and 3 of fig 7). But then you should probably
use an F-test, not a T-test.
Hope somebody can give a more definitive answer than this one...
Best wishes
Alle Meije
|