Hi,
In my study there are 2 genetic groups and I want to compare separately the genetic effect gender-wise.
e.g. construct a contrast comparing only brain areas with larger effect in GeneticGroup1 than GeneticGroup2 for only females.
assuming my design (flexible factorial, A -genetic groups, B-gender) my design is A1 A2 B1 B2 A1B1 A1B2 A2B1 A2B2
so in order to see larger effect in genetic group 1 than group 2 (A1>A2), I define the following T contrast:
1 -1 0 0 0.5 0.5 -0.5 -0.5
due to interaction A1 > A2 only for the females and A1 <= A2 for the males, so how do I define something similar for only the females (i.e. A1 > A2, only for females - B2)
On the same page, when searching for interaction I can define two T contrasts:
0 0 0 0 -1 -1 1 1
or
0 0 0 0 1 1 -1 -1
these two yield different (complimentary?) results, and I fail to understand whether this means the "direction" of the interaction effect and what does this actually mean?
Thanks
|