Dear Alfonso,

I'm very grateful for your comment.
May I still ask for clarity how to test for a "group by sex" interaction (only between-subject contrasts)?

When having the following strings:
- "cases": coded as 1 and controls as 0
-"controls": coded as 1 and cases as 0
-"males": coded as 1 and females as 0
-"females": coded as 1 and males as 0

will the contrast [1 -1 -1 1] test for such interaction, and [0.5 -05 0.5 -0.5] for the main effect of sex?

Thank you.
Lucas