Hi there,
I have a bit of a problem with a design that includes 3 groups. I also have a covariate (split across columns) and a confound (single column). Rather than testing individually for each pair of groups whether there is an interaction (directed) with my covariate using ttests, I wanted to simply see whether there is an interaction across groups if that makes sense (undirected). I am not quite sure how to do this, I guess I simply do an F test across some contrasts (see picture attached). Would that give me this?
Thanks very much for your input.
Cheers,
Torsten
|