Hi Aga,
> >> EV1 Ev2
> >> -1 0
> >> -3 0
> >> -5 0
> >> 0 3
> >> 0 1
> >> 0 5
> >> And then skip -D?
> >
> > Not quite - you need to demean each groups values
> before padding with
> > zeros. And you DO need the -D option.
>
> Let's assume I demeaned the values for each group
> separately and the design
> matrix is correct. What does it really give me to split the
> two subgroups
> into 2 separate EVs? Should this be always done if the
> means of these
> subgroups for these EV differ?
You're answering a different question: if you pool both groups in a single demeaned ev, you're looking for a significant linear correlation of this beh ev with the FA values across both groups.
But if you split them in two separate evs, padded with 0, then you're looking where beh is significantly correlating with FA *differently* between your two groups (i.e. the slopes for group1 beh and FA is significantly different than between group2 beh and FA).
Hope it makes sense,
Gwenaelle
|