Hi Gwenaelle,
"then you should indeed split the gender covariate as you did, but then *separately* demean EV3 (so gender in group 1 *only*) and EV4 (gender in group 2 *only*)."
Thanks. I thought as much and did just that.
My problem now is in the contrasts, if I want to regress out gender interactions do I make contrasts of EV3 and EV4 or do the EV1 and EV2 contrasts take into account the covariates in EV3 and 4? The manual is a little fuzzy about this.
Thanks
Suman
|