I am conducting a regression analysis to look at the relationship between a behavioural score and brain volume. My sample is made of 2 patient groups: A and B. The model includes the following variables:
DV = grey matter volume IV = behavioural score Covariates = age, gender and total inter cranial volume
I am interested in the main effect btw grey matter volume and behavioural score regardless of whether someone belongs to group A or group B. However, I do know that the covariate "group" is a moderating variable. If I understand the effect of moderating variables correctly then including the covariate "group" without its' interaction with the behavioural score will mask the main effect. And this is were my confusing begins:
a. am I right with the above assumption? b. if yes, do I remove the covariate "group" from the model? c. if no, how do I add it to the model?