I am working on the review of a manuscript. The reviewer wants effect sizes reported. We have a linear model with an interaction term and we're reporting marginal means for smokers vs non-smokers on another 3 categorical variable.
I decided to use Cohen's d, but the problem is that I am not sure how to calculate it for the interaction term. I have read through the similar questions posted, but I they do not explain how to calculate the pooled standard deviation estimate.
