YHey! Thanks for answering. No, I meant moderation analysis. So as I found the onlycondition is there must be a relationship between the dependent and independent variable and the ones for regression analysis of course. However, I feel al little bit lost at this point. That is why I want to ask and see it as a list from a source and also find out the steps for controlling them.
No real rules. Effect modification is the same as interaction, which can be on the additive or multiplicative scale. Example, risk for lung cancer in persons exposed to cigarettes and asbestos potentially being bigger than individual effects. Effects can be synergistic or antagonistic, I guess there could be some unfaithfulness as well (conflicting effects).
Usually address by controlling for both variables or stratifying.