Why do some journal editors request that we present the main effects when the interaction is also significant? Is "for sake of completeness" a good enough reason?

More Genevieve Allaire-Duquette's questions See All
Similar questions and discussions