Centering IVs for checking interaction effects in regression analysis

Attention: I have just skimmed an article (Echambadi, R. and Hess, J.d. (2007) Mean-Centering Does Not Alleviate Collinearity Problems in Moderated Multiple Regression Models, Marketing Science 26 (3), 438-445) which clearly states that the centering of the IVs does not impact the results of a regression even though collinearity issues are not indicated by the VIF anymore. Thus, centering does not have any pos. effect on the results of a regression and my question seems unwarranted.
Can anyone tell me instead how to check for an interaction term if I cannot test for it with by multiplying two continuous IVs? Thank you!

Dear board,

I'm currently conducting a research for my thesis und struggle a little bit with the statistics. Right now I want to check whether there is a sign. interaction effect between two IVs. I have read that in order to avoid multicollinearity issues, I could center both IVs by subtracting their mean, and then I could multiply both IVs to create an interaction variable. Then I would enter all three variables in a regression to see whether there are sign. effects.
Is this course of action legitimate? If yes, is there a good way to justify this course of action?
Thank you very much for your support!

Kind regards,

Last edited: