I've learned moderated mediation analysis from this book chapter. However, I'm looking for a method for "mediated moderation" and I cannot find it. For example, we are trying to reduce the correlation between X and Y. We already see our treatment T has significantly reduced the correlation, but we need to figure out whether this effect was mediated by variable M. How can we do such an analysis in R?

  • $\begingroup$ A path diagram would help convey the hypothesis you want to test. Mediated moderation could mean different things, and is often the same model as moderated mediation (but the focus is on the significance of the interaction term's indirect effect, rather than on the degree to which the indirect effect varies across the moderator). But you might mean what these authors call "indirect moderation": arxiv.org/pdf/1701.08862.pdf $\endgroup$
    – Terrence
    Commented Dec 14, 2022 at 11:28
  • $\begingroup$ Thank you for your response. In our use-case, there exists a correlation between X and Y. Our treatment has significantly reduced this correlation. We are interested is the mediation path through which our treatment reduced the correlation. 1) Would you call this mediated moderation? 2) How would you model and analyze this mediation effect? $\endgroup$ Commented Dec 17, 2022 at 23:18


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.