5
$\begingroup$

I'm quite new to stats so this might be a problem of perspective on my part.

Let's say I have the following:

  • IV: How my friend group feels about helping people
  • DV: Whether or not I help someone

It has already been shown that IV interacts with DV.

Then someone else shows that $M$ moderates the IV-DV interaction:

  • $M$: How much I care about my friends' opinions

I am wondering if $N$ could moderate $M$:

  • $N$: How much I see a very simple thing I could do to help someone

Conceptually, $N$ would moderate the strength of the moderation $M$ on the IV-DV relationship. If the task is so small, it could change how much I care about my friends' opinions.

I find nothing on the web and nothing on this site about this (moderating a moderator).

I see that often (though not necessarily) $M$ will have an interaction on the DV. If that is the case, it might be more accurate to see that $N$ is just a moderator and it is moderating the $M$-DV relationship and $M$ is seen as the IV. But I'm unclear.

$\endgroup$

1 Answer 1

7
$\begingroup$

Moderated moderation is a well-established model in path analysis / structural equation modeling, which as an example from the link takes on this relationship:

enter image description here

Given the complexity of estimating the inferential estimates with such models, one usually needs something like PROCESS or structural equation model to estimate them.

$\endgroup$
0

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.