[R] Sensitivity Analysis for Moderated Mediation?

[Bert Gunter]

This post actually has nothing to do with R programming per se, and
hence is off topic here. Please post elsewhere, e.g. on
stats.stackexchange.com or other list that discusses "mediation
models".

Cheers,
Bert
Bert Gunter

"The trouble with having an open mind is that people keep coming along
and sticking things into it."
-- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )


On Mon, Dec 5, 2016 at 5:00 AM, Dominik Wyss <nikwyss at gmail.com> wrote:
> Dear group,
> medsens() is a fantastic method for testing sensitivity of mediation models
> estimated by the mediate(). (mediation package by Tingley et al; Version
> 4.4.5).
> However, I'm wondering whether medsens() is also appropriate for moderated
> mediation? And if not, is there an alternative procedure that allows to test
> sensitivity of moderated mediation analysis?
>
> In one of my study, I find strong theoretical and empirical support for
> moderated mediation (moderator is a dummy var; otherwise the model specs are
> similar to chapter 3.2
> ftp://cran.r-project.org/pub/R/web/packages/mediation/vignettes/mediation.pdf).
> However, my medsens() analysis finds only very poor Rho and R2 values.
> Before throwing away the model, I wonder whether medsens is appropriate for
> estimating sensitivity of moderated mediation models, normally entailing a
> moderation*mediator interaction term.
>
> And if medsens is not appropriate for moderated mediation: Would it be
> enough enlightening to test sensitivity of two unmoderated mediation models
> ran over two sub-datasets, split along the moderation dummy?
>
> many thanks for any advice.
> Dominik
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.