Bayesian Mediation?

BrittJane

Hi team.

Does JASP have a method for conducting a Bayesian mediation yet? It doesn't seem like it to me, but thought I'd check - and if not, is one in the works? Further - does anyone know how to do one with a simple mediation (3 variables + covariates) other than the BayesMed package in R?

Thanks!

EJ

Hi BrittJane,

Yes, it is in the works! We are waiting for some other software developments to pull the trigger on this one. When we do implement mediation, this will likely be based on the blavaan R package, so you could check that out

Cheers,

E.J.

jploenneke

EJ, have your thoughts changed about the method proposed in the paper from 2015 "A default Bayesian hypothesis test for mediation"?

In short

Bayes Linear Regression

X to M

M to Y (partialling out effect of X?)

posterior probability (x to m) x posterior probability (m to y)

then converting back to BayesFactor

Is this still acceptable in your mind?

EJ

My thoughts have not changed but also/mainly because I have not given it much thought. I will revisit this when we implement this in JASP! On our to-do list...

Cheers,

E.J.

jploenneke

EJ....just to be clear

in the second step of the process in that paper

it is m to y...partialing out x?

thanks in advance