Mattan S. Ben-Shachar

Oh, an easy paper! (🤬) I'll take a closer look in a couple of days (:

I have no idea. I guess more generally I'm asking if the statement in this message is true? https://github.com/easystats/bayestestR/blob/master/R/mediation.R#L184

```R library(brms) f predictors, e.g. with `effectsize::standardize()`. #> # Causal Mediation Analysis for Stan Model #> #> Treatment: hp #> Mediator : am #> Response : mpg #> #> Effect...

> to simulate values from the fitted models for treatment = 0 and treatment = 1, then compute indices based on that. How is that different than using a slope...

Ah. Fuck. But I guess that does make sense, right (assuming you chose reasonable values), due to the non-linear nature...

@bwiernik I need a sketchpad... I would have built that in ggplot for no reason...

So where does this stand? Should the function change in some way? How should the docs be updated?

@strengejacke ?

> I think that what easystats users expect here, rather some lengthy theoretical rant, is a practical perspective. You know me too well... But I must rant at least a...

\ I agree that results are often similar numerically, but I also think we shouldn't only strive to make analyses easy, but also push for the correct interpretation. I sometimes...