Dominique Makowski

Related to that (looking at Mattan's link), a long time ago I wanted to add [report support](https://github.com/easystats/report/issues/31) for `lavaan::modificationIndices()`, but a nice first step could be to support it by...

> you have thought about implementing an Olkin-Pratt estimator Besides the validity questions, do we have this estimator and, if not, is it straightforward to add?

> Or perhaps it would be cleaner to just have the one metrics argument and allow it to be either the current string vector or a named list of functions...

Indeed, we should probably add a step to get_data to sanitize the output right?

Actually modelbased computes something named "average marginal effects": ``` r model Average Marginal Effects #> #> Start | End | Petal.Length | Coefficient | SE | 95% CI | t(142.33)...

Many potential evolutions could be considered to avoid too many duplication and to pool maintaining efforts - sharing some common core low-level set of functions (via insight?) - modelbased swaping...

The main issue is fixed using your suggestion Mattan, even though I'm not sure whether it's the best solution to modify the predictions rather than, somehow, the levels. Because, for...

> What do you think of the viewport overlay I linked to? What do you mean?

Yes this is my goal, it is the same as above in my example; i.e., it uses `ggplot2::geom_dotplot`. The problem of that geom is that it doesn't have afaik good...

Awesome, thanks a lot for your input! ☺️ Let's go with ggdist's solution then