[Feature Request]: ggplot2 4.0 upgrade to fix quantile issue in geom_violin

[R180]

Description

tidyverse/ggplot2#6577 cites a longstanding feature of geom_violin wherein it plots quantile lines based of the density estimate for the violin rather than for the actual data. As a consequence, the violin's median line (for example) won't exactly match the median line in a box plot of the same data. ggplot2 4.0 changes this behavior so that the geom_violin quantiles are computed directly from the data, thus improving the consistency of quantile plotting across geoms.

Purpose

Improved plotting consistency across various geoms

Use-case

Producing a violin plot in JASP's plot builder.

Is your feature request related to a problem?

Not exactly a problem.

Is your feature request related to a JASP module?

Descriptives

Describe the solution you would like

I'm simply asking the JASP team be mindful of the soon-to-be released version 4 of ggplot2, and maybe use the current ggpolt2 development version to do some pre-testing for compatibility with JASP.

Describe alternatives that you have considered

None.

Additional context

Illustration of the current quantile-mismatch between geom_boxplot and geom_violin.

[vandenman]

We will update once ggplot2 releases it's next version, thanks for letting us know! I'm not aware of their release schedule though, so for that I would keep an eye on their GitHub repository. They also have quite a few breaking changes listed in the development branch.

[vandenman]

ggplot2 got released today (see https://cran.r-project.org/web/packages/ggplot2/index.html) so I'll get started on implementing the necessary changes (it's also causing some unit tests to fail).