Color aes across grouping vars

[rjazwiec]

It is a general idea.

I have dataset from clinical study. Data for two dosing regimens, on two different days. Several parameters from PK analysis to compare.
Trought whole report I want to keep consistent color meaning - orange for lower dose black for higher dose.
I use ggstatsplot::grouped_ggbetweenstats

It is easy to pass color aes when comparing by parameter you want to color:

grouping.var = day_l
color = dose_l

Gives great result (sorry for the rough blinding - restricted data and I don't want to generate dummy plots)

But if I want to test by day but still have consistent color aesthetics (both box in the left pane orange, both in the right black):

grouping.var = dose_l
color = dose_l

It generates error because at the moment of setting colors grouping variable is already dropped.

It left me with:

Is there any way to work around it except generating two separate plots and arranging it together?
Or can you consider modification that will leave grouping variable for both sets after filtering by it?

[rjazwiec]

Ok. I've tried to generate separate plots - but it also doesn't work. Drops all unused columns...

[Ameane]

Hello. I am new to the package and under the supervision of @katossky we are willing to solve this
issue, as part of an assignment at the ENSAI school of statistical engineering
(France). We have read the [code of conduct (https://github.com/IndrajeetPatil/ggstatsplot/blob/main/.github/CODE_OF_CONDUCT.md) and we will try to be as
autonomous as possible but we may ask for guidance from now and then. We
have a solid background in statistics and a reasonable one in computer science.
If any maintainer thinks this issue is too ambitious for a beginner, please tell
us so and - if you have time - guide us to a project / issue were our help may
be more useful.

[katossky]

@Ameane You may want to create fake data and replicate the problem of @rjazwiec on these data. Then try to define what the exeected behaviour exactly is. Then walk back from there to suggest a solution changing the package's code.