Describe the solution you'd like
A clear and concise description of what you want to happen.
It's great that ggbetweenstats includes an effect size, but practically speaking, the effect size for K-W/ANOVA is often not very useful insofar as it's hard to interpret. It says that there's a difference between at least one group and the other groups. An effect size for the pairwise comparisons is much more interpretable since it's a single group comparison to a single group. For K-W, I'm using Dunn's test for the pairwise and then Vargha and Delaney's A for the effect size. I think it would be a good feature to include as an option because it means basically all of the important information about the chart is in the graphic itself.
How could we do it?
A description of actual ways of implementing a feature.
S. Magnifiaco has put together a straightforward implementation for Vargha & Delaney's A as multiVDA in rcompanion. I'm using that to generate the figures and then putting it into the caption together with paste and a few variables. My current implementation is very kludgey but should give you a sense of what I mean.
Describe the solution you'd like A clear and concise description of what you want to happen. It's great that ggbetweenstats includes an effect size, but practically speaking, the effect size for K-W/ANOVA is often not very useful insofar as it's hard to interpret. It says that there's a difference between at least one group and the other groups. An effect size for the pairwise comparisons is much more interpretable since it's a single group comparison to a single group. For K-W, I'm using Dunn's test for the pairwise and then Vargha and Delaney's A for the effect size. I think it would be a good feature to include as an option because it means basically all of the important information about the chart is in the graphic itself.
How could we do it? A description of actual ways of implementing a feature. S. Magnifiaco has put together a straightforward implementation for Vargha & Delaney's A as multiVDA in rcompanion. I'm using that to generate the figures and then putting it into the caption together with paste and a few variables. My current implementation is very kludgey but should give you a sense of what I mean.