faceting in dabestR

[Andy A.]

in my drosophila genetics research lab my students and I generate a large number of control-test datasets that we plot as scatterplots to show differences in the means between the two groups. we generate extremely long tidy data tables, with each observation associated with several factors that each have two or more levels (control vs test, male vs female, multiple genotypes).  we currently manage the data visualization with extensive faceting in R, so we can create these long tables and let R break out the different case/control dyads in facets. I'd love to use estimation stats to do this but would prefer not to assign group numbers to every permutation of the existing factors - the beauty of faceting is how flexible and efficient it is at generating pairs of groups for analysis. I'm curious if anyone has a way to apply estimation stats to data like this - each panel is an unpaired control-test dyad.

Many thanks!

Andy Arsham

Bemidji State University

[Jonathan Anns]

Hi Andy, apologies for the slow response. 

We agree this would be a cool feature to include in our packages, thanks for the suggestion! One could recreate similar styles in Python with a multi-subplot figure and a for loop, but this requires more manual coding as you mention. We have noted it down as a future addition. However, we are a small team with a variety of fixes and additions currently being worked on, so we cannot commit to adding this feature in the near future.  

~ Estimationstats team