Why you shouldn't rely on box plots too much

[@rappa753]

If you enjoyed this video and want to level up your R skills even further, check out my latest video courses: 📍Data Cleaning Master Class at data-cleaning.albert-rapp.de/ 📍Insightful Data Visualizations for "Uncreative" R Users at arapp.thinkific.com/courses/insightful-data-visualizations-for-uncreative-r-users

[@udaykumarravada8820]

Very informative!! I like rain clound plots along descriptive stats and sometimes I use also adjusted box which is much better

[@blaisepascal3905]

Really clear, nice video with lot of efforts!

[@rappa753]

Thank you 🤗

[@iwoaugustynski9265]

I like beeswarm plots for that purpose

[@ChongFrisbee]

My expectiation for the wiskers of a boxplot is to show 1.5 times the interquartile range. Meaning the distribution of the first and last quartiles also don't really affect the plot (and the wiskers are guaranteed to be symetrical). I don't really have a problem with boxplots, but I agree that trying to understand a distribution of data looking only at a boxplot is silly. Violin plots, on the other hand, I actually think are actively harmful. This is, of course, a personal opinion. If people find them useful then they find it useful. But the alternatives where you mixed more visualizations instead of mirroring the density plot where really good!

[@rappa753]

In what sense do you mean that violins are actively harmful? 🤔

[@ChongFrisbee]

@@rappa753 I possibly worded that too strongly, I don't think they are harmful in general, but if I, personally, need to understand a distribution it is more difficult if I see a violin instead of a regular density plot. By mirroring the density plot, the violin exacerbates the density around the mode and other density local maximums. I have to actively undo the bias this effect generates in my mind

[@rappa753]

@@ChongFrisbee that's a good point. Haven't thought about this.