Answers are grounded in this paper and cite their sources.
A variant of the boxplot is proposed in which the sides contain the information of a percentile plot (which is equivalent to the empirical cumulative distribution function). Unlike boxplots, there is no question about how long to draw the whiskers, nor is there loss of information due to grouping. Side-by-side comparisons of distributions are especially effective. In spite of including more detail, the impact on statistically-untrained readers remains similar to that of traditional boxplots.
No takes yet. Share an insight, caveat, or question.
Esty et al. (2003) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: