Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Alternatives to box plots

From   Nick Cox <>
To   "" <>
Subject   Re: st: Alternatives to box plots
Date   Thu, 21 Mar 2013 16:16:46 +0000

Worth noting that John Tukey in his Exploratory data analysis (Reading, MA, Addison-Wesley, 1977) did include a detailed example making it clear that box plots could no do justice to bimodality.

On 21 Mar 2013, at 13:24, Nick Cox <> wrote:

A long and concurrent thread starting at

centres on t tests and other methods applied to a problem with
obstetrics data. One of various issues raised is the merits of box
plots, including postings by David Hoaglin, Jay Verkuilen and myself.
My evaluation is that our views are much closer than may have

I am taking this out of that thread so that the graphical issues can
be a little more easily discussed by anyone so inclined,

If box plots are oversold -- or at least (vacuously) not a panacea --
then alternatives include

-dotplot- (official)   allows annotation with mean or median

-stripplot- (SSC) allows marginal boxes and bars; more options than -dotplot-

-devnplot- (SSC) a plot with ANOVA-like flavour showing individual
values relative to means (or any other summary of level)

-qplot- (SJ) quantiles!

*   For searches and help try:
*   For searches and help try:

© Copyright 1996–2016 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index