Re: st: Alternatives to box plots

[Nick Cox <njcoxstata@gmail.com>]

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 <njcoxstata@gmail.com> wrote:

> A long and concurrent thread starting at
>
> http://www.stata.com/statalist/archive/2013-03/msg00838.html
>
> 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
> appeared.
>
> 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!
>
> Nick
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/