Thanks for the compliment. I haven't rummaged around to try to find the
precise recipe behind the graph in question, but it looks similar in
content to a strip plot with added mean and median. I'd like to see the
justification for the design. I'd rather show the data a little more
directly.
With -stripplot- from SSC, you can go
sysuse auto
egen mean = mean(mpg), by(foreign)
egen median = median(mpg), by(foreign)
stripplot mpg, over(foreign) vertical stack center height(0.2) ///
addplot(scatter mean median foreign, ms(Sh Dh) msize(*2 ..)) ///
legend(on order(2 "mean" 3 "median") pos(3) col(1))
Nick
[email protected]
Fred Wolfe
Yesterday, Andrew Gelman placed this on his blog:
"I'd love if someone else were to write my article, tentatively titled
"Better than a boxplot," with the following abstract: "We demonstrate
graphical options that dominate the boxplot. We hope that, once these
alternatives are understood, boxplots are never used again." But I
have a horrible feeling I'm going to have to write this article
itself."
One response was this:
Jesse | October 9, 2009 10:10 PM | Reply
Here is the first image that comes up when you google it:
http://www.mathworks.com/matlabcentral/fx_files/23661/3/distributionPlot
.png
The one on the bottom left is pretty great.
Great, indeed. Perhaps Nick Cox, the great Stata GraphMaker, or
someone else could make a Stata program to do this?
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
