I didn't add a comment that I had in mind on the use of
a upward-pointing triangle, motivated
by the idea of a fulcrum about which a
beam supporting a histogram would balance.
Many introductory texts and courses
stress the mean as a balance point (centre
of gravity) for a distribution. Those
with some physics background will recall
that there is a very good reason for calling the
mean a moment measure.
Others may be reminded of the principle of a
see-saw
teeter totter
bascule
Wippe
or whatever you call it in your language.
I am indebted to Vince Wiggins for the
transatlantic translation (see-saw = teeter
totter) and also for implementing arrows.
In the statistical literature, the main idea
was discussed by
Doane, D.P. and Tracy, R.L. 2000.
Using beam and fulcrum displays to explore
data. The American Statistician 54: 289--90.
An earlier reference is
Helsel, D.R. and Hirsch, R.M. 1992.
Statistical methods in water resources.
Amsterdam: Elsevier. p.5
The whole of that textbook (which has
much worthwhile material on graphics) is
available free as a series of .pdf documents from
http://water.usgs.gov/pubs/twri/twri4a3/html/pdf_new.html
A more versatile -beamplot- program will be released via SSC
next month. For the help file, I would be grateful for
any further literature references to the use of means as
fulcrums (fulcra?) in descriptive or exploratory data graphics.
(Mentions of the mean as a theoretical balance point I
trust to be too frequent to be worth citation.)
Nick
[email protected]
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
