[S] S: Pie Charts

John Maindonald (john.maindonald@anu.edu.au)
Mon, 16 Nov 1998 08:52:36 +1100 (EST)


I think we need to distinguish between an interest in grabbing
attention, and the accurate presentation of information. It seems to
me legitimate, in the right time and place, to do a bit of attention
grabbing. Strict purity in this regard might mean that we would never
do work for an advertising agency, or even for a promotional brochure.
Colourful pie charts can be great attention grabbers. I guess that is
their appeal for seminar presentations. One hopes that in scientific
papers the interest is more in accurate presentation of information.
Scrutiny of the obfuscating use of *'s and p-values etc. in much of
the stuff that gets into print is not too encouraging in this respect.

One attraction of pie charts, for the perceptually sensitive, is that
(1) there are no endpoints as in the divided bar; no categories are
priveleged.
(2) unlike a dotplot form of representation; it is immediately clear
how the total (100%) is formed.

Does anyone know of experimentation with dividing up a thickish
circumference? One is then using a line to represent a 1-dimensional
quantity. Do users decode this form of representation more accurately?

John Maindonald email : john.maindonald@anu.edu.au
Statistical Consulting Unit, phone : (6249)3998
c/o CMA, SMS, fax : (6249)5549
John Dedman Mathematical Sciences Building
Australian National University
Canberra ACT 0200
Australia
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news