Fwd: Re: [S] Pie diagrams anyone?

Edward C Malthouse (ecm@nwu.edu)
Sun, 15 Nov 1998 14:46:10 -0600


I tell my students that the only good reason to use a pie chart is if a
client or boss specifically asks for one. In presenting results from my
work, I've never found a situation where pie charts gave a better
presentation. With that said, I'd like to make some confessions:

I have an old-fashioned wristwatch with three arms. This is a visual
display of time akin to a pie chart. There is no necessity to encode time
using these angles/vectors. The display adds no information over a digital
representation of the time. I even have difficulty determining the time
from a clock or watch with arms but no numeric labels. I suppose that many
years ago there were mechanical advantages to having arms, but I'm fairly
certain that this is no longer the case. My only reason for selecting a
watch of this kind is that I find the display more attractive than those of
digital watches.

The speedometer in my car also has a needle rather than a digital display.
Again, speed is being encoded with an angle. Even with the numeric labels
along the arc, I'm sure my ability to decode speed is much less accurate
than my ability to read a digital. Yet, I still prefer the needle-arc
variety.

These are two instances where I have some simple information to present --
a single number without making any sorts of comparisons -- and accuracy is
not a big concern. If I were performing a physics experiment involving
speed or time where accuracy is important, I'd be better off with a digital
display. In these situations, perhaps a pie-like representation is more
attractive visually than the number itself or a bar/dot format (although I
still prefer a bar thermometer over a dial one).

Ed Malthouse

Edward C. Malthouse, Ph.D.
Assistant Professor
Integrated Marketing Communications
Medill School of Journalism
Northwestern Universtiy
Evanson, IL 60208-1290
Tele: 847-467-3376
Fax: 847-491-5925
-----------------------------------------------------------------------
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