[S] brief replies to the pie-chart critics

Dave Krantz (dhk@paradox.psych.columbia.edu)
Fri, 13 Nov 1998 12:22:47 -0500 (EST)

Frank Harrell, Nov 13 1998:
Subject: [S] Pie Charts and Type III Sums of Squares

> Regarding whether pie charts should have been included in S-PLUS,
> my feeling is that they are worth the same as Type III sums of squares.
> Maybe Type III sums of squares should always be depicted using pie charts.

I like Type III sums of squares much more than I like pie charts!
Every tool can be misused; Why not try to avoid the misuses,
rather than throwing out the tools?

Todd Taylor, Nov 13 1998
Subject: Re: [S] Pie diagrams anyone?

> I can't think of a case where some other display wouldn't work better.

Robert Garrett's example is pretty compelling, and I imagine others
can be found.

> Their major weaknesses from my point of view
[ list deleted to save space ]

I agree with most of what Todd says here, but would put much of
it less emphatically.

> Their major advantage, of course, is that managers and most
> clients think they like them.

I think I like coffee ice cream, but maybe I really don't.

Seriously, this is something that deserves attention and study.
In some cases, one can do as Eric Gibson recommends, gently
offering an alternative that one thinks is better; and sometimes,
the client will agree. Where the client does NOT agree that the
alternative is better, it MAY be an opportunity to learn something
about the mental tasks that the client and his or her audience
are engaging in, vis-a-vis the data.

Jim Hodge, Nov 13 1998:
Subject: Re: [S] pie charts

> Dave Krantz wrote
> > it is still true that
> > many clients really like pie charts, and it behooves us to
> > think hard about why this is so;
> A propos of this, a story from grad school days: When I was working in
> the stat consulting clinic, I had more than one person come into my office
> saying that they'd been using SPSS because they were in the social
> sciences, and "it's the statistical package for the social sciences, isn't
> it?"
> Why do *you* figure people like tilted 3D pie charts and 3D bar charts so
> much, when they're manifestly perceptual disasters?

I don't know; this seems like something to think HARD about.
Answers that come off the top of the head, and ones that dismiss
most managers and clients as "stupid", seem suspect to me.

Dave Krantz
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