Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: new schemes for bargraphs


From   "Nick Cox" <[email protected]>
To   <[email protected]>
Subject   st: RE: new schemes for bargraphs
Date   Wed, 23 Jul 2003 17:21:57 +0100

Dimitriy V. Masterov

> I would like to find or make a scheme for bar graphs that contains
> bar fill patterns like polka dots, stripes and other highly
> distinguishable patterns. I put a lot of results on the
> web, and I don't think that most black and white printers can really
> distinguish between the subtle shades of grey that result from
> using many categories.
>
> If a scheme with good B-W patterns exists, I have not been
> able to find
> it. If it does not, is there a way to expand the custom
> colors palette to
> include geometric patterns?
>
> I would be most grateful is someone could point me in the
> right direction.

There was a thread on this a while back initiated
by Edwin Leuven. See the archives starting on 4 April 2003.

In essence, Stata 8 does not, as least as far as I know,
support polka dots, or straight stripes, or wavy stripes,
or indeed anything else other than uniform tones.
Various statements, smoke signals and silences from Stata
Corp don't indicate any interest in implementing any of
these in future.

Stata long since signed up the minimalist school of
statistical graphics, which eschews such effects.
Matters of taste aside, one argument is that
such effects produce vibratory impressions in at least
some readers.

I'll add my usual reference at this point to
Edward Tufte's books:

http://www.edwardtufte.com/tufte/

I see that he has just produced a booklet,
"The cognitive style of PowerPoint", which
remains me about the extremely terse
precis of a sermon once elicited from a
listener:

"What did the minister preach about?"

"Sin."

"What did he say in particular?"

"He was against it."

On Dmitry's (serious) point, I think it is often
true that different colours degraded to black-and-white
are not easily distinguishable. The implication of this is
that other means must be found for distinguishing
different categories, including text labelling,
different symbols on dot charts, or that users
seek to produce graphs in other software.

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/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index