problem... As a matter of fact, it pays to browse the help files of
well-known commands such as -ta- occasionally. Browsing -h tabulate oneway-
recently, I noticed you can have the command sort according to the
frequencies displayed (but I failed to notice the -plot- option). Thanks for
pointing it out, Nick.
Martin Weiss
_________________________________________________________________
Diplom-Kaufmann Martin Weiss
Mohlstrasse 36
Room 415
72074 Tuebingen
Germany
Fon: 0049-7071-2978184
Home: http://www.wiwi.uni-tuebingen.de/cms/index.php?id=1130
Publications: http://www.wiwi.uni-tuebingen.de/cms/index.php?id=1131
SSRN: http://papers.ssrn.com/sol3/cf_dev/AbsByAuth.cfm?per_id=669945
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Nick Cox
Sent: Friday, June 27, 2008 3:02 PM
To: [email protected]
Subject: RE: st: Quick display of categorical data
Martin's answer focuses on tabular displays. Here are seven ways to get
a graph. There will be others.
Official Stata:
tabulate rep78, plot
histogram rep78, discrete
dotplot rep78
quantile rep78
The last is best for ordinal variables.
User-written commands:
stripplot rep78, stack
catplot hbar rep78
qplot rep78
-stripplot- and -catplot- are on SSC. -qplot- is on the SJ website.
-qplot- is best for ordinal variables.
See also a discussion:
SJ-4-2 gr0004 . Speaking Stata: Graphing categorical and
compositional data
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . N.
J. Cox
Q2/04 SJ 4(2):190--215 (no
commands)
discusses graphical possibilities for categorical and
compositional data
That is the public domain via
<http://www.stata-journal.com/sjpdf.html?articlenum=gr0004>
Nick
[email protected]
Martin Weiss
Try this:
sysuse auto, clear
foreach var of varlist rep78 for{
ta `var'
inspect `var'
}
Admittedly quick and dirty...
Zitat von "Michael I. Lichter" <[email protected]>:
> I'm beginning to analyze
> some survey data, and much of the data consists of responses to
> questions on a 1-5 scale, a few responses on a 1-3 scale, and a few
> responses to limited nominal scales. I have too many cases (1,000+) to
> learn anything from a stemplot (although I suppose I could sample 1/10
> of the cases for this purpose), which isn't meant for categorical data
> anyway, and it's way, way too cumbersome to individually create and
> print bar charts using Stata graphics (unless there's something I
don't
> know about how to do this, which is likely). Ideally, what I'd like is
> a display something like this:
>
> Some inter |
> esting var |
> iable | Freq. Percent Cum.
> ------------+----------------------------------- Never, ever |
> 315 30.79 30.79 30 X X
> Maybe once | 204 19.94 50.73 X
X
> Occasionlly | 149 14.57 65.30 20 X X
X
> Pretty oftn | 55 5.38 70.67 X X X
X
> Always! | 300 29.33 100.00 10 X X X
X
> ------------+----------------------------------- X X X X
X
> Total | 1,023 100.00 1 2 3 4
5
>
> I would happily settle for just the ASCII bar chart, and a horizontal
> bar would be fine. The point is to be able to get a sense of the data
> at a glance, without wasting a bunch of paper.
*
* 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/
*
* 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/
