Re: st: -tabplot- updated on SSC

[Nick Cox <njcoxstata@gmail.com>]

Thanks very much for the compliments.

The UCLA page you cite was the result of my Stata friends there
sampling some of my work but was not ever complete, nor I guess
intended to be.

A list of my Stata packages in the public domain is given in the help
for -njc_stuff- (SSC), which is fairly up-to-date, but also includes
some obsolete or superseded material. A better guide is the
-njc_best_stuff- help file in the same package.

My FAQs and articles in the STB and SJ can be found as usual by

. search cox, author [historical]

Nick

On Sun, Oct 21, 2012 at 5:19 PM, Ariel Linden, DrPH
<ariel.linden@gmail.com> wrote:
> Nick's genius is unsurpassed!
>
> As far as I am aware, the following site is the only place where one can
> find the entire collection of Nick Cox's artwork.
>
> http://www.ats.ucla.edu/stat/stata/faq/graph/njcplot.htm
>
> Nick, is this site updated with all your graphs? Is there another location
> to find them as well?
>
> Ariel
>
>
> Date: Sat, 20 Oct 2012 09:45:29 +0100
> From: Nick Cox <njcoxstata@gmail.com>
> Subject: st: -tabplot- updated on SSC
>
> Thanks as usual to Kit Baum, the -tabplot- package on SSC has been updated.
>
> - -tabplot- works in Stata 8 up except that as advertised in
>
> <http://hsphsun3.harvard.edu/cgi-bin/lwgate/STATALIST/archives/statalist.120
> 8/Author/article-918.html>
>
> any users of Stata 8 or 9 downloading it will need to rename the
> tabplot.sthlp file to tabplot.hlp for it to be readable by their
> Stata.
>
> - -tabplot- offers an alternative to stacking bars or arranging them
> side by side; it arranges them in table form, in what on occasion I
> have called a table plot, or what can be called a one-way, two-way or
> three-way bar chart.
>
> By default -tabplot- calculates and shows cell frequencies (percents
> etc.) for one, two or three categorical variables, but with a little
> effort it can be subverted to show any cell summary for such a
> cross-classification. There are several small virtues to such a
> format. One is that values can be shown alongside bars, thereby
> hybridising graph and table. Another is that very small frequencies
> (more generally values) are more evident to inspection. (In a stacked
> bar chart, for example, such values are easy to miss as thin slivers
> of a different colour.)
>
> The update is twofold. First, I continue to add references to
> occurrences of this kind of plot and remain very open to new
> references, especially old or interesting ones. Although when you see
> such a plot, the idea will seem very simple and obvious to you, they
> remain far less common than (e.g.) stacked bar charts.
>
> Second, Marcello Pagano kindly alerted me to a small but careless bug,
> which I have fixed. (If it had bit you, you would have noticed.)
>
> Incidentally, some readers might be aware of -tableplot- (SSC) and
> wonder about the difference. -tableplot- is not being maintained by
> me; the intention is that what you can do with -tableplot- you can do
> with -tabplot-, although with different syntax. It's best to switch
> and learn the difference, as -tableplot- is going nowhere new.
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/