Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: AW: Presenting categorical data


From   "Martin Weiss" <martin.weiss1@gmx.de>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: AW: Presenting categorical data
Date   Wed, 1 Apr 2009 10:48:11 +0200

<> 

Tim may want to try Nick`s

*************
ssc d tabplot
*************

whose help file says " It is mainly intended for representing a two-way
contingency table of categorical variables."

HTH
Martin


-----Ursprüngliche Nachricht-----
Von: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag von Tim
Gesendet: Mittwoch, 1. April 2009 10:35
An: statalist@hsphsun2.harvard.edu
Betreff: st: Presenting categorical data

Dear Statalist

I have a small dataset (n=51 with many missing values). There are about 
17 variables, all categorical with 6 to 10 categories in each (although 
some categories can be sensibly combined). I also have a list of 24 
comparisons that the researcher would like to look at.

It seems to me the only viable approach is to simply display the data in 
a way that highlights interesting features. True?

So I would like to know what commands I should be investigating, whether 
built-in or user-written.


As an example, here is one tabulation from the dataset:


Use policy |         Knowledge of PEG
     guide |      Poor   Adequate  Above ave |     Total
-----------+---------------------------------+----------
         1 |         0          6          2 |         8
         2 |         6          2          1 |         9
         3 |         0          3          1 |         4
         4 |         0          7          1 |         8
         5 |         0          7          3 |        10
-----------+---------------------------------+----------
     Total |         6         25          8 |        39

Knowledge was actually a 5-point scale from None to Excellent.
Use policy guide is 3 categories of No (don't need to; no guide exists; 
guide not suitable) and 2 categories of Yes.
I think the interesting result is that all the people with poor 
knowledge claimed that no guide exists.
The table can show this (and of course will be better with proper 
labelling), but I am sure there are better ways to present this.

-scatter- with -jitter- gives me a graph that I think is a good starting 
point. But then adding categorical labels to the axes becomes a hassle. 
I expect -scatter- is not the right tool -- what more appropriate 
commands exist?

Or is my whole approach to this dataset wrong?

Tim

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index