Re: st: Tables and row/column percentages

[Rufus Browning <browning@sfsu.edu>]

Luciana,

You're right about the inefficiency of dozens of separate tabulations. Two 
suggestions: use -table- to generate means that are proportions, and use 
-svytabs-.

USE -table- TO GENERATE MEANS=PROPORTIONS

If your "dependent" variable is 0,1 or can sensibly be made 0,1 another 
possibility is to use the -table- command to generate "means" (i.e., 
proportions) of that variable within categories of row, column 
(supercolumn, superrow) variables. -table- takes pweights, e.g.

. table district villtype [pw=weightvar] , contents(mean marriage) row col 
format(%5.3f)

Then I like to see the corresponding table of frequencies:

. table district villtype if marriage != . , c(freq) row col

USE SVYTABS TO COMBINE MANY CALLS TO -svytab-

I've also found -svytabs- by Michael Blasnik to be helpful for some kinds 
of problems:
package svytabs from http://fmwww.bc.edu/RePEc/bocode/s
------------------------------------------------------------------------------------------------------------------------
TITLE
      'SVYTABS': module to Create Tables for export Combining Multiple 
Survey Tabulations

DESCRIPTION/AUTHOR(S)
      svytabs generates a variety of tables using calls to Stata's
      svytab command.   Its advantages over svytab are in convenience
      and formatting.  A single  svytabs command can generate a table
      that would require many svytab commands  and tedious assembly of
      results.  By default, tables of percentages are  created in a
      comma delimited format that can be saved to a text file for
      importing to other documents.  Statistically significant
      differences can  be flagged with asterisks.  Standard errors,
      confidence intervals, design effects, and population counts can
      also be produced. Detailed output can  be logged while tables are
      saved in separate files.

Hope this helps.

Rufus Browning
browning@sfsu.edu

At 11:56 AM 10/28/2003 -0500, LUCIANA SURAN wrote:

> Dear Statalist,
>
> I have a question concerning the "table" command, and actually, the usage of
> tables in general in STATA.  I am workign with a dataset in which I would
> like to examine variables such as marriage by a number of different
> indicator variables, such as district, village type (i.e. intervention vs.
> control), age, etc.   Also, I need to examine the data using a pweight.
>
> I was wondering if there is anyway to get row/column percentages using the
> "table"" command.  I know that I could get these percentages using the "tab"
> command, but I find it inefficient to run lots of tabs and have the output
> be scattered all over the place.
>
> In sum, I was wondering if there is anyway to get row/column percentages for
> a number of different variables in one step.
>
> I would appreciate your help very much.
>
> ~Luciana

*
*   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/