Carsten Baumann
>
> I am trying to summarize results from a survey with about 28,000
> participants. What I am trying to do seems simple, however, I need help.
> A former colleague of mine used to do similar work in SPSS using the
> AGGREGATE command and then OUTFILE with a BREAK by (sub)categories. Any
> tips on how to do this are greatly appreciated.
>
> I am working with student survey results with the following variables:
>
> WTSKUL (value 0 to 2)
> BUILDING (value 20 to 8000)
> GRADE (value 6 to 12)
> SEX (value 1 to 2)
> ETHNICITY (value 101 to 109
> SCALE1
> SCALE2
> ...
> SCALE110
>
> The scales have values (or SCORES) from 0 to 8
>
>
> What I need is to generate a long file with a COUNT by BUILDING
> by CATEGORY
> (Grade, Sex, or Ethnicity) by SCORE for each SCALE. That is, how many
> FEMALES (CATEGORY 1) in BUILDING == 20 received a SCORE == 1 on SCALE1.
> For example:
>
> BUILDING CATEGORY SCALE SCORE COUNT
> 20 1 SCALE1 1 458
> 20 1 SCALE1 2 657
> 20 2 SCALE1 1 20
> 20 101 SCALE1 1 54
> 20 109 SCALE1 4 333
>
> repeat with each SCALE
>
> repeat for each BUILDING
>
> The individual cases must be weighted using WTSKUL.
>
Ulrich Kohler has made one suggestion. Another
is to use -contract-. Actually, this is the
same suggestion, as -contract- is just
Ulrich's idea wrapped up as a single command.
You can weight with WTSKUL as long as that is an integer.
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/
