Re: st: Calculating percentages (proportions) of indicator variables

[Abhimanyu Arora <abhimanyu.arora1987@gmail.com>]

I am sure there must be better ways. But one long tedious way could be to

-generate 4 (indicator) variables for each response
-by country generate total for each response corresponding to the
above variables, lets say bi (i=1-4)
-generate total obs var for each country bys country:gen new_var=_N

-and within a forvalues loop (i=1/4)
generate percentage_share=bi/new_var if response==i




On Mon, Aug 9, 2010 at 2:39 PM,  <rado645-bg@yahoo.de> wrote:
> Dear all,
>
> I hope that you would help me with the following technical problem.
>
>
> I have a dataset based on a question with four different possible answers. The
> question takes the form of an indicator variable with 4 different numeric values
> (1,2,3,4).
>
> I have data for individuals from several countries.  I need to generate a new
> variable that contains the fraction (i.e. percentage share) of each response  to
> the question within each country. However, I cannot find any online source on
> calculating percentages with mutliple responses in Stata (more than two).
>
> Example:
>
> country: response:
>
> 1               3
> 1               3
> 1               2
> 1               1
> 1               4
> 1               3
> 2               1
> 2               1
> 2               2
> 2               3
> 2               4
> 3               2
> 3               2
> 3               4
> 3               3
> 3               1
>
>
>
> *
> *   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/