Re: st: Socio-economic multi-variable analysis

["Michael Blasnik" <michael.blasnik@verizon.net>]

I assume that you variables are named q48a-q48k since you can not start a 
variable name with a number in Stata (aside--why do people seem to misquote 
their own data so often on Statalist??).  Anyway, this should do what you 
want and create one long list with lines dividing it by number of yes 
answers.

egen numyes=rsum(q48q-q48k)
sort numyes
l idno numyes , sepby(numyes)

M Blasnik

----- Original Message ----- 
From: "Todd Swarthout" <todd.swarthout@medecinsdumonde.net>
To: <statalist@hsphsun2.harvard.edu>
Sent: Friday, December 08, 2006 9:17 AM
Subject: RE: st: Socio-economic multi-variable analysis



> Im sorry, let me try to explain again...I suspect the answer is quite 
> easy.
>
> I have 350 questionnaires, each with an ID number (variable = idno). Each 
> questionnaire can have up to 11 total yes responses to the 11 questions 
> having to do with socio-ecinomic status (variables 48a - 48k). I would 
> like to know for each ID number how many yes ('1') respondes there are 
> within these 11 questions...without having to do a command for each ID 
> number.
>
> Maybe something like "list idno if 48a-48k 'contain' 1 yes 
> responses"...and then perhaps I could do this command for each possible 
> number of 'yes' responses.
>
> "list idno if 48a-48k 'contain' 1 yes responses"
> "list idno if 48a-48k 'contain' 2 yes responses"
> "list idno if 48a-48k 'contain' 3 yes responses"
>
> Etc...
>
> In the end Id like the table
>
> ID No./ No. 'yes' responses in questions 48a - 48k
> /
> /
*
*   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/