Re: st: -weight- spss: weight by -- telling stata the weight

[Steve Samuels <sjsamuels@gmail.com>]

These variables can't be used to compute weighted averages unless the "weight" variables are normalized to sum to 1. And even then, weighted averages and standard errors for subgroups will be wrong.

In Stata support for weights is built into most commands (and into all the survey commands, as Stas suggests). Typing "help weight" would be a good start.

Steve
sjsamuels@gmail.com

On Feb 2, 2012, at 3:06 PM, Stas Kolenikov wrote:
> 
> See
> 
> help survey
> 
> and
> 
> help svyset
> 
> On Thu, Feb 2, 2012 at 2:46 PM, David Cefskimal
<david.cefskimal@googlemail.com> wrote:
> dear statalist,
> 
> please help with hopefully simple problem:
> 
> imagine that you have 4 variables in your dataset
> 
> weight
> age
> sex
> country
> 
> in spss if you say "weight by weight" computer will know, that now all
> your variables should be multiplied with the variable weight. so in
> your computer's "mind" the new dataset will look like:
> 
> age*weight
> sex*weight
> country*weight
> 
> in stata there is nothing like that? Do I really need to generate
> variables I use with weights each time, like
> gen (v1)=v11*w1
> 
> THANKS
> *
> *   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/



-- 
Stas Kolenikov, also found at http://stas.kolenikov.name
Small print: I use this email account for mailing lists only.

*
*   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/