Re: st: Weights in survey design

[Steven Samuels <ssamuels@albany.edu>]

population, the  survey design and what were the primary and  
(possibly) second stage and later sampling units?
In a HH or telephone survey, ordinarily the PSU's would be some kind  
of geographic areas,  and the sampling strata for PSU's  cannot be  
sex, as your setup implies.

 Other questions: were the weights post-stratified or raked in any  
way to reflect the population totals?  How did you "reset" the weights?

Steven
On Mar 18, 2007, at 6:46 PM, Jason Ferris wrote:,


> I have a large dataset with weights calculated as PPS based on  
> household
> size, stratified by sex.  The age group respondents are from 16-64.
>
>
>
> I am interested in looking at data only from those aged 16-24.  I can
> use the subpop command "subpop(if age>=16 & age<=24)" for all the
> commands.  But I am wondering if I can drop all other cases (keep if
> age>=16 & age<=24) and the 'reset' my weights based only on those aged
> 16-24.
>
>
>
> In the original form (with all data) I have the following summary  
> data:
> (note the survey design is quiet a simple one)
>
> Svyset
>
>
>
>       pweight: pps
>
>           VCE: linearized
>
>      Strata 1: sex
>
>          SU 1: <observations>
>
>         FPC 1: <zero>
>
>
>
> . svy: tab sex
>
> (running tabulate on estimation sample)
>
>
>
> Number of strata   =         2        Number of obs      =      8664
>
> Number of PSUs     =      8664        Population size    =      8664
>
>                                       Design df          =      8662
>
>
>
> -----------------------
>
>       sex | proportions
>
> ----------+------------
>
>    female |       .5046
>
>      male |       .4954
>
>           |
>
>     Total |           1
>
> -----------------------
>
>   Key:  proportions  =  cell proportions
>
>
>
> If I select the subgroup (age 16-24):
>
> . svy,subpop(if age<=24): tab sex
>
> (running tabulate on estimation sample)
>
>
>
> Number of strata   =         2        Number of obs      =      8664
>
> Number of PSUs     =      8664        Population size    =      8664
>
> Subpop. no. of obs =       999
>
> Subpop. size       = 1438.7586
>
> Design df          =      8662
>
>
>
> -----------------------
>
>       sex | proportions
>
> ----------+------------
>
>    female |       .4599
>
>      male |       .5401
>
>           |
>
>     Total |           1
>
> -----------------------
>
>   Key:  proportions  =  cell proportions
>
>
>
>
>
> When I reset my weights with data only representing those 16-24  
> years of
> age (ie., as if this was the way I original designed my study) I  
> get the
> following results:
>
>
>
> . svy: tab sex
>
> (running tabulate on estimation sample)
>
>
>
> Number of strata   =         2        Number of obs      =       999
>
> Number of PSUs     =       999        Population size    =       999
>
> Design df          =       997
>
>
>
> -----------------------
>
>       sex | proportions
>
> ----------+------------
>
>    female |       .4655
>
>      male |       .5345
>
>           |
>
>     Total |           1
>
> -----------------------
>
>   Key:  proportions  =  cell proportions
>
>
>
> As it can be seen there is now a difference in the proportions between
> using subpop and resetting my weights.  Is this a problem?
>
>
>
>  Jason
>
>
> *
> *   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/
*
*   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/