Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: svyset problem 1


From   Peter Muhlberger <pmuhl1848@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: svyset problem 1
Date   Wed, 23 Sep 2009 23:14:00 -0500

Hi Steve:  That's very helpful, thanks!  Now the error msg makes sense.

There was no stratified random sampling of schools (psu's) in this
project, so I didn't specify strata.  I had hoped to be able to use
post-stratification as a way to reweight the data so it would look
more like the population described in my auxiliary data (an
alternative to calculating pweights by hand).  My 'fpc' information
came from what I know about the post-strata N's.  I can omit the fpc
option, and the analysis will run.  Will have to see if this really
gives me the same result as the pweights would.

Peter

On Wed, Sep 23, 2009 at 8:51 PM,  <sjsamuels@gmail.com> wrote:
> 1. The fpc values must be constant within sampling strata, not
> post-strata. Stata believes that there is just one sampling stratum
> (i.e. the whole population)  in your design, because you did not set
> the strata() option.
>
> 2. Finite population corrections are appropriate only for descriptive
> analyses.  Omit the fpc() option, if you are doing  regressions or
> hypothesis tests.   See references at:
> http://www.stata.com/statalist/archive/2009-02/msg00806.html  .
>
> -Steve
>

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index