Statalist


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

Re: st: svyset problem 2: using svy with partially complete surveys


From   Peter Muhlberger <pmuhl1848@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: svyset problem 2: using svy with partially complete surveys
Date   Thu, 24 Sep 2009 13:35:00 -0500

Hi Michael:  Thanks!  Yes, this is clearly the way Stata treats it.
Are you also saying, however, that this is theoretically the way it
has to be?  Supposing I have no strata (err, one big strata) and 5
post-strata, wouldn't it be conceivable to correct for the number of
known population units in each post-strata?  If I keep working w/ this
kind of data, I guess it'll be time for me to pick up a book that
carefully derives each of the formulas behind svy.

Peter

On Thu, Sep 24, 2009 at 7:55 AM, Michael I. Lichter
<mlichter@buffalo.edu> wrote:
> Peter,
>
> The FPC derives from sampling and has nothing to do with any
> poststratification adjustments you apply. If you have a single stage,
> clustered, unstratified design, you should have only one FPC value, equal to
> the sampling fraction. You do not want to calculate poststratum-specific
> FPCs.
>
> Michael
>
*
*   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