Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: Survey question


From   Stas Kolenikov <skolenik@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Survey question
Date   Tue, 4 May 2010 13:11:56 -0500

On Tue, May 4, 2010 at 12:37 PM, Michael Norman Mitchell
<Michael.Norman.Mitchell@gmail.com> wrote:
>  We have a sampling frame of people (a list of say 7000 people) and we are
> going to sample of say 300 of them. All people in the sampling frame will
> have an equal probability of being selected.
>
>  Assume that we do not want to compute "totals" (i.e, counts, frequencies)
> in the population, and just focus on things like percentages, means, and
> estimation commands (like regress). In such a case, is there any need or
> benefit to calculate -pweights- and use the -svy- commands?

Even if you use SRS, there's still a small gain in precision because
of the finite nature of the population captured by the finite
population correction, 1-300/7000. (If you sampled 7000 out of 7000,
your uncertainty will be zero, which is what the fpc is reflecting,
among other things.) You don't need anything at all for this, just

svyset, fpc(7000)

-- 
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/


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