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: How to set calibrated weights


From   Steve Samuels <sjsamuels@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: How to set calibrated weights
Date   Fri, 26 Oct 2012 12:01:50 -0400

You are not done yet, Veronica, so hang in there a bit longer. The
793 count can only occur only if the same cluster number occurs in more
than one stratum. That should not happen to primary sampling units, so
there is a mistake some place. The practical consequence is that
standard errors will be wrong, because it is the -svyset- number that
counts.


You locate the problems with the following code:
***************************************
preserve
bys cluster w2_gc_dc: keep if _n==1
bys cluster: gen nstrat= _n
tab nstrat  //duplicates are nstrat >1
sort cluster
list cluster w2_gc_dc if nstrat>1 //show problem
save problem, replace
restore
*************************************
Send your findings to the survey organization and ask them what is going on.


Steve



On Oct 26, 2012, at 9:41 AM, Veronica Galassi wrote:

Dear Steve,



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