Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: sample weights of 2 related surveys


From   "Sayer, Bryan" <BSayer@s-3.com>
To   "'Marina Fe B. Durano '" <marina.f.durano@stud.man.ac.uk>, "'statalist@hsphsun2.harvard.edu '" <statalist@hsphsun2.harvard.edu>
Subject   st: RE: sample weights of 2 related surveys
Date   Wed, 26 Mar 2003 09:39:26 -0500

Sampling weights are generally designed to reflect the composition of a
specific population.  If those two (or more) populations are mutually
exclusive but combined represent some useful population then you just need
to concatenate the two datasets, being mindful that the scale of the weights
remains the same.

But if the populations overlap, then you would need to rescale the weights
of the overlap portion, preferably along the same categories that the
original weights were created. 

But this specific instance sounds like they are not designed to be analyzed
together.  Otherwise, I think they would already be joined.  Possibly the
two sets of questions are very different.  In which case you simply analyze
them seperatly.

Bryan Sayer
Statistician, SSS Inc.

 

-----Original Message-----
From: Marina Fe B. Durano
To: statalist@hsphsun2.harvard.edu
Sent: 3/26/03 7:46 AM
Subject: st: sample weights of 2 related surveys

Hello, all! I have a question on what to do with the sample weights when
combining cases of two related surveys.

Situation:
A labor force survey was conducted. Each household was asked wether a
person
left for abroad in the last 5 years. If yes, another set of questions
was
asked. There are two datasets: one on the domestic labor force and the
second on the labor force that went abroad.

Both datasets use the same household sampling frame. Cases for each
dataset
have sample weights based on population density and adjusted for
enumeration
areas that were not enumerated, non-interview households, and population
projection. The two datasets are treated separately with their own set
of
sample weights.

Question:
If I combine the two datasets to get total labor force in and out of
country, how do i recompute the sample weights?

thanks,
marina


Marina Fe B. Durano
Dover St. Building, Room N.3.10
School of Economic Studies
University of Manchester
Manchester M13 9PL
United Kingdom
Tel: +44(161)2754516




*
*   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/



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