[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: svyset where PPS and two stage cluster sampling is used

From   "arka roy" <>
Subject   st: svyset where PPS and two stage cluster sampling is used
Date   Wed, 25 Jun 2008 10:05:55 -0700

   I am working on Stata 9 with a dataset which lists educational
outcomes of individuals.The dataset was formed as
follows:first the population was divided into a number of
strata(districts),then from each district PSUs(villages) were selected
using probability proportional to size(PPS) with replacement and from
each village SSUs(households) were selected using SRS without
replacement.I need district level estimates-so I am working with one
district at a time.I also calculated the sampling weight for each
observation.To svyset the data I used the following command:

svyset vilage[pw=sampwt] || household,fpc(nhouseholds)

However Stata gives a note which states that since stage 1 is sampled
with replacement, all further stages will be ignored.But the fact is
that I have clusters of unequal sizes.So if the future stages are
ignored,would not there be a problem?I looked in the statalist
archives but could not find any solution.Could you please advice how
to get around this problem?Thanks

Arka Roy Chaudhuri
Indian Statistical Institute
*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index