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

st: How to svyset for the SIPP


From   Stephen Mennemeyer <smenneme@uab.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: How to svyset for the SIPP
Date   Thu, 10 Feb 2005 17:17:07 -0600

Dear Statalisters:

Can anyone give me some guidance on how to analyze the Survey of Income
and Program Participation ( SIPP), especially the 1996 version, to take
account of the complex sample design?  I have access to both SAS/SUDAAN
and Stata.

Assuming for the moment that I use the svy commands in Stata and that I
want to do longitudinal analysis, I think I want to use the svyset
command  as follows:

svyset wpfinwgt, strata(gvarstr)

where wpfinwgt is the longitudinal weight for individuals and gvarstr is
the "variance stratum code".

I  am confused about whether I can or should do anything with the
options for PSU and FPC.

According to the SIPP Manual
http://www.census.gov/apsd/techdoc/sipp/sipp96l.pdf  page 8-1:

"The 1996 Panel of the SIPP sample is located in 322 Primary Sampling
Units (PSUs), each
consisting of a county or a group of contiguous counties. Within these
PSUs, living quarters
(LQs) were systematically selected from lists of addresses...."

As far as I can tell, there is no SIPP variable for the PSU. The  PSU
code is scrambled inside the ssuid variable (the household ID number)
but I do not think there is any way to tell which ssuids came from  the
same PSU.

However, from reading the Stata 8 manual  at U [30] p. 346-347 I wonder
if I should use the command:

svyset wpfinwgt, strata(gvarstr) psu(ssuid) fpc(epppnum)

where epppnum is the individual person identifier within the household.

I think this is wrong but my logic here is that the SIPP is sampling
individuals who are "clustered"  in households where every member of the
household is interviewed. I am particularly concerned about the remark
in the Stata Manual U 30.2.2 p. 347 "For example if our PSUs were were
households and we included every member of the household in our study,
then a finite population correction term would be appropriate where the
households are sampled using simple random sampling without replacement
in each stratum"

Guidance would be much appreciated.

-- 
Stephen T. Mennemeyer Ph.D.
University of Alabama at Birmingham
School of Public Health
Dept. of Health Care Organization and Policy

U.S. Mail:
1530 3rd Ave. South 330 RPHB
Birmingham, Al 35294-0022

Express Delivery:
330 Ryals Public Health Building
1665 University Blvd.
Birmingham, Al  35294-0022

Phone: (205) 975-8965
FAX (205) 934-3347
e-mail: smenneme@uab.edu

*
*   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