An official website of the United States government
Here’s how you know
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
Secure .gov websites use HTTPS
A lock (
) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.
Two different maximum overlap procedures were used in the 1990’s redesign to overlap with the 1980’s design. The first procedure, for designs which select one primary sampling unit (PSU) per stratum, was used for the Current Population Survey (CPS) and the National Crime Victimization Survey (NCVS). This procedure is detailed in Ernst (1986). For CPS, the procedure resulted in an average increase in expected overlap of .26 PSU/stratum and for NCVS the procedure resulted in an average increase in expected overlap of .30 PSU/stratum. The second procedure, designed for two-PSU-per-stratum designs, was used for the Survey of Income and Program Participation (SIPP). The procedure resulted in an average increase in expected overlap of .94 PSU/stratum. This procedure is detailed in Section 3.1 of Ernst (1989). Modifications allowing for different PSU definitions in the 1980’s and 1990’s designs are given in Ernst and Ikeda (1992). The modified procedure was the procedure that was implemented. Two 1990’s strata were not overlapped because they contained too many PSUs. The overlap procedure used for SIPP employs a methodology that is more effective in increasing overlap than the procedure used for CPS and NCVS. However, the SIPP procedure requires the assumption of stratum-to-stratum independence in the initial design, an assumption only met by SIPP among these three surveys.
Share
Related Information
WORKING PAPER
Statistical Research Reports and StudiesSome content on this site is available in several different electronic formats. Some of the files may require a plug-in or additional software to view.
Top