MPC Member Publications

This database contains a listing of population studies publications written by MPC Members. Anyone can add a publication by an MPC student, faculty, or staff member to this database; new citations will be reviewed and approved by MPC administrators.

Full Citation

Title: Working Papers Including the ASEC Oversamples in Linked CPS Data Including the ASEC Oversamples in Linked CPS Data

Citation Type: Journal Article

Publication Year: 2025

DOI: 10.18128/IPUMS2025-01

Abstract: The Current Population Survey (CPS) is an important source of labor force data, and its Annual Social and Economic Supplement (ASEC) provides additional information about health insurance coverage, income, and poverty in the United States. The ASEC questionnaire is administered to 1) all respondents to the March CPS Basic Monthly Survey (BMS), 2) an oversample of Hispanic households from 1976 forward, and 3) an oversample of households with children or non-white household members from 2001 forward. The CPS also has a short-term panel component. Creating CPS ASEC panel datasets by linking files from adjacent years allows for the examination of short-run change in health insurance coverage, income, and other topical areas. Many common methods used to link the CPS ASEC files only include March BMS respondents in the ASEC panels; however, many oversample individuals can also be linked. The main challenges to including CPS ASEC oversample records in ASEC panels are the multi-step process required for linking, bridging changing household identifiers over time, and handling duplicated household identifiers and false positive matches that arise due to the recycling of Census Bureau household identifiers prior to 2005. This paper describes our approach for addressing the complexities and challenges of linking ASEC oversamples across adjacent years and the method by which these oversample records were incorporated into the custom IPUMS CPS linking keys CPSID, CPSIDP, and CPSIDV for the period from 1989 to 2024. Properly including these records in linked ASEC data substantially increases the linked sample size. From 1976-2001, linked ASEC sample sizes are between 4 and 9% higher, and from 2002 forward, linked ASEC sample sizes are more than 30% higher than when only the March BMS respondents are included in the panel.

Url: https://doi.org/10.18128/IPUMS2025-01

User Submitted?: No

Authors: Rodgers, Renae

Periodical (Full):

Issue:

Volume:

Pages:

Countries:

IPUMS NHGIS NAPP IHIS ATUS Terrapop