Integrating, Preserving, and Disseminating Linked CPS Data

点击翻译按钮获取中文摘要

 DESCRIPTION (provided by applicant): This proposal seeks continuation funding to enhance and expand IPUMS-CPS, featuring integrated data, dissemination software, and associated metadata that dramatically simplifies longitudinal analyses of Current Population Survey (CPS) data, eliminating redundant research effort and opening new opportunities for discovery. IPUMS-CPS freely disseminates data through an innovative user interface that provides powerful tools for search, discovery, research design, and data access. The proposed project will provide researchers with flexible access to integrated and well-documented longitudinal data across six decades of CPS surveys, including all basic monthly surveys from 1976 through 2020, Annual Social and Economic Supplement (ASEC) files from 1962 through 2020, and all other surviving topical supplements from 1967 through 2020. IPUMS- CPS serves the scientific enterprise by reducing wasteful duplication of effort (e.g., in linking files and harmonizing variables), eliminating common technical errors (e.g., in variance estimation), making findings easier to replicate, and encouraging and facilitating sophisticated and powerful new longitudinal analyses in many research domains. To accomplish these goals, we propose six specific tasks: (1) Clean, verify, and de-identify recovered CPS files from the 1960s, 1970s, and 1980s in preparation for disclosure review and integrate them into the public- use IPUMS collection; (2) Expand the database by integrating data from nearly 78 new files that will be released between 2015 and 2020, including all CPS monthly and supplement data as well as data we recover in collaboration with the Census Bureau; (3) Collaborate with the Census Bureau to add IPUMS-CPS linking keys to restricted CPS files for use in the Census Bureau's Research Data Center network; (4) Develop additional new keys for linking CPS observations over time, systematically compare the linkages resulting from different approaches, and provide users with documentation about the keys and differences between them; (5) Create new tools for data access and data processing; (6) Provide user support and perform outreach at conferences, expand online data analysis capabilities, host in-depth training workshops, and implement a new plan for long-run preservation of the microdata and metadata. The proposed work will be carried out by a team of highly-skilled researchers with unparalleled expertise in data integration, record linkage, and the CPS. Collaborators include leading researchers from the Minnesota Population Center, the U.S. Census Bureau, the Bureau of Labor Statistics, the Unicon Research Corporation, and NORC. The project will employ cutting-edge software innovations, extensive usability testing, and established metadata standards.