Administrative Records for Survey Methodology 1
|Convenor:||Dr Asaph Young Chun|
|Affiliation:||US Census Bureau|
Incorporation of administrative records have long been regarded as a way of improving the quality and interpretability of surveys and censuses and of controlling the rising cost of surveys (Chun and Scheuren, 2011). The increasing number of linked datasets, such as Health and Retirement Study in the U.S., National Educational Panel Study in Germany, and Understanding Society in UK, are accompanied by growing empirical evidence about the selectivity of linked observations. The extent and pace of using administrative data varies from continent to continent and from country to country. This is partly due to differential concerns about privacy, confidentiality, and legal constraints, as well as variability in acceptance and implementation of advances in statistical techniques to control such concerns.
The primary goal is to control data quality and reduce total survey error. This session will feature papers that implement "total administrative records error" and “total linked data error” methods and provide case studies and best practices of using administrative data tied to the survey life cycle (Chun and Larsen, a forthcoming Wiley book). The session invites papers that discuss fundamental challenges and recent advancements involved in the collection and analysis of administrative records, integration with surveys, censuses, and auxiliary data. We also encourage submission of papers discussing institutional collaboration on linked data, sustainable data access, provision of auxiliary tools and user support. For example, papers in this session include, but are not limited to the following topics:
1.Innovative use of administrative data in household surveys and censuses to improve the survey frame, reduce nonresponse follow-up, and assess coverage error.
2.Quality evaluations of administrative data and quality metrics for linked data
3.Recent advancements in processing and linking administrative data with survey data (one-to-one) and with multiple sources of data (one-to-many).
4.Recent methods of disclosure limitation and confidentiality protection in linked data, including linkages with geographical information.
5.Bayesian approaches to using administrative data in surveys, censuses, small area estimation, and nonresponse control.
6.Implementation of new tools that facilitate the use of linked data by simplifying complex data structures or handling inconsistent information in life-course data
7.Strategies for developing and maintaining a user-friendly infrastructure for the analysis and dissemination of linked data and solutions for collaboration
8.Applications that transform administrative data into information that is useful and relevant to policymaking in public health, economics, science and education.