Pandata Systems to Enhance Survey and Census Systems: Employing Administrative Data to Better Inform Public Policy

• Activity Number: 51
• Type: Invited
• Date/Time: Sunday, July 31, 2011 : 4:00 PM to 5:50 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Young Chun*+ and Fritz Scheuren
• Companies: U.S. Census Bureau and NORC at the The University of Chicago
• Keywords: Administrative data ; public policy ; paradigm ; Register-based census ; paradata ; pandata

Abstract

Administrative data have long been looked to as a way of sustaining or increasing the quality and interpretability of surveys and censuses, while controlling for the rising cost of data collection. In this paper we revisit this familiar topic proposing a "pandata systems" approach. Pandata are the purpose-driven data produced by integrating strengths of multiple data files that link administrative data to paradata and metadata. A modeling navigator is used which guides selection of variables and ties data fitness to purposes. We demonstrate the merits plus shortcomings of our pandata systems approach, with examples fit to purposes of public health, education and economic policy research, in striking a balance between cost and information quality. We assess indicators of information quality such as relevance, coverage, timeliness and comparability.