Online Program Home
  My Program

Abstract Details

Activity Number: 319 - The New Multiple Data Sources Paradigm for Federal Statistics: Progress and Prospects
Type: Invited
Date/Time: Tuesday, August 1, 2017 : 10:30 AM to 12:20 PM
Sponsor: Committee on National Statistics, NAS
Abstract #322074 View Presentation
Title: The New Multiple Data Sources Paradigm for Federal Statistics: Progress and Prospects
Author(s): Robert Groves* and Amy O'Hara* and Premkumar Natarajan* and Jerry Reiter*
Companies: Georgetown University and Stanford University and University of Southern California and Department of Statistical Science, Duke University
Keywords: multiple data sources ; federal statistics ; big data ; data linkage

Federal surveys face declining response rates, tighter budgets and other factors that will likely erode the quality of these surveys as we move forward. Panelists will discuss recent advances toward a new paradigm of integrating multiple data sources in producing federal statistics to take the place of the still dominant survey-only paradigm. This new paradigm requires a more comprehensive set of standards and the creative use of existing data-"big data" sources, commercial data, administrative records, and historical data-together with ongoing surveys. These will be essential to help ensure that federal statistics will continue to be relevant, cost-effective, and timely. The stage will be set by a discussion of guiding principles, prospects and challenges involved in this new paradigm. Panelists will provide specific examples of ongoing work in identifying important alternative sources of data, and in overcoming existing barriers to make these data available for approved use in programs and academic research. Discussion will include approaches to measure more carefully the error structure of estimates produced through linking multiple sources of data.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2017 program

Copyright © American Statistical Association