Current Projects
Search
 

Home
Search for Projects
View Projects
Project Title
by Subject/Focus Area
by Board/Committee
by Major Unit
Provisional Committee Appointments Open for Formal Public Comments
by Last Update
Meeting Information
Conflict of Interest Policy
Committee Appointment Process
FAQ
  Project Information

Project Information

 Printer Friendly Version

Project Title:

Using Multiple Data Sources and State-of-the-Art Estimation Methods in Federal Statistics: Frameworks, Methods, and Assessment
PIN: DBASSE-CNSTAT-13-09        

Major Unit:

Division of Behavioral and Social Sciences and Education

Sub Unit: Committee on National Statistics

RSO: Harris-Kojetin, Brian

Subject/Focus Area: Behavioral and Social Sciences


Project Scope
An ad hoc panel of nationally renowned experts in social science research, computing technology, statistical methods, privacy, and use of alternative data sources in the United States and abroad will conduct a study with the goal of fostering a paradigm shift in federal statistical programs. In place of the current paradigm of providing users with the output from a single census, survey, or administrative records source, a new paradigm would use combinations of diverse data sources from government and private sector sources combined with state-of-the art methods to give users richer and more reliable statistics leading to new insights about policy and socioeconomic behavior. The motivation for the study stems from the increasing challenges to the current paradigm, such as declining response rates and increasing cost and burden for surveys.
The panel will prepare two reports as part of this study:

First Report
The first report will discuss the challenges faced by the federal statistical system; the current paradigm of providing users with the output from a single census, survey, or administrative records source and that paradigm's increasing disadvantages for meeting user needs; and the foundational elements needed for a new paradigm.
More specifically, the first report will:
• review federal statistical agencies' current paradigm for producing national statistics and challenges to this paradigm;
• review federal statistical agencies' legal frameworks and mechanisms for protecting the privacy and confidentiality of their data and examine challenges to those frameworks and mechanisms;
• review federal statistical agencies' legal frameworks and mechanisms for providing access to underlying data to researchers to foster transparency, replicability of statistical series, and for policy and social science research and examine challenges to those frameworks and mechanisms;
• examine federal statistical agencies' access to alternative sources of data for federal statistical programs, the organizational structures sustaining access, and the impediments to access;
• discuss the characteristics of a new paradigm for federal statistical programs that would combine diverse data sources from government and private sector sources with state-of-the art methods to give users richer and more reliable statistics; and
• discuss the foundational elements needed for a new paradigm.
The first report will contain findings and conclusions from the panel's deliberations and recommendations for steps needed to lay the foundation for a new paradigm.

Second Report
The second report will propose approaches for implementing a new paradigm that would combine diverse data sources from government and private sector sources with state-of-the art methods to give users richer and more reliable statistics.
The second report will:
• assess alternative approaches for implementing a new paradigm that would combine diverse data sources from government and private sector sources;
• evaluate concepts, metrics, and methods for assessing the quality and utility of alternative data sources, analogous to the “total error” framework used for surveys;
• evaluate statistical models for combining data from multiple sources;
• examine metrics and methods for evaluating the quality of combined-information estimates;
• evaluate alternative designs of statistical processes that foster privacy protections, transparency, objectivity, timeliness, replicability, efficiency, and continuity of statistical series; and
• identify priorities for research needed for federal statistical agencies to advance a multiple-data-sources paradigm.
The second report will contain findings, conclusions, and recommendations for actions toward implementing a new multiple-data-sources paradigm for federal statistics.

The project is sponsored by Laura and John Arnold Foundation and the President's Committee.
The approximate start date for the project is 3/26/2015.
The report will be issued at the end of the project in approximately 24 months
This statement of task was revised on 07/21/2016.


 
Project Duration: 30 months    

Provide FEEDBACK on this project.

Contact the Public Access Records Office to make an inquiry, request a list of the public access file materials, or obtain a copy of the materials found in the file.


Committee Membership
Committee Membership

Meetings
 Meeting 1 - 09/04/2015
 Meeting 2 - 12/15/2015
 Meeting 3 - 02/25/2016
 Meeting 4 - 06/01/2016
 Meeting 5 - 09/23/2016
 Meeting 6 - 12/14/2016
 Meeting 7 - 03/16/2017

Reports

Reports having no URL can be seen
at the Public Access Records Office