Creating and Validating Synthetic Data (NCSES/Census, Annual Business Survey) (ABSSyn)

Award Date: 09/28/2023
Project Number: ABSSyn-23-N06
Project Title: Creating and Validating Synthetic Data (NCSES/Census, Annual Business Survey)
Project Performers: Knexus Research Corp., RTI International, Dr. Gillian Raab (University of Edinburgh), Dr. Amy O’Hara (Georgetown University), Dr. Alan Karr (Temple University and AFK Analytics)
Award Amount: $997,916.00
Objective: The objective of this project is to test and compare methods for creating synthetic data to support a tiered access model; explore the use of synthetic data for evidence-building; and test the use of verification metrics in validating estimates produced from synthetic data. The Annual Business Survey (ABS) is an annual survey conducted by the U.S. Census Bureau on behalf of multiple sponsors, one of whom is the National Center for Science and Engineering Statistics. One objective in disseminating ABS data is the production of a public use microdata file that could be used by researchers, policy makers, and public-data users. To address disclosure concerns, a synthetic datafile option is being explored. This project would build on this work to inform the National Secure Data Service Demonstration Project (NSDS-D) and support tiered access through the development of a publicly available dataset that could be used without the barriers involved in accessing the restricted-use data.
Period of Performance: September 2023 – March 2026
Point of contact: Christine Task| Knexus Research Corp. | [email protected]

 

Request for Solutions (Closed)

Overview Release Date Submission Date Documents
The Government has an interest in testing and comparing methods for creating synthetic data to support a tiered access model; exploring the use of synthetic data for evidence-building; and testing the use of verification metrics in validating estimates produced from synthetic data. 07/10/2023 07/31/2023