Association of Public Data Users 2010 Conference Opportunities and Challenges for the U.S. Statistical System and ERS’ Role Katherine R. Smith Administrator,

Association of Public Data Users 2010 Conference

Opportunities and Challenges for the U.S. Opportunities and Challenges for the U.S. Statistical System and ERS’ RoleStatistical System and ERS’ Role

Katherine R. SmithAdministrator, Economic Research Service, USDA

Challenges Facing the U.S. Statistical System in the Next Decade

Why Encourage Micro Data Access?

• “Data are valuable only to the extent that they can be put to practical use….Improvement in data access coming this fall will make ARMS data tremendously valuable, while maintaining the strong disclosure and data security features that protect the respondents” Katherine Smith, Director, Resource Economics Division, Economic Research Service

• “Alternative means of dissemination …..may increase access for some researcher or facilitate analysis that combines data sets”

• “The Advisory Committee recommends conducting a pilot study of making the ARMS micro data available through the National Institutes of Standards and Technology and National Opinion Research Center (NIST-NORC) data enclave”

NORC Data Enclave: Mission

• Promote access to sensitive microdata

• Protect confidentiality (portfolio approach)

• Archive, index and curate micro data

• Encourage researcher collaboration / virtual collaboratory

• Leverage cutting edge technologies (e.g., metadata preparation & documentation best practices)

Enclave Sponsors

• U.S. Department of Commerce • (NIST-TIP)• U.S. Department of Agriculture (ERS/NASS)• Ewing Marion Kauffman Foundation• National Science Foundation• Annie E. Casey Foundation • Center for Medicare and Medicaid Services

Data Enclave Today

• Approximately 200 active researchers

• Approximately 500 on enclave listserv

• Increasingly accepted as a model for providing secure remote access to sensitive data:

• U.K. Data Archive (Secure Data Service), Council on European Social Science Archives (CSSSDA), University of Pennsylvania, Columbia University, University of Chicago, Chapin Hall Center for Children

• Emphasis on building and sustaining virtual organizations/ researcher collaboratories

• Developing customized XML metadata driven tools (for archiving, disclosure control processing)

Select Tools

• Stata• StatTransfer • SAS • SPSS• “R” Project for Statistical Computing• MATLAB• MPlus• LimDep / NLogit• Microsoft Office 2007• LISREL• Adobe PDF Reader• IHSN Microdata Management Toolkit• SuperSTAR• Notepad ++

Create single-indicator map: Fast food restaurants per capita

Child food insecurity by state and counties in which 20 percent or more of low income preschool children are obese

Create layered maps: Child food insecurity/obesity

April 2010 Behavioral Economics Conference

A New Survey on Food Purchases

Linking Administrative Data

SCOP cloud-based virtual data center with agency access to shared software, metadata, quality control protocols, and publicly available statistical data

(May 2010)

Statistical Agency #1

Statistical software

Statistical Data Set #1; metadata

Statistical software

Statistical Data Set #2, #3; metadata

Statistical software

Statistical Data Set #N; metadata

Statistical Agency #2

Statistical Agency #N

Partners and Collaborators,

Data.gov

Cloud providers could host different public statistical data sets with access controls for specific software products as identified by parent agency.

ERS Distance Learning

Contact Information

• Katherine (Kitty) R. Smith• Administrator• Economic Research Service• [email protected]• 202-694-5000