Linking Federal Data on Agriculture: Pathways toward Better Understanding of
Agricultural Sector Dynamics and Program Effectiveness: A Workshop355 E Street, SW, Washington, DC
October 2, 2015
Co-sponsored by the USDA’s Economic Research Service and the
Council of Professional Associations on Federal Statistics
Information on aspects of U.S. agriculture is abundant. The USDA alone conducts surveys of farm practices, farm and ranch operators, and farm households, maintains geographical data bases on farmed land, tracks animal health and veterinary practices, and maintains records on participants in conservation, price support, lending, nutrition assistance and crop insurance programs. Federal agencies beyond USDA collect and maintain in geographical information systems data on the physical and quality features of natural resources potentially affected by farming and ranching. IRS data on Farm Sole Proprietorship are collected on Schedule F of Form 1040. Other IRS administrative data on farms include farm net income or loss. The Census Bureau collects data on a wide spectrum of agricultural and agriculturally-related businesses. BLS collects workforce statistics, unemployment and earnings data, work-related fatalities, injuries and illnesses, import and export prices, and workplace trends for crop production and animal production sectors and industries providing support for agriculture.
Each statistical survey, physical survey, and administrative data set has unique uses and value on its own. But the ability to link and examine cross-series phenomena opens huge possibilities for a better understanding of independent agricultural program effectiveness, cross-program interactions, and sectoral dynamics. Recent research papers in the economics literature on inequality, social mobility, and health care policy have demonstrated the value of building data sets that include administrative data, information about local conditions, and links to survey data.
These data sets can be very large (approaching big data sizes) and take advantage of IT improvements. Building new data requires investment not just in IT, but also in the definition and consistency of terms, bases for accurate linkage, and access protocols. For example including program participation data and location information on farmers or ranchers could raise privacy questions. However, many potential strategies capitalize on existing information and provide a way to add value to current information and meet research needs with stable or declining federal budgets for statistical activities. The overall goal of the project is to achieve better data linkage for socially beneficial illumination of agricultural structure and performance and agricultural program effects and interactions.
A Workshop was held October 2, 2015. Its objectives were to
- Highlight potential benefits of federal agricultural data linkage and integration;
- Explore the bases for successes in agricultural data linkage and integration
- Pinpoint obstacles to wider success through discussion between data users and agency producers
- Identify needed investments and other actions that could be taken to enhance socially beneficial agricultural data linkage and integration
The program for the Workshop, a list of workshop participants, and copies of slide presentations from the workshop are available here. A report on the workshop and resultant recommendations is forthcoming.
PRESENTATIONS: (please click on the presentation title, PDF will open in a new tab)
9:30 Presentations on and discussion of conservation and farm program linkage
Moderator – Marca Weinberg, Director, Resource and Rural Economic Division, ERS
- Roger Claassen and Vince Breneman, Economic Research Service, "Developing spatial data on agri-environmental outcomes linked with farm and farm program data." (pdf, 934 KB)
- Patrick Flanagan, Natural Resources Conservation Service (NRCS), Resource Inventory Division, “Linking the National Resources Inventory to the World” (pdf, 779KB)
- Josh Woodard, Cornell University, "Building a data warehouse to enable analysis of farm, conservation and risk management programs"
- Mike Woodside, U.S. Geological Survey, “Nutrient Levels in Streams and Rivers Are Changing--Unraveling "why" requires spatial and temporal information on agricultural conservation actions”
12:45 Presentations on and discussion of linked statistical analysis of structure and performance
Moderator: James MacDonald, Chief, Structure, Technology, and Productivity Branch, ERS
- Ani Katchova, Ohio State University, “Strategies to Link Farm Business and Household Data Over Time” (pdf, 354 KB)
- Kirk White, Center for Economic Studies, Bureau of the Census “Opportunities to link Census and USDA Data” (pdf, 177 KB)
- Brent Hueth, University of Wisconsin “Capitalizing on Census Bureau and other non-USDA Statistics on Agricultural Business” (pdf, 1.9 MB)
- Richard Dunn, University of Connecticut “Potentials and Pitfalls of Administrative Data in Studying the Agricultural Services Sector” (pdf, 856 KB)