662 – Statistical Challenges in Combining Survey and Administrative Data
Combining Survey and Administrative Data to Produce Official Statistics
Andreea L. Erciulescu
National Institute of Statistical Sciences
Nathan B. Cruze
USDA National Agricultural Statistics Service
Balgobin Nandram
Worcester Polytechnic Institute, Worcester, Massachusetts
Combining survey and ancillary data to produce official statistics is gaining interest at federal agencies and policy makers due to its efficiency, i.e., an increase in reliable estimates at detailed disaggregation levels, a decrease in allocated budgets and an increase in usability of available data. The United States Department of Agriculture's National Agricultural Statistics Service publishes annually more than 400 reports, including county-level estimates that play an important role in the allocation of funds in some agricultural programs. In this paper, small area estimation modeling approaches are considered that borrow information across areas and from ancillary data, to produce reliable county-level agricultural estimates. Challenges in assessing the quality and the usability of different data sources are discussed.