Evaluating a New Approach for Estimating the Number of U.S. Farms with Adjustment for Misclassification

• Activity Number: 572
• Type: Contributed
• Date/Time: Wednesday, August 6, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Survey Research Methods Section
• Author(s): Denise Abreu*+ and Stephen Busselberg and Andrea C. Lamas and Wendy Barboza and Linda J. Young
• Companies: USDA/NASS and USDA/NASS and USDA/NASS and USDA/NASS and USDA/NASS
• Keywords: Misclassification errors ; area frame ; list frame ; logistic regression

Abstract

In recent years, the National Agricultural Statistics Service (NASS) began a research effort to address an undercount in the estimate of the U.S. number of farms derived from its annual June Agricultural Survey (JAS). Misclassification of farm status was found to be a major cause of the undercount. NASS has evaluated a host of measures and methods to assess, quantify, and account for this misclassification. The approach derived from this process employs record linkage techniques, logistic regression, and NASS's annual list sampling frame. The methods developed and the subsequent results are presented here.