Is It Feasible to Use a Sampling List Frame to Evaluate Misclassification Errors of an Area-Frame-Based Survey?

• Activity Number: 226
• Type: Topic Contributed
• Date/Time: Monday, August 1, 2011 : 2:00 PM to 3:50 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Denise A. Abreu*+ and Andrea C. Lamas and Hailin Sang and Pam Arroway and Kenneth Kyle Lopiano and Linda J. Young
• Companies: National Agricultural Statistics Service and National Agricultural Statistics Service and National Institute of Statistical Sciences and North Carolina State University and University of Florida and University of Florida
• Keywords: Misclassification Errors ; Area Frame ; List Frame ; Record Linkage ; Re-screening Survey

Abstract

During the past three years, the National Agricultural Statistics Service (NASS) has made an effort to address, quantify and adjust for an undercount in the number of farms indication on its annual June Area Survey (JAS), which is based on an area frame. This undercount is a direct result of the misclassification of agricultural tracts as non-agricultural. The 2007 Census of Agriculture mailing list (CML) was evaluated as a potential source to assess misclassification on the 2007 JAS. This evaluation revealed that the CML was a rich source from which to quantify the undercount of farms on the JAS. However, the CML is only available every five years and misclassification on the JAS should be assessed each year. Independently of the area frame, NASS maintains a list of agricultural operators, referred to as the list frame. Yearly list-based samples are selected from the list frame. In addition, the list frame serves as the foundation for building the CML. The list frame is updated on an on-going basis and operators are categorized as either active or inactive. This paper discusses the feasibility of using the list frame to assess misclassification on the JAS.