Adjusting an Area Frame Estimate for Misclassification Using a List Frame

• 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): Andrea C. Lamas*+ and Denise A. Abreu and Hailin Sang and Kenneth Kyle Lopiano and Pam Arroway and Linda J. Young
• Companies: National Agricultural Statistics Service and National Agricultural Statistics Service and National Institute of Statistical Sciences and University of Florida and North Carolina State University and University of Florida
• Keywords: Misclassification ; Area Frame ; List Frame ; Classification and Regression Tree ; Receiver Operating Characteristic curve

Abstract

In recent years, the National Agricultural Statistics Service (NASS) evaluated a variety of approaches to adjust for misclassification in its annual June Area Survey (JAS), which is based on an area frame. This misclassification is a direct cause of an undercount in the number of farms indication produced by the JAS. One approach to correct for this undercount is to use NASS's sampling list frame, which is independent of the area frame. However, recent studies showed that there are farm status inaccuracies on the list frame. These are active records that are not associated with farms. If the list frame farm status inaccuracies are not addressed, the adjusted JAS number of farms indication could become biased upwards. Using Classification and Regression Tree (CART) models, the probability that a list frame record is active can be obtained. This paper evaluates methods for classifying each