Using Paradata to Design the Quarterly Tax Sample

• Activity Number: 51
• Type: Invited
• Date/Time: Sunday, August 3, 2014 : 4:00 PM to 5:50 PM
• Sponsor: Survey Research Methods Section
• Author(s): Brian Dumbacher*+ and Carma R. Hogue
• Companies: U.S. Census Bureau and U.S. Census Bureau
• Keywords: Paradata ; Sample design ; Response propensity ; Logistic regression

Abstract

The Quarterly Survey of Selected Non-Property Taxes (F-73) is a sample survey conducted by the Governments Division of the U.S. Census Bureau that collects data on local government revenue from non-property taxes. During its 2009 sample design, F-73 experienced a response rate that was lower than the U.S. Census Bureau's standard. In 2013, the Governments Division redesigned the sample and narrowed the scope of the survey from 11 taxes to three to try to reduce respondent burden, increase the response rate, and increase data quality. In this paper, we describe how paradata, which are data about the data collection process, were used in part of the sample redesign to select units with a high estimated response propensity. We also describe evaluation simulations that were performed to validate the redesign and fine-tune parameters.