Nonresponse Mitigation in the Quarterly Summary of State and Local Tax Revenues

• Activity Number: 542
• Type: Contributed
• Date/Time: Wednesday, August 7, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Government Statistics Section
• Author(s): Joseph Barth*+ and Courtney Hill and Bac Tran
• Companies: U.S. Census Bureau and U.S. Census Bureau and U.S. Census Bureau
• Keywords: nonresponse ; calibration ; sample survey ; adaptive design ; response propensity ; government statistics

Abstract

The Quarterly Summary of State and Local Tax Revenues (QTax) was designed to estimate quarterly property, sales, personal income, corporate income, and other taxes for state and local governments. The Bureau of Economic Analysis uses these estimates to develop estimates of the Gross Domestic Product (GDP). In this paper, we discuss the methodologies considered when the redesigned local non-property component of QTax produced lower response rates than expected for several quarters. Instead of the traditional approach of conducting Non-Response Follow Up (NRFU) on all non-responding units, an adaptive design where a sample of chronic non-responders was selected for focused follow-up. Calibration estimators were considered, both with and without a response propensity model. We consider not only traditional calibration estimators in which auxiliary data are available either at the population level (InfoU) or only for the full sample (InfoS), but also the case where estimated values are used in the calibration. In this paper, we present simulation studies validating these approaches as well as the underlying theory.