Strategies for Subsampling Nonrespondents for Economic Programs

• Activity Number: 160
• Type: Topic Contributed
• Date/Time: Monday, August 4, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Survey Research Methods Section
• Author(s): Stephen Kaputa*+ and Laura Bechtel and Daniel Whitehead and Jenny Thompson
• Companies: U.S. Census Bureau and U.S. Census Bureau and U.S. Census Bureau and U.S. Census Bureau
• Keywords: Quadratic program ; Unit response rate ; Response mechanism ; Nonresponse adjustment

Abstract

Adaptive design strategies for data collection can increase the quality of response data under a reduced survey budget. In this framework, the U.S. Census Bureau is investigating nonresponse subsampling strategies, including a systematic sample of nonrespondents sorted by a measure of size, for usage in the 2017 Economic Census. Design constraints include a mandated lower bound on the Census unit response rate, along with targeted industry-specific response rates. This paper presents research on allocation procedures for subsampling nonrespondents, given a systematic subsample. We consider two approaches: (1) equal-probability sampling and (2) optimal allocation with constraints on unit response rates and sample size with the objective of selecting larger samples in industries that have initially lower response rates. Using the Annual Survey of Manufactures (ASM) sample as our original population, we present a simulation study that examines the cost, variance, relative bias, and unit response rates for the proposed allocations, assessing each procedure's sensitivity by varying the program-level sampling rate, the response mechanism, and the nonresponse adjusted estimator.