Boosting Sample Representation and Response Rate of an ABS Random Household Survey via Blending a Non-Random Approach: An Innovation to Reach the Hard-to-Reach

• Activity Number: 82 - Contributed Poster Presentations: Government Statistics Section
• Type: Contributed
• Date/Time: Monday, August 3, 2020 : 10:00 AM to 2:00 PM
• Sponsor: Government Statistics Section
• Author(s): Urban Landreman*
• Companies: Hennepin County
• Keywords: ABS; sampling; health; blended data

Abstract

SHAPE is a population-based health survey for Hennepin County, Minnesota since 1998 - www.Hennepin.us/SHAPE. With response rates declining over time, it was harder to be confident that the survey respondents represented the population as a whole. To address this, responses from a sample of adults applying for County services were blended with the ABS household survey data. The results look quite promising as a way to inexpensively increase the representativeness of the survey respondents and as a model that other government agencies can use to address steadily lower response rates.