Obtaining Population Representation with an Address-Based Sampling Design

• Activity Number: 654
• Type: Topic Contributed
• Date/Time: Thursday, August 5, 2010 : 10:30 AM to 12:20 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Michael Link*+ and Jennie Lai and Charles Shuttles
• Companies: The Nielsen Company and The Nielsen Company and The Nielsen Company
• Keywords: Address Based Sampling ; Survey Methodology ; Sample Frames ; Representation ; Mixed Mode Surveys

Abstract

Obtaining a representative sample of completed interviews is a key goal of all survey efforts. With the advent of address-based sampling (ABS) -- sampling addresses from a database with near universal coverage - it is critical that researchers develop survey methodologies which work in tandem with sample designs to achieve this goal. Using a large ABS mixed-mode study of 150,000 respondents conducted in 2009, we explore three key questions: (1) What are the advantages provided by an ABS design to help in targeting hard-to-reach groups (younger adults, Hispanic, Black, and cell phone only)? (2) What are the challenges in reaching these groups when an ABS design is used? and, (3) How successful have efforts been in achieving representative samples? The findings further both our understanding of the strengths and limits of an ABS survey design and highlight key areas of continued research.