Methodological Issues in the Conversion from RDD to Address-Based Sampling

• Activity Number: 512
• Type: Topic Contributed
• Date/Time: Wednesday, August 5, 2009 : 2:00 PM to 3:50 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Michael Link*+ and Gail Daily and Charles Shuttles
• Companies: The Nielsen Company and The Nielsen Company and The Nielsen Company
• Keywords: Address based sampling ; RDD ; sampling ; cell phones ; survey methodology

Abstract

Address-based sampling (ABS), the use of a comprehensive address database for sampling residences, is viewed as a viable alternative to random digit dialing. In November 2008, the Nielsen TV Ratings Diary moved from a landline telephone to an ABS frame, becoming the first major survey to make this important transition. The move was designed to solve three critical problems: (1) increasing noncoverage bias due to exclusion of cell phone only households and unlisted landline numbers in zero listed telephone banks, (2) number portability, and (3) declining representation of key demographic groups, particularly younger adults. The pioneering ABS design used a multi mode data collection approach, including Web, mail, and telephone recruitment tools. More than 500,000 addresses were sampled for the November 2008 implementation. We assess the success of this transition.