Optimizing Call Strategies in RDD: Differential Nonresponse Bias and Costs in REACH 2010

• Activity Number: 278
• Type: Topic Contributed
• Date/Time: Tuesday, August 5, 2003 : 2:00 PM to 3:50 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Whitney Murphy*+ and Colm A. O'Muircheartaigh and Rachel M. Harter and Carol Emmons
• Companies: NORC at the University of Chicago and University of Chicago and NORC at the University of Chicago and NORC at the University of Chicago
• Keywords: RDD ; nonresponse bias ; call strategies

Abstract

For the REACH 2010 project, NORC investigated the relative efficiency of different call strategies by defining subgroups of cases based on their intermediate call status (promising and unpromising). Using the relative marginal productivity of additional calls to the two types of cases, we devised a call strategy for the survey. Calls per completes were determined at different stages, as were marginal and average cost, and cut-offs were established based on the results. Implications for response rates and sample sizes were determined and evaluated. At the time of the above analyses, interview data were not available. We examine the impact the imposed cut-offs had on the resulting interview data. This paper thoroughly compares the interview data for promising and unpromising cases to determine whether any differential nonresponse bias was introduced into the data by following only promising screener cases past the minimum call cut-off. Using costs and mean square error, we evaluate last year's decisions in light of the data.