Assessing the Effectiveness of Using Panel Respondent Mode Preferences

• Activity Number: 12
• Type: Invited
• Date/Time: Sunday, July 29, 2007 : 2:00 PM to 3:50 PM
• Sponsor: Social Statistics Section
• Author(s): Thomas B. Hoffer*+ and Karen Grigorian and Ron Fecso
• Companies: NORC at the University of Chicago and NORC at the University of Chicago and National Science Foundation
• Keywords: respondent mode preferences ; multimode surveys ; response rates ; data quality

Abstract

This paper uses data from the Survey of Doctorate Recipients (SDR), an ongoing panel survey of doctoral scientists and engineers, to assess whether and how mode preference data should be used to improve panel survey participation and data quality. In the 2003 SDR, respondents were asked to indicate whether they would prefer to complete via paper, Web, or CATI in the next (2006) round. Most indicated a preference for the Web mode but honoring that preference and the supporting system of e-mail prompts runs a risk of less-cooperative subjects more easily deflecting prompts. Such cases may then wind up in the relatively expensive CATI mode, and defeat the purpose of honoring preferences. We analyze response rate and quality indicators from the 2001, 2003, and 2006 surveys to assess the extent to which honoring respondent mode preferences had a positive or negative effect on survey outcomes.