When Less is More: Are Reluctant Respondents Poor Reporters?

• Activity Number: 165
• Type: Contributed
• Date/Time: Monday, August 9, 2004 : 2:00 PM to 3:50 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Ting Yan*+ and Roger Tourangeau and Zac Arens
• Companies: JPSM, University of Maryland/Abt Associates and University of Maryland and University of Maryland
• Keywords: response rate ; nonresponse error ; measurement error ; response bias

Abstract

Survey research has been witnessing declining response rates across a wide range of surveys. As a result, extensive resources have been invested in boosting response rates under the assumption that a higher response rate will lead to lower survey error. However, the extra attempts to reduce nonresponse are costly. Given limited budgets, survey researchers might be more confident about allocating a larger share of survey budgets to decreasing nonresponse if they could be assured that an improved response rate does in fact lower survey error. In fact, the empirical evidence in this regard is mixed. Some studies show respondents interviewed on early calls differ from those interviewed on later calls. However, other recent studies seem to demonstrate that large differences in response rate have only minor effects on cross-sectional analyses. This study, analyzing a national RDD survey data that measures the public knowledge of and attitudes towards science, replicates the work of Curtin et al (2000) by looking at the impact of response rates on key survey variables from another topic area. In addition, this study extends that work by exploring the relationship between the nonresponse errors and the measurement errors in a national survey. It will focus on answering the question whether reluctant respondents have more or less measurement error.