Predictors of Retention and Re-Engagement in the Health and Retirement Study

• Activity Number: 252
• Type: Contributed
• Date/Time: Monday, July 30, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Mary Beth Ofstedal*+ and Mick Couper and Kuang-Tsung Chen
• Companies: University of Michigan and University of Michigan and University of Michigan
• Keywords: panel surveys ; panel attrition and retention ; health and retirement study

Abstract

Retention of respondents in panel studies has become increasingly difficult and costly, yet is critical for the study's long-term success. Understanding the factors that are associated with panel retention is important for targeting field interventions to improve retention and reduce costs and for assessing and accounting for attrition bias. We use data from the Health and Retirement Study (HRS), a panel survey of older Americans that began in 1992, to examine factors that are associated with continuous participation and with re-engagement of participants who have missed one or more of the survey waves. We pool the data across waves and estimate two logistic regression models: one predicting current wave response among previous wave non-respondents (the re-engagement model) and a second predicting current wave response among previous wave respondents (the retention model). Predictors include study design factors; prior wave effort, burden and response outcomes; respondent and interviewer characteristics. Findings will help identify areas for improvement in the administration of the HRS and may have implications for the design and administration of panel surveys more generally.