Effects of a Monetary Incentive in a National Randomized Survey of Medicare Beneficiaries

• Activity Number: 314
• Type: Contributed
• Date/Time: Wednesday, August 14, 2002 : 10:30 AM to 12:20 PM
• Sponsor: Section on Survey Research Methods*
• Author(s): Judith Lynch*+ and Lauren McCormack and Steven Garfinkel and Scott Scheffler and Sherry Terrell
• Affiliation(s): RTI International and RTI International and American Institute of Research and RTI International and Centers for Medicare & Medicaid Services
• Keywords: Monetary Incentives ; increasing response rates ; Medicare Beneficiaries ; Mixed-Mode Surveys ; Surveys

Abstract

This paper describes results of an experiment to test the effects of using a $10 incentive on survey response rates and costs in a national randomized survey of 6,154 Medicare beneficiaries. We found that the incentive significantly increased beneficiary participation in this government-sponsored mixed mode (mail with telephone follow-up) survey. The overall completion rate was 76.4 percent--80.3 percent for the incentive group and 73 percent for the no incentive group, a statistically significant difference of 7.3 percentage points. Using logistic regression we modeled beneficiaries' propensity to respond to the survey controlling for other factors available on the sampling frame and found that the incentive increased the odds of responding by 39 percent. We observed that the difference in response rates between the incentive and no incentive groups decreased after each major survey event, suggesting that an incentive has the greatest impact earliest in the data collection period. By shifting responses to earlier stages of the survey, the incentive reduced the survey cost by more than $4, but not enough to fully offset the cost of the $10 incentive.