Encouraging Record Use in an Online Survey

• Activity Number: 476
• Type: Contributed
• Date/Time: Wednesday, August 1, 2012 : 8:30 AM to 10:20 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Mick Couper*+ and Mary Beth Ofstedal and Sunghee Lee
• Companies: University of Michigan and University of Michigan and University of Michigan
• Keywords: online survey ; income and assets ; measurement error ; record use

Abstract

Like many surveys, the Health and Retirement Study (HRS) encourages respondents to use records to answer questions on financial data such as assets, mortgage payments, and the like. The question is how respondents react to that encouragement, and what effect it has on the quality of data they provide. In the 2009 HRS Internet survey, a random half was prompted to consult records at the start of the relevant section. The encouragement increased reported use of records from 38.7% to 46.7%. In the 2011 survey, the encouragement was moved to the advance letter for a random subset. This gives respondents sufficient time to gather the relevant documents, but may also encourage nonresponse. The prompt increased reported record use from 45.6% to 55.4%; however, the group that was prompted has a lower response rate to the survey (75.5%) than those not prompted (78.3%). In this paper we examine the effects of such encouragement on compliance (self-report of record-use) in the two surveys, and on the quality of the data reported (item missing data, rounding, etc.).