An Evaluation of Respondent Selection Methods for Household Mail Surveys

• Activity Number: 365
• Type: Contributed
• Date/Time: Wednesday, August 10, 2005 : 8:30 AM to 10:20 AM
• Sponsor: Section on Survey Research Methods
• Author(s): Michael Battaglia*+ and Michael W. Link and Martin R. Frankel and Pamela Giambo and Ali H. Mokdad
• Companies: Abt Associates, Inc. and Centers for Disease Control and Prevention and Abt Associates, Inc. and Abt Associates, Inc. and Centers for Disease Control and Prevention
• Keywords: Respondent selection ; Mail surveys ; Random-digit-dialing surveys ; Response rates

Abstract

Few empirical studies have assessed the effectiveness of techniques for random selection of respondents within households in mail surveys of the general public. Much of the literature in this area focuses on list sample mail surveys, where a respondent's identity often is known ahead of time. Less is known about respondent selection in mail surveys where household composition is largely unknown. As part of a larger assessment of alternatives to random-digit dialed (RDD) telephone surveys, a six-state pilot study was conducted in conjunction with the Behavioral Risk Factor Surveillance System (BRFSS), testing the utility of a mail survey using a random sample of residential addresses. Three alternative respondent selection techniques were tested: 1) the next birthday method, 2) collection of data from all eligible adults in the household, and 3) a nonprobability approach in which household members selected one adult to complete the questionnaire. A telephone followup was conducted with a subsample of households to assess compliance with respondent selection rules. The analysis compares response rates, respondent demographics, and compliance assessment across the three groups.