The presentation discusses some of the issues in transitioning an in-person survey to a web/paper survey. It reports on two recent studies that tested a range of design options including incentives, mail contact protocols and questionnaire length. One survey is cross-sectional and the other is a longitudinal survey in its fifth wave. The issues for these two surveys, which are quite different, will be reviewed and contrasted.

Dr. Paul P. Biemer is Distinguished Fellow of Statistics at RTI International and Associate Director of Survey Research and Development for the Odum Institute at the University of North Carolina in Chapel Hill.  A PhD in Statistics, Biemer has 35+ years of experience in survey methodology, complex survey design, and data analysis. He has published over 100 articles and book chapters and has co-authored or co-edited seven books. His current research examines the relationships between survey design and survey error, particularly nonresponse and measurement errors. He is a co-developer of a number of innovations in survey methodology including:

  • Adaptive Total Design (ATD)
  • The ASPIRE system for survey monitoring and quality improvement for official statistics
  • Audio Computer Assisted Self-Interviewing (ACASI) and Computer Audio Recorded Interviewing (CARI).

He pioneered the field of latent class analysis and mixture modeling for survey error evaluation.  At UNC, he established the Certificate Program in Survey Methodology which he directs.  He is a Fellow of the ASA and the AAAS and an Elected Member of the ISI. He holds a number of awards for his contributions to the fields of statistics and survey methodology, including the W.S. Connor Award, H.O. Hartley Award, the Morris Hansen Award and the Roger Herriot Award.