University of Michigan
Survey Research Center
About the Webinar
Prevention scientists have been investigating precision medicine, which builds upon heterogeneity in response to treatments. In parallel, survey methodologists have begun to recognize that samples are heterogeneous with respect to response to recruitment protocols. This heterogeneity has been acknowledged by at least two new classes of survey design—responsive and adaptive. Responsive design is an approach to dealing with uncertainty about key survey design parameters. Responsive survey designs identify potential risks related to costs or errors, develop indicators for tracking these risks, and then plan design changes for controlling these costs or errors. These design options are triggered if the indicators cross prespecified thresholds. Adaptive survey design, on the other hand, starts from knowledge about likely outcomes across heterogeneous subgroups in the sample, and implements different survey designs for different subgroups. This presentation starts from a definition of the basic principles of responsive and adaptive designs and then provides concrete examples of the implementation of these designs. These examples are drawn from a variety of settings, including face-to-face, telephone, and mixed-mode surveys.
About James Wagner
Dr. Wagner is a Research Associate Professor at University of Michigan’s Survey Research Center. His research is in the area of survey non-response, including indicators for the risk of non-response bias and adaptive/responsive survey design. He is the co-author of a new book (2017) entitled "Adaptive Survey Design." He has published articles in a variety of journals including Public Opinion Quarterly, Statistics in Medicine, Journal of the Royal Statistical Society, Journal of Official Statistics, and others. Dr. Wagner is also the Associate Director for the Michigan Program in Survey Methodology. He teaches courses on sampling, statistics, and research design.