John Finamore, SRMS Publications Officer
The ASA SRMS webinar series will continue for a third year with four webinars organized by section education officer, Marilyn Seastrom. The first, “Practical Tools for Nonresponse Bias Analysis” by Kristen Olson of the University of Nebraska-Lincoln and Jill Montaquila of Westat, will take place April 24.
This webinar will provide an overview of methods that may be used to help address the Office of Management and Budget guidelines for conducting nonresponse bias studies when response rates in surveys are less than 80% or there is reason to suspect estimates are biased due to nonresponse. Practical tools will be described and examples used to illustrate the methods. The advantages and disadvantages of the methods will be presented, and the value of having multiple approaches will be highlighted. The need to devise strategies for nonresponse and its analysis in the planning stage, prior to completing the survey, will be emphasized.
Olson is an assistant professor of survey research and methodology and sociology. She has been at the University of Nebraska-Lincoln since 2007, and her areas of research include nonresponse bias and nonresponse adjustments, the relationship between nonresponse and measurement errors, and interviewer effects. Olson’s research has appeared in Public Opinion Quarterly, the Journal of the Royal Statistical Society Series A, Sociological Methods and Research, Field Methods, Social Science Research, and Survey Research Methods. She is currently serving as conference chair for MAPOR and has taught short courses on nonresponse bias studies for AAPOR, DC-AAPOR, SAPOR, and the Joint Program in Survey Methodology (JPSM). Olson is also editor of the Research Synthesis section of Public Opinion Quarterly. She earned her BA in mathematical methods in the social sciences and sociology from Northwestern University, her MS in survey methodology from the JPSM at the University of Maryland, and her PhD in survey methodology from the University of Michigan.
Montaquila is an associate director of the statistical staff and senior statistician at Westat and a research associate professor in the JPSM at the University of Maryland. She is a Fellow of the American Statistical Association. Her research interests include various methods for evaluation of nonresponse bias, random digit dialing survey methodology, and address-based sampling. Montaquila has given short courses on approaches for nonresponse bias analysis for DC-AAPOR, SAPOR, and JPSM. She has served as president of the Washington Statistical Society and is chair-elect of SRMS.
Roundtables and CE Course for JSM 2012
The ASA SRMS also will sponsor two roundtables and one Continuing Education course at JSM 2012 in San Diego, California.
Lars Lyberg of the University of Stockholm, Sweden, will lead the roundtable discussion “Total Survey Error in Practice.” We are all aware that only a fraction of error sources are taken into account when margins of survey error are presented. Should we approach this problem by attempting to estimate mean squared error components, improving survey processes so we gradually approach more ideal bias-free processes, or both? How should we handle the survey design problem when we have to deal with a multitude of nonsampling error sources, some of which defy expression? How should issues regarding survey quality that go beyond textbook sampling formulas be communicated to and discussed with users and clients? This roundtable will allow the opportunity to discuss these and other related questions.
Joe Sakshaug of the Institute for Employment Research, Nuremberg, will lead the roundtable “Challenges with Linking Survey and Administrative Data Set.” Linking survey and administrative data is an attractive option for substantive researchers interested in studying complex phenomenon and survey methodologists who use such data to study data quality issues and reduce data-collection costs. However, there are several barriers to linkage that can potentially reduce the quality of the linked data, including the effects of non-consent bias and matching errors. This roundtable aims to promote and discuss the joint use of administrative and survey data with a focus on linkage issues, including biases in linked data sets, methods of asking for linkage consent, and direct and indirect matching methods. Results from observational and experimental linkage studies will be provided using multiple data sets from the United States and Germany.
Finally, Frauke Kreuter of the University of Michigan will teach the Continuing Education course “Paradata in Survey Research.” Survey data are increasingly collected through computer-assisted data-collection modes. As a result, a new class of data—called paradata—is now available to survey methodologists. While the type of available paradata varies by mode, all share one feature: They are a byproduct of the data-collection process capturing information about that process. This course covers the great potential of paradata for social survey research. It will provide an introduction and overview of methodological issues involved in the collection and analysis of paradata. Research examples will be discussed, including the use of paradata to monitor fieldwork activity, guide intervention decisions (e.g., through responsive design), and addressing various total survey error components. Case studies will draw attention to the challenges in automated data capturing and modeling of the complex structure of paradata.
Leave a Reply