IASS Webinar 50: Multiple frame methods and designs for combining data sources
| Date | 26 Mar 2025 |
| Time | 13:00 GMT+01:00 - 14:30 GMT+01:00 |
| Level of instruction | Intermediate |
| Instructor |
Sharon Lohr
|
| Registration fee | |
Multiple frame surveys, in which samples are selected independently from overlapping sampling frames, can improve population coverage, increase sample sizes for subpopulations of interest, and reduce costs. I review classical multiple frame survey theory for probability samples and extensions to non-probability samples. Assumptions for estimators from multiple frame surveys to be unbiased with the claimed variance are strong, and when the assumptions are not met estimators from a multiple frame survey may have higher mean squared error than estimators from a high-quality single frame survey. I look at multipurpose designs that can detect violations of assumptions in addition to providing estimates of population characteristics.
Instructors
About the instructor
Sharon Lohr has published widely about survey sampling, design of experiments, and statistical methods for education, public policy, law, and crime. She is the author of numerous articles in statistics journals and of the books Sampling: Design and Analysis, now in its third edition, and Measuring Crime: Behind the Statistics. Formerly Dean’s Distinguished Professor of Statistics at Arizona State University and a Vice President at Westat, she is now Professor Emerita at Arizona State University and a statistical consultant and writer.
Sharon is a Fellow of the American Statistical Association and an Elected Member of the International Statistical Institute, and currently serves as a member of the Committee on National Statistics of the U.S. National Academies of Sciences, Engineering, and Medicine. She was the inaugural recipient of the Gertrude M. Cox award for contributions to statistical practice, and has been honoured by being selected to give the Hansen, Deming, and Waksberg Lectures.