Choosing Sample Sizes for Multilevel and Longitudinal Studies Analyzed with Linear Mixed Models
Methods: Mind the Gap
Keith Muller, Ph.D.
University of Florida College of Medicine
Deborah Glueck, Ph.D.
University of Colorado School of Medicine
About the webinar
Planning a reproducible study requires selecting a sample size that is expected to ensure adequate statistical power. Advances in statistical methods and free point-and-click software have made it easier to select a sample size for clustered and longitudinal designs with linear mixed models. The software, a suite of training modules and reference materials are freely available online (www.SampleSizeShop.org). The software interface and training materials are intended for health scientists and have been validated in face-to-face and online courses. Dr. Keith Muller and Dr. Deborah Glueck discuss what the software and training cover, what they don’t, and what they hope to add. A longitudinal randomized trial is used to illustrate the power analysis process.
About Keith Muller
Dr. Muller is a biostatistician specializing in the design and analysis of multilevel and longitudinal research. He is the first author of two books on the theory and practice of linear models. He has co-authored about 150 peer-reviewed publications, about half of which are health research collaborations. Dr. Muller is a Fellow of the American Statistical Association. He received a bachelor’s and master’s degree in psychology from Bradley University, as well as a doctorate in quantitative psychology and a master’s degree in theoretical statistics from the University of North Carolina at Chapel Hill. Postdoctoral employers have included the University of Illinois (1976-1977, Educational Methods), Bell Laboratories (1977-1978, Human Factors), the University of North Carolina (1978-2006, Biostatistics), and the University of Florida (2006-present, health outcomes and biomedical informatics). About half of Dr. Muller’s efforts have been supported as co-investigator or principal investigator of research funded by the National Institutes of Health (NIH). He has also served on standing and ad hoc review committees for the NIH.
About Deborah Glueck
Dr. Glueck is Senior Biostatistician at the Lifecourse Epidemiology of Adiposity & Diabetes Center (LEAD) at the Colorado School of Public Health; and professor of pediatrics at the University of Colorado Denver Medical School. Dr. Glueck holds an AB cum laude from Harvard College in Mathematics, as well as an MA and Ph.D. degrees in biostatistics from the University of North Carolina at Chapel Hill. She completed a postdoctoral fellowship in health services research at Robert Wood Johnson Medical School. Dr. Glueck has focused throughout his career on power and sample size for multilevel and longitudinal studies, as well as the analysis of longitudinal epidemiological cohort studies and randomized controlled clinical trials. Dr. Glueck was multiple principal investigator (MPI) for an R01 for the development of new analytical methods for life course studies, including the implementation and development of new techniques to model developmental trajectories in young people (R01GM121081); was MPI of a G13 grant funded by the National Library of Medicine for a book on power and sample size; and was MPI of an R25 funded by the National Institute of General Medical Sciences for developing a short course and a massive online open course on power and sample size.
Last updated on April 29, 2022