Prof. dr. Donald Hedeker

University of Chicago

Two-stage mixed-effects location scale (MELS) models for intensive longitudinal data

Abstract: Intensive longitudinal data are increasingly encountered in many research areas.  For example, ecological momentary assessment (EMA), experience sampling method (ESM), and/or mobile health (mHealth) methods are often used to study subjective experiences within changing environmental contexts. In these studies, up to 30 or 40 observations are usually obtained for each subject over a period of a week or so, allowing one to characterize a subject’s mean and variance and specify models for both. In this presentation, we focus on a smoking study of dual users (i.e., both combustible and electronic cigarette users) using EMA where interest is on characterizing changes in mood variation associated with these nicotine products, and whether subjects’ mood response can predict future nicotine product use.  At the first stage, the MELS model includes random subject effects for the mean (i.e., location), which characterize subjects’ differential mood response to combustible and electronic cigarettes.  A random effect for subjects’ variability (i.e., scale) in mood responses is included to characterize subjects’ mood consistency/erraticism.  These random location and scale effects are used in a second stage regression, both linear and multinomial, model to predict future nicotine product use.   Since the random effects are estimates, repeated draws from the posterior distribution of the random effects for each subject are utilized in the second stage model (i.e., plausible value replications), with results averaged across these repeated draws.  A software program, MixWILD, which facilitates this two stage modeling approach, is described.

Bio: Donald Hedeker is Professor of Biostatistics in the Department of Public Health Sciences at the University of Chicago. His research focuses on the development and application of advanced statistical methods for clustered and longitudinal data, with particular expertise in mixed-effects modeling and intensive longitudinal designs. He is the primary author of several freely available software programs for mixed-effects analysis, and co-author (with Robert Gibbons) of the influential textbook Longitudinal Data Analysis (Wiley, 2006). More recently, his work has advanced methods and software for the analysis of intensive longitudinal data, including data collected through mobile health (mHealth) and ecological momentary assessment (EMA) studies. These developments have provided researchers with powerful tools to analyze data featuring high temporal resolution and within-person dynamics. Professor Hedeker was elected Fellow of the American Statistical Association in 2000 and has served as Associate Editor for Statistics in Medicine since 2006. His contributions have been instrumental in shaping contemporary approaches to longitudinal and multilevel data analysis in the health and behavioral sciences.