Digital Module 16: Longitudinal Data Analysis
Recorded On: 01/21/2021
In this digital ITEMS module, Dr. Jeffrey Harring and Ms. Tessa Johnson introduce the linear mixed effects (LME) model as a flexible general framework for simultaneously modeling continuous repeated measures data with a scientifically-defensible function that adequately summarizes both individual change as well as the average response. The module begins with a non-technical overview of longitudinal data analyses drawing distinctions with cross-sectional analyses in terms of research questions to be addressed. Nuances of longitudinal designs, timing of measurements, and the real possibility of missing data are then discussed. The three interconnected components of the LME model: (1) a model for individual and mean response profiles, (2) a model to characterize the covariation among the time-specific residuals, and (3) a set of models that summarize the extent that individual coefficients vary, are discussed in the context of the set of activities comprising an analysis. Finally, they demonstrate how to estimate the linear mixed effects model within an open-source environment (R). The digital module contains sample R code, diagnostic quiz questions, hands-on activities in R, curated resources, and a glossary.
Keywords: fixed effect, linear mixed effects models, longitudinal data analysis, multilevel models, population-average, random effect, regression, subject-specific, trajectory
Jeffrey R. Harring
Jeff is a Professor in the Measurement, Statistics and Evaluation Program within the Department of Human Development and Quantitative Methodology at the University of Maryland, College Park. There, he teaches introductory and intermediate graduate level statistics courses and advanced quantitative methods seminars in longitudinal data analysis, mixture modeling, simulation design and statistical computing. Jeff has taught several multi-day workshops on the application of longitudinal methods using R and SAS statistical software most recently at the National Center of Educational Statistics (NCES) in Washington D.C. Prior to joining the program faculty in the fall of 2006, Jeff received an M.S. degree in Statistics and completed his Ph.D. in the Quantitative Methods in Education from the University of Minnesota. Before that, Jeff taught high school mathematics for 12 years. He has published nearly 100 articles and book chapters, co-edited three volumes and co-authored a book. His research focuses on linear and nonlinear models for repeated measures data, structural equation models, finite mixtures of both linear and nonlinear growth models and extensions of these methods to multilevel data structures.
Contact Jeff via firstname.lastname@example.org
Tessa is a Ph.D. candidate in the Measurement, Statistics and Evaluation Program within the Department of Human Development and Quantitative Methodology at the University of Maryland, College Park. She received her Master of Science in Educational Research from Georgia State University. Tessa currently serves as a project coordinator for the Synthetic Data Project (SDP) of the Maryland Longitudinal Data System Center, an Institute of Education Sciences (IES) funded project aimed at assessing the feasibility of and implementing a system for synthesizing statewide longitudinal data in order to increase data access for researchers and policy analysts while minimizing risk of data disclosure. For the SDP, Tessa conducts research on the feasibility of reproducing nested data structures in longitudinal synthetization models. Outside her work with the SDP, Tessa’s research has centered around creating and improving statistical methods for analyzing complex data structures in longitudinal contexts. This includes modeling time as an outcome in latent growth models, accounting for similarities among schools when modeling student mobility in longitudinal studies, and exploring the development of ensembles of social networks in the classroom over time.
Contact Tessa via email@example.com