Mixed-effects Models Using R

By Douglas Bates, University of Wisconsin

Abstract:

Mixed-effects models are used in the analysis of data that have multiple levels of variation in the response. Common examples of such data are longitudinal data, where there are multiple measurements gathered over time on a set of subjects or other experimental units, and organizational data, where the subjects are grouped into one or more levels of organizational units, such as squad, platoon and company. Linear mixed models, which are also called variance components models, or random slopes models or panel-data models, have been used extensively in the analysis of such data and recently methods have been developed for extensions such as generalized linear mixed models and nonlinear mixed models.

In an STTR project for the U.S. Army Medical Research Activity we have developed new computational methods for such models and provided an implementation in R (www.r-project.org), a freely-available environment for statistical computing and graphics. In this tutorial we will introduce the data handling and graphics capabilities of R and describe the theory and practice of the analysis of longitudinal and organizational data using R.

The short course is a free service offered to conference registrants. No additional fees are required beyond conference registration.

 

Return to ACAS Home Page