Multilevel models are known by many synonyms (e.g., hierarchical linear models, general linear mixed models). The defining feature of these models is their capacity to provide quantification and prediction of random variance due to multiple sampling dimensions (across occasions, persons, or groups). Multilevel models offer many advantages for analyzing “imperfect” longitudinal data, such as flexible strategies for modeling change and i ndividual differences in change, examining time-invariant or time-varying predictor effects, and the use of all available complete observations, even if unbalanced across subjects. This four-day workshop will serve as an applied introduction to multilevel models for longitudinal data. Beginning from a base of general linear models, we will examine how to use fixed and random effects to describe the effects of time within studies of individual change (i.e., growth curve models) or individual fluctuation (i.e., daily diary designs). We will also examine how individual differences can be predicted by time-invariant predictors.
The primary software package utilized for instruction will be SAS, but reference examples using SPSS, STATA, and Mplus are also available. Participants should be familiar with the general linear model (e.g., analysis of variance and regression), but no prior experience with multilevel models or knowledge of advanced mathematics (e.g., matrix algebra) is assumed. However, it will be helpful to read chapters 1-7 of the course textbook prior to attending the workshop.
The in-person portion of this workshop will focus on concepts, analyses, and interpretation of output from computer programs used in longitudinal analyses. Each day has two parts: a morning session from 8:30am to 11:00am and an afternoon session from 2:30pm to 5:00pm. All lecture materials will be available both online and in hard-copy format.
From 11:00am to 2:30pm is open time. Go to lunch, visit the beach, lounge by the pool, or sit and work through the examples presented during the day in the course classroom. Individual consultation sessions with the instructor are available during this time during each day of the workshop.
Additional Course Materials
Augmenting the in-class portion of the workshop are supplemental materials available online that support the primary textbook. The textbook for the course will be Longitudinal Analysis: Modeling Within-Person Fluctuation and Change, which participants should purchase separately. These materials include data, syntax, and output for all textbook examples in SAS, SPSS, STATA, and Mplus.
Finally, all workshop lectures are recorded and videos of the course will be available online shortly after the workshop is completed. Online content is available to registered participants in advance of the workshop and continues to be available to participants up to one year after the workshop.
Anticipated Schedule of Topics
- Day 1:
- Introduction to Multilevel Longitudinal Models (chapter 1)
- Review of General Linear Models and Repeated Measures ANOVA (chapter 3)
- Describing Within-Person Fluctuation over Time via Alternative Covariance Structure Models (chapter 4)
- Day 2:
- Describing Within-Person Change over Time (chapter 5)
- Polynomial and Piecewise Random Effects Models for Change (chapter 6)
- Day 3:
- Time-Invariant Predictors of Fluctuation (chapter 7)
- Time-Invariant Predictors of Change (chapter 7)
- Day 4:
- Multilevel Longitudinal Models for Non-Normal Outcomes
- Individual Work Sessions
About the Instructor
Lesa Hoffman has a joint appointment as an Associate Professor of Quantitative Methods in the Child Language Doctoral Program and as an Associate Scientist and Director of Quantitative Development in the Research Design and Analysis Unit, both within the Schiefelbusch Institute for Life Span Studies at University of Kansas. Dr. Hoffman received her Ph.D. in Cognitive and Quantitative Psychology at the University of Kansas in 2003. She joined the faculty of the University of Kansas in August of 2014, having previously served on the faculty for eight years in the Department of Psychology at the University of Nebraska-Lincoln. Dr. Hoffman’s program of research seeks to empirically examine and to thoughtfully disseminate how developments in quantitative psychology can best be utilized to advance empirical work in psychology, human development, and other social sciences. Dr. Hoffman’s research program has been funded by the National Institute of Health and the National Science Foundation, and has been published in journals such as Multivariate Behavioral Research, Psychology and Aging, and the Journal of Speech-Language-Hearing Research. Dr. Hoffman is the sole author of the course textbook, Longitudinal Analysis: Modeling Within-Person Fluctuation and Change. For more information, please visit her website: http://lesahoffman.com.
Workshop Location: Omni Hilton Head Oceanfront Resort
The Omni Hilton Head Oceanfront Resort boasts an elegant contemporary design throughout the property’s spacious guest rooms, public areas, meeting rooms and the unmatched white sand beaches joining the cool blue Atlantic Ocean. Tropical gardens run throughout the Hilton Head beach resort, providing a serene and tranquil place where you can choose from a variety of activities during your stay.
For hotel information and registration, please visit https://www.omnihotels.com/hotels/hilton-head.
Discount for Multiple Workshop Enrollment
Enrolling in more than one workshop? Enrolling as a group? We now offer a 10% discount on the total price of enrollment for in-person workshops.
Enter twoworkshops in the coupon code section of the check-out page to receive the discount.
Please contact Lesa Hoffman at firstname.lastname@example.org.