Latent Growth Models
Latent growth models (LGMs) allow for an analysis of dynamic processes that unfold over time. Rooted in the structural equation modeling framework, LGMs can be used to characterize trajectories of social processes across cases, examine predictors of the trajectories, and connect different trajectories to outcomes. This course introduces LGMs with a focus on the intuition underlying the models and applications of the models. Examples, exercises, and code will be provided in R and Stata. Knowledge of R or Stata and linear regression is recommended.
Day 1: Longitudinal data structures and unconditional latent growth models
Day 2: Nonlinear and conditional growth models
Day 3: Effect decomposition and outcomes of growth processes
Dr. Shawn Bauldry is an Associate Professor of Sociology and a Core Faculty in Advanced Methods at Purdue University. He received a Ph.D. in Sociology and an M.S. in Statistics from the University of North Carolina at Chapel Hill. He has published papers on a range of topics related to Structural Equation Models and has co-authored a book on Confirmatory Factor Analysis in the Sage QASS series. He has taught graduate-level statistics courses for over 10 years and has developed numerous workshops and courses on SEMs, causal inference, and multilevel and longitudinal analyses. He is a recipient of the Purdue Sociology Department Graduate Mentoring Award.
On-demand. Lecture slides, lab code, and recordings are available to view at your convenience.