Full day short course (Monday, July 11; 10:00AM-5:30PM)
Structural equation modeling (SEM) is a very general statistical technique widely used technique in social and behavioral sciences. The frequency properties of popular SEMs are often the focus of empirical investigations using simulated data, and new SEMs are often presented with a cursory Monte Carlo simulation study to verify their practical applicability.
This course explains how to conduct Monte Carlo simulations in R. Beginning with fundamental data generation from a population model that fits within the SEM framework, special attention is paid to functions provided by lavaan to facilitate simulation, and to the simsem package that includes many special features for generating complex data (e.g., fixed exogenous covariates, missing data mechanisms, random parameters) and analyzing simulation results. The simsem package can utilize both lavaan and OpenMx for data generation or analysis, as well as custom functions provided by the user (e.g., using Mplus via the MplusAutomation package). Additional topics include generating discrete and multilevel SEM data.
R syntax is provided for all path- and factor-model examples. Other topics are included among the many vignettes available at http://simsem.org/
All instruction and example syntax will utilize the R software.
About the Instructor
Terrence D. Jorgensen
Terrence D. Jorgensen, PhD, teaches SEM as an assistant professor of methods and statistics within the Department of Child Development and Education at the University of Amsterdam. He maintains the R packages semTools and simsem, and contributes to the lavaan and blavaan packages.