Boris O. Sokolov
(National Research University "Higher School of Economics")

Introduction to SEM and SEM Software

Abstract: The course is intended to give an introduction to general principles and techniques of structural equation modeling (SEM) and their implementations in two popular SEM software packages, an R package lavaan (Rosseel 2012, 2013) and MPLUS (Muthen & Muthen 2012). The topics covered by the course are confirmatory factor analysis (CFA), measurement invariance, path models, structural equation models. In addition, practical issues of estimation, visualization and presentation of various types of SEM models are discussed. Prior knowledge of statistical programming language R is desirable, but is not a necessary requirement.

Kenneth A. Bollen
(University of North Carolina at Chapel Hill, USA)

Latent Curve Models: A Structural Equation Perspective

Kenneth Bollen is Henry Rudolph Immerwahr Distinguished Professor of Psychology and Neuroscience & Sociology, University of North Carolina at Chapel Hill, USA.

Abstract: Latent Curve Models (LCMs) are an increasingly popular approach to analyze longitudinal data. Though the models go by many names (e.g., growth curve modeling, latent growth models, latent trajectory models), they all refer to statistical models for longitudinal data that allow each individual in the sample to have distinct over-time trajectories of change. These patterns of change are summarized in relatively few parameters. The parameters in turn are modeled as functions of other variables.

With the growing availability of longitudinal or panel data, social science applications of and interests in LCM have increased. The formulations and estimation of these models have proceeded in several ways. In the workshop the LCMs will be analyzed from the perspective of structural equation modeling with latent variables. Although the lecturer will present simple regression based procedures that are helpful in the early stages of LCM, most of the discussion will make use of Structural Equation Models (SEMs).

The major topics of the course are: an overview of trajectory models & a review of SEMs, unconditional latent curve models (LCMs), nonlinear LCMs, conditional LCMs, the analysis of groups, multivariate LCMs, and latent variable LCMs.