Session: Structural Equation Modeling (Research that Promotes Sustainability and (re)Builds Strengths (January 15 - 18, 2009))

Structural equation modeling (SEM) appears to be the fastest growing multivariate statistical method in the educational, social and behavioral sciences (Hershberger, 2003). SEM is also known as covariance structural analysis, covariance modeling, and causal modeling. SEM enables one to test complicated theoretical models that involve not only direct effects but also indirect effects. SEM also permits the inclusion of both observed and unobserved variables (e.g., factors). The workshop will introduce structural equation modeling and reasons of its popularity. The workshop will help researchers 1) identify research questions in a SEM and 2) how to interpret a SEM. SEM is a family of techniques instead of a single method. Some of the different SEM techniques that will be discussed include: • Path analysis • Confirmatory factor analysis • SEM with latent variables • Mediation • Multi-sample SEM • Mean and covariance structural analysis • Longitudinal Models (Latent growth curve models) The workshop will introduce various techniques of SEM and demonstration using computer software.