PSY6551

Structural equation models are a class of statistical techniques that incorporate regression analysis, path analysis, confirmatory factor analysis, & full-scale models integrating both measurement & structural components. These techniques are useful for both experimental & non-experimental data; for cross-sectional data sets; for group comparisons; & for longitudinal data sets, including the modeling of growth curves. In general, it is a large sample size approach, though recent advances have been made for SEM to become more applicable in small samples through Bayesian & other alternate estimation methods. This course picks up where an introductory course to structural equation modeling leaves off. It focuses on how to deal with funny dependent variables, complex measurement issues, interactions, time models, Monte Carlo Simulations, power, & integrations of SEM & multilevel modeling.

006663

Fall and Spring