21 Sep Structural Equation Modeling – ONLINE
1:00 pm - 2:30 pm
statistics seminar | level: intermediate |
for questions related to this event, contact firstname.lastname@example.org
affiliation: Ghent University
Certain factors, measures or constructs (hereafter referred to as latent) are difficult if not impossible to directly measure yet are worthy of investigation. For example, in the social sciences, constructs such as intelligence, ability, trust, motivation etc. cannot be directly observed because there are no operational methods for doing so. However, objective observed measures exists that are indicators of these latent constructs even though they are associated with measurement error.
Structural equation modeling (SEM) provides a flexible framework that offers the opportunity to construct latent variables, estimate relationships among these variables, while ensuring that these relationships are free of the measurement error associated with the observed indicator variables.
In this seminar, we will briefly touch on five steps that characterize most SEM applications: Model formulation, model identification, model estimation, model evaluation and model modification with some examples using the lavaan package in R.