Applying Structural Equation Modelling in Education Research

Structural equation modelling (SEM) is a set of statistical techniques used to simultaneously examine the structure of interrelationships among observed and latent variables represented by a series of equations, while accounting for measurement error. This paper presents a non-technical introduction to SEM and provides an illustration on how the fit between an SEM model (based on theory) and data – in the context of higher education – can be assessed through a measurement and structural model, as well as validated with a holdout sample to evaluate generalisability. SEM can serve as a very useful and powerful tool for the analysis of education data to support teaching and learning.