This course is designed as an applied introduction to the use of the Stata software for estimating structural equation models. Structural equation modelling (SEM) is used widely by researchers in a diverse array of fields to find and test complex relationships amongst observed (measured) variables and latent (unobserved) variables and amongst the latent variables themselves. SEM subsumes other analytical techniques already performed in Stata such as regression, path analysis, factor analysis, and canonical correlation. This course is designed as an applied course using Stata software to run structural equation models. Detailed notes with worked examples and references will be provided as a basis for both the lecture and hands-on computing aspect of the course. This course primarily focusses on the application of SEM rather than the mathematics behind SEM and is broken up into Six Parts.
Part I: Fundamentals of Stata SEM: Introduction to data inspection, types of variables, data editing, exploratory analysis, and identification of potential problems with data.
Part II: Fundamentals of SEM. Extension of factor analysis and regression analysis using SEM; the advantages of SEM over conventional analytical techniques; the fundamentals underlying SEM; model conceptualisation, path diagrams, model specification and the Stata notation and diagrams.
Part III: Basic Stata SEM Models: Fitting different types of structural equation models in Stata (single-factor models, multiple factor models, confirmatory factor analysis (CFA) models and path analysis); and fitting SEM models using summary statistics data.
Part IV: Stata postestimation tests and predictions: Redisplaying results and obtaining standardized results, obtaining goodness of fit statistics, and performing hypothesis tests, including modification indices tests for omitted paths, relaxing constraints and for model simplification; and obtaining predicted values, including predicted factor scores.
Part V: Common problems in SEM. Problem data and difficult models including topics such as missing data, small samples, non-normal data, constraining parameters, non-positive definite matrices, negative error variances, unidentified and inadmissible models, modelling with and without reliability.
Dr Joanna Dipnall is an applied statistician with interests in the advanced statistical methods, including machine learning and deep learning techniques. She completed her Honours in Econometrics with Monash University and her PhD with IMPACT SRC, School of Medicine, Deakin University. Joanna works extensively with registry and linked medical data and collaborates extensively with the Faculty of IT at Monash to supervise Masters and PhD students to integrate artificial intelligence within health research. Joanna teaches within the Monash Biostatistics Unit and is the Unit Co-coordinator for the Monash Masters of Health Data Analytics course. Joanna has taught advanced statistical methods for many years at universities and for ACSPRI.
This course assumes that participants have (1) Familiarity with the Stata command language (2) sufficient understanding of statistics to be able to comprehend the material covered in the course outline, such as a basic grounding in multiple regression and factor analysis (3) some experience in using Microsoft Word and Excel or their equivalent (4) experience using a text editor such as Notepad, UltraEdit. While not a pre-requisite, participants with no previous exposure to structural equation modelling are strongly encouraged to first complete the course 'Fundamentals of Structural Equation Modelling'.
Course notes will be supplied. No specific references are suggested but participants might find the following text useful.
Bollen KA. 1989. Structural equations with latent variables. New York: John Wiley.
Schumacker RE & Lomax RG. 2010. A Beginner’s Guide to Structural Equation Modeling. Mahwah, New Jersey: Lawrence Erlbaum.