Fundamentals of Structural Equation Modelling

Structural equation modelling—or structural equations with latent variables—is a very general statistical model and widely used method. For example, SEM is used in fundamental disciplines such as the social, economic and psychological sciences, the biological sciences, and applied disciplines such as education, health and marketing.

 

This course provides an overview of the fundamentals of SEM. As well as the statistical theory, an overview of the many applications and capabilities of SEM is given.

 

 
Level 3 - runs over 5 days
Instructor: 

Dr Mark Griffin is the Director of Insight Research Services Associated (www.insightrsa.com), where Insight consists of Insight Business Analytics, and Insight Training and Events. Insight Business Analytics providing training and consulting across the areas of business, statistics and IT. Insight Business Analytics is a Business Partner of Microsoft and a member of the IBM's Partner World and Google's Partner Program. Insight Training and Events teaches 30 qualifications from Certificate I to Graduate Diploma level across the areas of business, information technology, health and social services, and civil and environmental engineering.

 
Mark is also an Industry Fellow with the School of Business, University of Queensland, and has established and written training materials for several of their courses in Business Analytics. Mark serves on the Executive Committee for the Statistical Society of Australia, was the Co-Chair of their National Conference in 2021, and is the Founding Chair of their Section for Business Analytics. Mark is also the Founding Chair of the Business Analytics Special Interest Group within the International Institute of Business Analysis. To date he has presented over 100 two-day and 40 five-day workshops in statistics around Australia

About this course: 

Structural equation modelling—or structural equations with latent variables—is a very general statistical model and widely used method. For example, SEM is used in fundamental disciplines such as the social, economic and psychological sciences, the biological sciences, and applied disciplines such as education, health and marketing.

 

SEM has become popular for several reasons, apart from its generality:

     (i) all SEM models can be represented visually,

     (ii) a standard notation helps researchers to communicate, and

     (iii) several software packages for estimating SEM models are readily available (e.g., AMOS, LISREL, Mplus, R).

 

This course provides an overview of the fundamentals of SEM. As well as the statistical theory, an overview of the many applications and capabilities of SEM is given. The course is not particularly mathematical, but instead places emphasis on the fundamental concepts of SEM and how it is used by applied researchers.

 

General aims of the course are for students to develop a readiness for using SEM software and to develop the requisite knowledge for applying SEM methods and models in an intelligent way. Note that participants may be invited to briefly present their own research on the last day of class. This exercise, along with the formal lecture material, might help participants to chart a direction forward in their study and application of SEM.

 

This course is designed as an introductory course for applied researchers and as such, is suitable for participants who want to develop a fundamental knowledge of SEM techniques. Some participants may want to follow this course by the applied and advanced SEM workshops taught by ACSPRI.

 

Course syllabus: 

Day 1

  • Review of linear regression

 

Day 2

  • Building regression and path models using an SEM framework
  • General stages of SEM modelling - specification, identification, estimation, testing and modification
  • Reporting SEM research
  • Exploratory factor analysis
  • Confirmatory factor analysis and latent variables

 

Day 3

  • Related groups models and higher order latent factors
  • Multilevel models
  • Review of logistic and Poisson regression

 

Day 4

  • Latent Growth Models
  • Diagnostics including assessing normality, outliers, linearity

 

Day 5

  • Accounting for missing data
  • Tests for mediation and moderation

 

Course format: 

This course may run in a computer lab, or you may be advised to bring your own laptop with specified software.

We will let you know in advance.

 

As part of this course you will be using data from the Australian Data Archive (ADA). These datasets are restricted, so you will need to apply to the ADA for access as a prerequisite for the course. In the weeks leading up to the course, ACSPRI will contact you with detailed instructions on how to do this.

 

Approximately half of the time during this course will be spent in PowerPoint presentations, and half of the time in computer demonstrations and self-paced computer exercises. The majority of the computer exercises will be conducted in AMOS, the remaining secondary exercises will be conducted in SPSS.

Recommended Background: 

Participants must have completed the course Fundamentals of Multiple Regression or an equivalent course at university level and/or have equivalent experience. Familiarity with analysis of variance, factor analysis or regression is desirable, but not strictly necessary. It is assumed that participants have little or no familiarity of structural equations with latent variables.

Recommended Texts: 

General Reading:

  • Bollen, Kenneth A. (1989). Structural Equations with Latent Variables. New York: John Wiley & Sons.
  • Kline, Rex B. (2005). Principles and Practice of Structural Equation Modeling. (2nd Ed.). New York: Guilford Press.
  • Schumacker, Randall & Lomax, Richard. (2004). A Beginner's Guide to Structural Equation Modeling. (2nd Ed.). Mahwah, N.J.: Lawrence Erlbaum Associates.
FAQ: 

Q: Do I have to have any prerequisites to do this course?

A: Yes, you must have completed Fundamentals of Multiple Regression or a University course at the same level

 

Q: What should I bring to the workshop?

A: Different workshop participants will be at different stages of their research. While we will explore prepared computer exercises with provided datasets during this workshop, some participants might want to bring their own data to this workshop to analyse as well. In addition workshop participants will also have the opportunity to present their own SEM models to the class (where this might involve current or future research). There is no requirement for participants to have their own data or to present their SEM models, but some participants may want to take advantage of this opportunity.

 

Participant feedback: 

Mark is an excellent lecturer and designed and balanced the course well!

 

Achieved my objective of kick starting my SPSS knowledge and starting with AMOS to get my DAB analysis under control

 

I can now build SEM models and Incorporate in my research and publications. It helped well to further my knowledge of SPSS

 

I came with a 0 zero knowledge of SEM and its application with real data. At the end of the course, I can keep my thoughts towards and work around SEM.

 

Great to learn content/theory and then work through problems with solutions

 

Was great, ample opportunity to practice skills and ask questions

 

The content is focused, straight to what I need, practice also makes it better. The instructor is amazingly supportive and friendly, knowledgeable.

 

Provided a clear and concise explanation of SEM fundamentals along with practical execises which enable us to gain experience in all elements taught within the course.

 

Mark is very helpful. He always answer all the questions. Spend time helping students to do exercising. Thank you so much Mark. You are a great teacher.

 

The fundamentals and ground work covered was exeptionally useful. I feel I now have a very solid understanding of regression, path analysis and SEM.

 

Notes: 

The instructor's bound, book length course notes will serve as the course texts.