An Introduction to Structural Equation Modelling

Years: 1,2,3


Description

Structural Equation Modelling (SEM) is a powerful multivariate statistical technique which enables researchers to examine several regression equations simultaneously. This session will provide an introduction to the key concepts involved in SEM, including latent, exogenous, and endogenous variables and their graphical notation. Students will also be introduced to the concepts of both the measurement and structural model, before being taken on a step-by-step journey through the process of data analysis, stopping off on the way to consider issues of model specification, data collection, model estimation, model evaluation, and model modification. The session will conclude with a demonstration of how to interpret the output of an SEM analysis and to report the findings/revealed model correctly using both text and appropriate diagrams/figures.

Access Instructions

Bookings for these online courses are via: https://bookwhen.com/ueaonlinetraining

The demand for these courses is high and you are advised to book your place as soon as the booking system goes live.

Information on the booking dates will be circulated by email.

Indicative Student Feedback for this Session

You have clarified SEM and given me so much more understanding of it. Although I've been reading up on it, you have connected so many dots for me. I believe I will also understand even better the analyses aspects of the papers I read (Univ. of East Anglia; Business).

Thanks for the session. I was concerned before it started that most of it would go completely over my head, but now the method seems less daunting and unachievable…For me, having a list of steps as a guide is probably the handiest thing (Goldsmiths).

Thanks for a really helpful lecture – I was already pretty familiar with SEM but it was interesting to hear about the different software packages available for this approach. I liked the explanation of the difference between the measurement and structural models, and solutions to the problem of missing data or multi-collinearity. The session clarified the idea of adding or dropping parameters, and which test to use for this, and I was not familiar with parcelling before, so it was good to be introduced to that concept. Also, the rule of thumb for sample size seems a really useful thing to know! Kim explained things at a really good pace and in a very clear way, and I liked how she answered people’s queries in real-time to clear up any confusion (Royal Holloway).

The Vitae Researcher Development Framework

This session maps on to Domain A of the Vitae Framework: Knowledge and intellectual abilities which includes the sub-domain areas of:

  • A1: Knowledge base
  • A2: Cognitive abilities
  • A3: Creativity
The Vitae Researcher Development Framework