MRes Criminology (Social Statistics) / Course details

Year of entry: 2024

Course unit details:
Structural Equation and Latent Variable Modelling

Course unit fact file
Unit code SOST70042
Credit rating 15
Unit level FHEQ level 7 – master's degree or fourth year of an integrated master's degree
Teaching period(s) Semester 2
Available as a free choice unit? Yes


This course unit introduces students to structural equation modelling (SEM), a family of models that encompasses regression, path/mediation analysis, factor analysis, and more. Emphasis is given to the flexibility engendered by the SEM approach, which integrates several methods that are often and unhelpfully presented as inflexible and stand-alone. The 'traditional' approach to SEM, based upon continuous observed variables and assuming continuous latent variables, is expanded to encompass models for categorical observed variables. The resulting modelling framework, termed generalized latent variable modelling, is a highly flexible, modular tool for modelling and testing complex social science data. The course also introduces students to the lavaan package in R, which can be used to estimate these models from data.


Unit title Unit code Requirement type Description
Introduction to Statistical Modelling SOST70011 Pre-Requisite Recommended


Students should have completed introductory/intermediate training in statistical analysis and research design, such that they are familiar with:

  • Non-experimental, survey-based research; its strengths and limitations.
  • Linear and logistic regression analyses; in particular the meaning the b coefficients.
  • The R software package, for fitting linear and logistic regression models.
  • Part time students must take ISM prior to the course


  • To introduce students to modern latent variable and structural equation modelling, so that they can specify, estimate, interpret and critically discuss a range of such models based on relevant research questions.
  • To introduce students to the lavaan library in R, which we will use to specify and fit a range of structural equation and latent variable models, including: confirmatory factor analysis, item-response theory models, mediation/path analysis, latent growth models.


Week 1: Introduction to SEM and causal analysis

Week 2: Confirmatory Factor Analysis I

Week 3: Confirmatory Factor Analysis I

Week 4: Mediation

Week 5: Item Response Theory I

Week 6: Item Response Theory I

Week 7: Measurement bias

Week 8: Missing data and practice models

Week 9: Multilevel CFA

Week 10: Review and practice models

Teaching and learning methods

Each week (except the first) we will give you some homework activities (something to read, and/or watch, and/or do). During the following session we will review and explore those activities, to check our understanding. It is imperative that you carry out the homework activities before the session, as the sessions are not lectures as such; they are a chance for us to ask each other questions to check our understanding of the material. The weekly sessions will be 2-hour classes consisting of review of materials, Q&A session, and hands-on practical exercises using R software. In the exercise the students will be required to carry out formative tasks designed to strengthen their understanding. Weekly back-up support will also be provided in the form of office hours.

Knowledge and understanding

Understand the nature of structural equation modelling and its relationship to other statistical methods, specifically regression, path, and latent variable models. Distinguish between and use models for categorical and continuous outcome variables. Identify the contexts when different structural equation models are appropriate.

Intellectual skills

Be able to critically evaluate examples of latent variable and/or structural equation modelling. Be able to translate conceptual theory/hypothesis into appropriate latent variable and structural equation models. Make appropriate scientific inferences from the results of structural equation models.

Practical skills

Use R to specify and fit a range of structural equation models to social datasets. Interpret the parameter estimates generated by different structural equation models.

Transferable skills and personal qualities

Write a report that synthesises evidence from relevant literature and the student’s own analysis; exercise self-management skills in terms of pacing workload and meeting deadlines; gain experience in analysing quantitative social data.

Employability skills

Analytical skills
Probabilistic and broader numerical skills/training. Statistical analysis and data handling skills. Practice in technical report writing.

Assessment methods

Formative assessments

  1. Understanding causality on DAGs (up to 300 words)
  2. Interpretation of SEM model results (up to 300 words)
  3. Formative assignment on SEM model building for causal hypothesis testing in R (a short coding assignment, equivalent to up to 300 words).

Summative assessments 

  1. 25% Understanding causality on DAGs assignment, using a multiple-choice test equivalent to a half-hour exam
  2. 25% SEM Model results interpretation assignment, using a multiple-choice test equivalent to a half-hour exam
  3. 50% SEM model building for causal hypothesis testing assessment, using a 1,500 word report on an analysis in R conducted by the student.

Feedback methods

Feedback available via Turnitin

Recommended reading

  • Kaplan, D. (2009). Structural Equation Modeling: Foundations and Extensions (2nd Ed.). Thousand Oaks, CA: Sage
  • Kline, K. (2018). Principles and Practice of Structural Equation Modelling (4th Ed.). New York: Guildford.                                                                            

Study hours

Scheduled activity hours
Lectures 20
Practical classes & workshops 10
Independent study hours
Independent study 120

Teaching staff

Staff member Role
Nicholas Shryane Unit coordinator

Additional notes



Return to course details