Course content

Whether the goal is to track educational inequalities (SDG 4), monitor health disparities (SDG 3), or compare social outcomes across groups and countries (SDG 10), the quality of research conclusions depends on the quality of measurement. Flawed or non-comparable instruments undermine the validity of findings, yet the methods for evaluating measurement quality are rarely taught systematically outside specialist programmes. This course addresses that gap.

This course introduces the theory and application of measurement models and their role in structural equation modelling (SEM), with a focus on applied research. You will learn to specify, estimate, and evaluate path models, factor models, and full SEM with latent variables, with particular emphasis on measurement invariance: whether a scale measures the same construct in the same way across groups, countries, or contexts. All analyses are carried out hands-on in the free, open-source software R, mainly using the lavaan package.

The course covers the following key topics:

  • Overview of latent variable models, measurement error, and the role of SEM in applied research
  • Path models and diagrams
  • Model specification, identification, estimation, and evaluation
  • Exploratory and confirmatory factor analysis
  • Full structural equation models with latent variables
  • Moderation and mediation models
  • Multigroup analysis and measurement invariance across groups and contexts

Learning Outcomes

  • Recognise the general principles of measurement models
  • Understand the key assumptions that underlie these models and methods
  • Understand what violations of their assumptions can mean for model selection and associated inferences
  • Understand how choices of estimation method, model specification, and fit evaluation shape research conclusions
  • Select, apply, and interpret the parameters of a measurement model for the research question at hand, for instance, in the context of structural equation modelling
  • Test key assumptions and offer possible solutions to violations
  • Test for measurement invariance across groups and interpret what non-invariance means for substantive conclusions
  • Write up and report the results of an analysis in line with good scientific practice
  • Analyse data with the help of the open-source software R (lavaan package)
  • Demonstrate proficiency in using measurement models to answer well-defined research questions, for instance, in the context of structural equation modelling
  • Interpret published scientific research that uses these models and methods
  • Evaluate the tenability of associated inferences and knowledge claims

Teaching

This course combines lectures and seminars with hands-on data analysis tasks in R. The course is offered in a hybrid format: master’s students at the University of Oslo attend on site, while students from other Circle U. universities participate fully digitally.

Obligatory course components are an 80% attendance requirement in teaching and the completion of two individual written assignments.

  • 17.03.2027 10:15–12:00 and 12:15–14:00
  • 19.03.2027 10:15–12:00
  • 31.03.2027 12:15–14:00
  • 02.04.2027 10:15–12:00
  • 14.04.2027 10:15–12:00 and 12:15–14:00
  • 16.04.2027 10:15–12:00
  • 21.04.2027 12:15–14:00
  • 23.04.2027 10:15–12:00
  • 28.04.2027 12:15–14:00
  • 30.04.2027 10:15–12:00
  • 05.05.2027 12:15–14:00
  • 07.05.2027 10:15–12:00
  • 12.05.2027 12:15–14:00
  • 14.05.2027 10:15–12:00
  • 19.05.2027 12:15–14:00
  • 21.05.2027 10:15–12:00

Lecturers

The course is taught by Prof. Ronny Scherer (Centre for Educational Measurement, University of Oslo) together with collaborators from Humboldt-Universität zu Berlin and King’s College London, giving students access to measurement expertise from three Circle U. universities.

Examination

To obtain 10 ECTS credits, students must complete the obligatory course components and pass an individual take-home examination (7 to 11 June 2027) covering all course material. The examination is conducted in English and graded on a scale from A to F. Successful completion of the obligatory course components is required before sitting the exam.

Admission requirements

Admission requires enrolment in a master’s programme at a Circle U. university.

Recommended previous knowledge: data management (wrangling and auditing), descriptive statistics, data visualisation (e.g., plots, tables, diagrams), statistical inference, and basic use of the software R. No prior experience with structural equation modelling is required.

Ranking of applicants

There are 30 available study places. The course is compulsory in the master’s programme in Assessment, Measurement and Evaluation. Students enrolled in this programme are given priority; the remaining places are open to Circle U. students.

If there are more eligible Circle U. applicants than available places, places will be allocated by random draw, giving all eligible candidates an equal chance to participate. Circle U. students accepted to this course are exempt from paying tuition fees.

Contact

Course related questions:

For questions related to funding, ECTS recognition or mobility, please contact the Circle U. office at your home university:

Ready to apply?

Complete the online form below to apply.
Apply by 5 January 2027

The alliance of 9 European Universities