NMAK16018U Structural Equation Models
MSc Programme in Statistics
The course is an introduction to latent variable models. We introduce Item Response Theory (IRT) models, focusing mainly on the Rasch model, Confirmatory Factor Analysis (CFA) models, and Structural Equation Models (SEM’s). The exercises will be a mixture of theoretical problems and data analysis. The course covers the following topics:
- General measurement models (including Rasch model and CFA)
- Conditional and marginal estimation
- Model identification
- Evaluation of model fit
- Path analysis
- Causal inference
- Structural equation models
- Measurement error in covariates
- An introduction to implementations of the methodology in R.
At the end of the course the student will have knowledge about different types of latent variable models, and will have the knowledge to
- Explain the assumptions underlying the models
- Interpret the parameters of the models
- Discuss model identification and be able to determine if two models are identical
The student will acquire skills necessary for applying latent variable models to real data, decide on which model to use and which analysis to perform. The student will have the skills to utilize theoretical results in the practical analysis, including how complex models can be specified.
At the end of the course the students will have the competence to
- Evaluate the fit of measurement models (including Rasch model and CFA)
- Estimate the parameters of structural equation models
- Use latent variable models to adjust for measurement error in covariates
- 7,5 ECTS
- Type of assessment
- Oral examination, 30 minutes without preparationEvery week a few statistical problems will be given. Students will in turn present solutions in class followed by a plenary discussion of these solutions. The student's own solutions will form the basis of the first part of the oral examination. The second part will be devoted to a general discussion of the contents of the course.
- All aids allowed
- Marking scale
- 7-point grading scale
- Censorship form
- No external censorship
Several internal examiners
Same as ordinary.
Criteria for exam assesment
The student must in a satisfactory way demonstrate that he/she has mastered the learning outcome of the course.