Multilevel and Structural Equation Modelling
Beschrijving
Course goals
TESTING AND COURSE AIMS
The multilevel modeling part of this course is tested with
- Home assignment(s) (20%), which consists of multiple exercises spread out over multiple weeks and that students make in groups
- An individual exam (30%), in which students have to show they developed knowledge and understanding of multilevel modelling, and can perform the correct analyses and interpret results in a limited amount of time
With this assignment and exam we measure the extent to which
- Students have developed knowledge and understanding of multilevel modeling (Knowledge and understanding)
- Students have learned to translate research questions into (aspects of) multilevel models, and specify relevant multilevel models in R (Knowledge and understanding)
- Students can translate research questions into sequence of models to be run (Learning skills)
- Students can interpret output related to multilevel modelling in terms of relevance and statistical significance (Applying knowledge and understanding)
- Students can relate multilevel modelling output to research question (Applying knowledge and understanding)
The structural equation modeling part of this course is tested with
- an assignment (10%), which consists of multiple exercises spread out over multiple weeks and that students make in groups
- an individual test (40%), in which students have to show they can perform the correct analyses and interpret results in a limited amount of time
With this assignment and test we measure the extent to which
- Students have developed knowledge and understanding of structural equation modeling (Knowledge and understanding)
- Students have developed knowledge and understanding of the computer program Mplus (Knowledge and understanding)
- Students have learned to translate research questions into (aspects of) structural equation models, and specify relevant structural equation models in Mplus (Knowledge and understanding)
- Students can translate research questions into sequence of models to be run (Learning skills)
- Students can use specialized software (Learning skills)
- Students can interpret Mplus output in terms of relevance and statistical significance (Applying knowledge and understanding)
- Students can relate Mplus output to research question (Applying knowledge and understanding)
- Students can understand and interpret the TECH1 output from Mplus (Knowledge and understanding)
- Students can represent a model using the SEM matrices (Applying knowledge and understanding)
- Students can interpret Mplus output in terms of model fit (Applying knowledge and understanding)
- Students can relate Mplus output to research questions (Applying knowledge and understanding)
Content
Two techniques that are often encountered are multilevel modeling (MLM) and structural equation modeling (SEM).
MLM is appropriate for handling nested data, for instance, patients in hospitals, or occasions in people. MLM can be used to study the within cluster and the between cluster relationships between an outcome variable and predictors. The course gives a broad introduction to multilevel modelling, followed by a number of advanced topics. The multilevel part will cover the two and three level multilevel model, analyzing longitudinal data, and contextual effects using multilevel models.
SEM covers both factor analyses and path analyses. It thus combines latent variable modeling with mediation analysis in a broad framework that allows for a large variety of models and research questions. In the course, we cover how to translate a theory into a model and how to specify, fit and evaluate models, including factor models, multiple group factor analysis (i.e., measurement invariance), mediation models, and longitudinal models including latent growth curve and cross-lagged panel models. In the lab meetings Mplus is used.
Reviews0 reviews
Heb jij dit vak gevolgd?
Deel je ervaring met toekomstige studenten. Inloggen met je Universiteit Utrecht mailadres duurt één minuut.
Schrijf een review