Course Description
Multilevel models with linear, non-linear, and growth outcomes.
Other topics include estimation, reliability, model building
strategies, intra-class correlation, plausible values, fixed, and
random effects. Focus is on conceptualizing, conducting,
interpreting, and writing up multilevel analyses, as well as
understanding relevant statistical and practical issues.
Athena Title
Multilevel Modeling
Prerequisite
ERSH 8320 or ERSH 8320E
Semester Course Offered
Offered every year.
Grading System
A - F (Traditional)
Student learning Outcomes
- Students will demonstrate their understanding of the topics in this course by completing a series of computer-based assignments.
- Students will demonstrate their understanding of the topics in this course by presenting published studies that utilize a multilevel statistical model, and by participating in class discussions.
- Students will demonstrate their ability to carry out a study that employs a multilevel model by conducting their own study using a dataset of their choice.
- Students will be able to present the results of their study both orally in class and in the form of a research paper.
Topical Outline
- Week 1: Introduction to multilevel models and their applications
- Week 2: Survey of Multilevel models, One-way Analysis of Variance with Random Effects and means-as-outcomes
models, applications in organizational research
- Week 3: Analysis of Covariance models and centering
- Week 4: Random-coefficient and slopes-as-outcomes models
- Week 5: Estimation and hypothesis testing
- Week 6: Model building and assessment
- Week 7: Individual growth models
- Week 8: Piecewise growth models
- Week 9: Three-level models
- Week 10: Non-linear models: Bernoulli outcome
- Week 11: Non-linear models: Multinomial outcome
- Week 12: Non-linear models: Order categories and count outcomes
- Week 13: Residual analysis: Using model residuals to answer research questions
- Week 14: Multilevel models for studying school effects and program evaluation
- Week 15: Cross-classified random effects models
- Week 16: Multilevel Structural Equation Models
