Multilevel Models for Cross Sectional Data: Summer 2014 (ICPSR)

Upcoming Workshops:

Thank you for visiting my course notes. Here are some upcoming opportunities to learn from me and my colleagues in person:

Currently no workshops are planned

Applied Multilevel Models for Cross Sectional Data
Workshop Syllabus

ICPSR Summer Workshop in Boulder, Colorado
July 14 – 18, 2014

Presented by:

Jonathan Templin, Ph.D. (
Associate Professor, Department of Psychology and Research in Education
Achievement and Assessment Institute
University of Kansas

Teaching Assistant:

Meghan Sullivan
Graduate Student: Department of Psychology and Research in Education
Research, Evaluation, Measurement, and Statistics Program
University of Kansas

Course Description

Multilevel models are powerful statistical models that partition multiple sources variation that may be present due to dependencies in data. Also known as hierarchical linear models mixed effects models, multilevel models extend traditional linear models (such as regression or analysis of variance) to analyses where data structures are clustered, nested, or hierarchical in nature. This workshop presents an introduction to multilevel models featuring their use in cross-sectional analyses. By attending the workshop, participants will gain an understanding of the multilevel modeling approach and will be able to evaluate and conduct basic multilevel model analyses.

The topics spanned in the workshop will be discussed in an integrated framework, with the first day being a review of general linear models beginning with unconditional models and the rules of model comparisons. The second day will feature two-level models: adding random components and adding single predictors, including a discussion of predictor centering techniques. The third and fourth day will be spent on multilevel models with multiple predictors and models with three or more levels. The final day will be spent discussing advanced topics: multilevel models with multivariate predictors and crossed random effects models.

The primary software package used for instruction will be SAS, but some reference examples using SPSS, Mplus, and R will be provided. The course will also include daily opportunities for hands-on practice and individual consultation. Participants should be familiar with ANOVA and regression, but no prior experience with multilevel models or knowledge of advanced mathematics is assumed.

Overall Course Files
Zipped Folder of All Syntax FilesZip File
All Lecture Slides PDFPDF File
Schedule of Topics
DateLecture SlidesSyntax FilesData Files
Monday, July 14thIntroduction to Multilevel Models and Hierarchical Data
The General Linear ModelSAS Syntax
Simple, Marginal, and Interaction Effects in GLMsSAS SyntaxSAS Data
Statistical Distribution Assumptions of GLM/Maximum Likelihood
Lab 1: Introduction to Data Manipulation in SASExample Zipped Folder
Tuesday, July 15thMultilevel Models – a Guiding ExampleSAS Syntax
SAS Syntax Handout
Data File (CSV)
Centering Predictors and Variance Decomposition SAS Syntax
SAS Syntax Handout
SPSS Syntax
SPSS Syntax Handout
Mplus Syntax Handout
Data File (CSV)
Random Slopes, Cross-Level Interactions, InterpretationsSAS SyntaxData File (CSV)
Lab 2: Fitting Single-Predictor Multilevel ModelsExample Zipped Folder
Wednesday, July 16thComprehensive Overview of Multilevel Models
Two-Level Clustered Data – Students within SchoolsSAS Syntax Handout
SPSS and STATA Syntax Handout
Two-Level Cross-Classified Data ModelsSAS Syntax Handout
Three Level Models (Part 1)SAS Syntax
Lab 3: Fitting Multi-Predictor Multilevel ModelsExample Zipped Folder
Thursday, July 17thThree Level Models (Part 2): Clustered Longitudinal Designs
Three Level Models ExampleSAS Syntax Handout
SPSS and STATA Syntax Handout
Multivariate Normal Distribution and Multivariate AnalysesSAS Syntax
Multilevel Models in Matrix Form
Lab 4: Multivariate ModelsExample Zipped Folder
Friday, July 18thMultivariate Multilevel ModelsSAS Syntax
Generalized Multilevel Models (Non-Normal Outcomes)SAS Syntax
Generalized Multivariate Multilevel ModelsSAS Syntax