University of Oregon, College of Education

Workshop on Multilevel Models


This workshop will provide an introduction to multilevel modeling.  Many social and natural phenomena have a nested or clustered organization that results in data with dependencies associated with group or cluster membership.  Hierarchical Linear Models (HLM) provide a method for correctly analyzing such data as well as a means to study relationships that cross levels. Workshop participants will learn foundational principles and concepts in HLM and will have hands-on practice using software to apply and interpret basic models.  The workshop will cover multilevel data structures, intraclass correlation, model building, centering, model testing, fixed and random effects, two and three level models, longitudinal growth models, statistical power and design planning, and the use of HLM in cluster randomized trials.

Please direct any questions to Joe Stevens at stevensj@uoregon.edu.


Before the workshop, participants should:
  • Read the following:
Osborne, J. W. (2000). Advantages of Hierarchical Linear Modeling, PARE, available at:  http://pareonline.net/getvn.asp?v=7&n=1

Hox, J.  (1995). 
Multilevel Analysis, Read chapters 1 and 2.

Raudenbush & Bryk, 1986, Sociology of Education, available at:
  http://www.uoregon.edu/~stevensj/HLM/raudenbush.pdf

Willms & Raudenbush (1989). Longitudinal HLM study of school effects, available at: 
http://www.uoregon.edu/%7Estevensj/AofC/willms&raudenbush.pdf

  • Additional Resources:
Goldstein, H. (1995). Multilevel Statistical Models

Raudenbush, et al. (2007).  Strategies for improving precision.  Educational Evaluation and Policy Analysis.

Hedges & Hedberg (2007). Intra-class correlation values for planning.  Educational Evaluation and Policy Analysis.


  • Download a copy of the workshop PowerPoint slides for note taking in pdf format (one slide/page) or as multiple slides/page

  • If you want to do hands-on activities in class:
    • Download and install the free student version of HLM
    • If you are interested in power in cluster randomization or HLM studies, download and install the Optimal Design Software
If you want to participate in hands-on activities, bring a laptop with wireless connectivity and install the HLM and optimal design software. 

Back to my Home Page