Multilevel Modeling of Non-Normal Data
A 2-Day Seminar Taught by Donald Hedeker, Ph.D.
Multilevel models are increasingly used for analysis of clustered and longitudinal data, and methods for continuous outcomes are commonly used and applied. However, many research studies have non-normal outcomes, for example, outcomes that are dichotomous, ordinal, or nominal. Although methods for such non-normal outcomes have been available for quite some time, they are perhaps not as routinely applied as models for continuous outcomes.
This workshop will focus on analysis of dichotomous, ordinal and nominal multilevel outcomes. Both clustered and longitudinal data will be considered, and the following models will be described: multilevel logistic regression for dichotomous outcomes, multilevel logistic regression for nominal outcomes, and multilevel proportional odds and non-proportional odds models for ordinal outcomes. The latter models are useful because the proportional odds assumption of equal covariate effects across the cumulative logits of the model is often unreasonable.
Who should attend?
If you plan on analyzing multilevel data (either clustered or longitudinal) and have categorical outcomes of interest, this course is for you. Analyzing multilevel categorical data is necessary in many research fields where normally distributed continuous outcomes are not obtained or relevant.
Participants should be thoroughly familiar with multiple linear regression, and have some knowledge of logistic regression.
In all cases, methods will be illustrated using software, with SAS, Stata, and SuperMix examples and syntax. Some familiarity with reading in data and performing basic statistical analyses in either SAS or Stata is recommended.
This is a hands-on course with at least one hour each day devoted to carefully structured and supervised assignments. To do the exercises, you will need to bring your own laptop computer with a recent version of SAS or Stata and the free student version of SuperMix installed. SuperMix can be downloaded at http://www.ssicentral.com/supermix/downloads.html. Power outlets will be provided at each seat.
NOTE: If you have Windows 7 or older, please run Supermix as administrator. Windows 8 and more recent should run as a regular user.
LOCAtion, Format, And Materials
The course will meet Thursday, May 18 and Friday, May 19 at the Conference Chicago at University Center, 525 South State Street Chicago, IL 60605.
The class will meet from 9 to 5 each day with a 1-hour lunch break.
Participants receive a bound manual containing detailed lecture notes (with equations and graphics), examples of computer printout, and many other useful features. This book frees participants from the distracting task of note taking.
Registration and lodging
The fee of $995.00 includes all seminar materials. The early registration fee of $895 is available until April 18.
If you cancel your registration at least two weeks before the course is scheduled to begin, you are entitled to a full refund (minus a processing fee of $50).
Lodging Reservation Instructions
A block of guest rooms has been reserved at the The Congress Plaza Hotel, 520 S. Michigan Avenue, Chicago, IL 60605 at a special rate of $189 per night. This location is about a 5 minute walk to the seminar location. You can make a reservation online by clicking here. Alternately, you can call 312-427-3800 x 5025 or toll free at 1-800-635-1666 during business hours and identify yourself as part of the Statistical Horizons room block. For guaranteed rate and availability, you must reserve your room no later than Monday, April 17, 2017.
- Introduction to multilevel analysis
- Multilevel analysis of clustered dichotomous outcomes
- Multilevel analysis of clustered ordinal outcomes
- Testing of the proportional odds assumption
- Multilevel non-proportional odds models for ordinal data
- Multilevel analysis of longitudinal dichotomous outcomes
- Subject-specific and population-averaged estimates
- Marginalization of the subject-specific estimates
- Multilevel analysis of longitudinal ordinal outcomes
- Multilevel analysis of longitudinal nominal outcomes
“Dr. Hedeker is both knowledgeable and approachable; that means that the workshop not only contained a tremendous amount of useful information but also that he was able to answer specific individual questions. He explained complex concepts in easy to understand language and kept the tone of the workshop light throughout—an achievement after 8 hours of modeling. Thank you!”
Tracey Andrews, Hackensack University Medical Center
“This course provided excellent information about non-normal data that nicely complemented my graduate programs MLM course. Ultimately, the information provided in this course will help me to confidently pursue a larger breadth of research questions and for that I am very grateful!”
Michelle Haikalis, University of Nebraska-Lincoln
“Dr. Hedeker did a great job of taking a very complex topic and making it applicable to real-world data. He showed us how to do the analyses on three different platforms and explained the pros and cons of each. I feel much more equipped to analyze my non-normal multilevel data because of this seminar. Many of my questions were answered over the course of the two days.”
Lindsay Pitzer, Truth Initiative
“I was not sure about which statistical packages are more comprehensive on handling multi-level data with or without longitudinal information. Great to know SuperMix is so powerful to model multilevel data.”
Deepak Adhikari, University of Texas Medical Branch
“It was very useful especially with all the detailed examples which are real and related.”
Aseel Almansour, International Monetary Fund
“Dr. Hedeker made very complex material engaging, understandable and almost fun.”
Amy Watson, University of Illinois at Chicago
“It was indeed a great course. It distinguishes clearly between the multilevel model and longitudinal model. Also, why normal logistic model would not work in clustered structure.”
“Don provided very clear explanation of the material that made it much easier to understand. Specifically, he described the pros and cons of each model for non-normal data and the underlying assumptions of each one. The course was excellent!”
Natania Crane, University of Illinois at Chicago