Multilevel Modeling of Categorical Outcomes - Online Course
A 4-Week On-Demand Seminar Taught by
Donald HedekerMonday, May 4 —
Monday, June 1, 2026
Each Monday you will receive an email with instructions for the following week.
All course materials are available 24 hours a day. Materials will be accessible for an additional 2 weeks after the official close on June 1.
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, counts, 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, nominal, and count multilevel outcomes. Both clustered and longitudinal data will be considered, and the following models will be described: multilevel logistic regression for dichotomous outcomes, multilevel models for count data, multilevel logistic regression for nominal outcomes, multilevel Poisson regression for counts, 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 inconsistent with the data.
The course takes place online in a series of four weekly installments of videos, readings, and exercises, and requires about 6-8 hours/week. You may participate at your own convenience; there are no set times when you are required to be online.
This four-week course can be accessed with any recent web browser on almost any platform, including iPhone, iPad, and Android devices. It consists of several modules, which contain videos of the 4-day livestream version of the course in its entirety. There are also weekly exercises that ask you to apply what you’ve learned to a real data set.
There is also an online discussion forum where participants can post questions or comments about any aspect of the course. All questions will be promptly answered by Dr. Hedeker.
Computing
In all cases, methods will be illustrated using software, with SAS and Stata examples and syntax. Some familiarity with reading in data and performing basic statistical analyses in either SAS or Stata is recommended.
To do the exercises, you will need a computer with a recent version of SAS or Stata.
There is now a free version of SAS, called SAS OnDemand for Academics, that is available to anyone.
If you’d like to familiarize yourself with Stata basics before the seminar begins, we recommend following along with a “getting started” video like the one here.
Seminar participants who are not yet ready to purchase Stata could take advantage of StataCorp’s 30-day software return policy.
In all cases, methods will be illustrated using software, with SAS and Stata examples and syntax. Some familiarity with reading in data and performing basic statistical analyses in either SAS or Stata is recommended.
To do the exercises, you will need a computer with a recent version of SAS or Stata.
There is now a free version of SAS, called SAS OnDemand for Academics, that is available to anyone.
If you’d like to familiarize yourself with Stata basics before the seminar begins, we recommend following along with a “getting started” video like the one here.
Seminar participants who are not yet ready to purchase Stata could take advantage of StataCorp’s 30-day software return policy.
Who should register?
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.
You should be thoroughly familiar with multiple linear regression, and have some knowledge of logistic regression.
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.
You should be thoroughly familiar with multiple linear regression, and have some knowledge of logistic regression.
Seminar outline
1. Introduction to multilevel data
-
- What are multilevel data?
- Why do multilevel data analysis?
- Overview of multilevel models for categorical data.
2. Multilevel analysis for clustered binary outcomes
-
- Multilevel logistic regression model.
- Scaling of regression coefficients.
- Students in classrooms and schools example.
- Using Stata, SAS, and Supermix for multilevel analysis.
- Interpreting output, testing for cluster effects, creating plots of cluster effects.
- Fit of models to observed proportions.
3. Multilevel models for clustered ordinal outcomes
-
- Multilevel cumulative logistic regression model.
- Proportional odds assumption.
- Students in classrooms and schools example.
- Using Stata, SAS, and Supermix for multilevel ordinal analysis.
- Interpreting output, and testing for cluster effects.
- Fit of models to observed proportions.
- Non-proportional odds multilevel models.
4. Introduction to multilevel models for longitudinal binary outcomes
-
- Multilevel logistic regression model.
- Random intercept and trend models.
- Subject-specific and population-averaged estimates.
- Longitudinal psychiatric clinical trials example – descriptive statistics and plots.
5. Multilevel for longitudinal binary outcomes (continued)
-
- Longitudinal psychiatric clinical trials example – random intercept and trend models.
- Using Stata, SAS, and Supermix for multilevel analysis.
- Conditional and marginalized estimates.
- Interpreting output, testing for random subject effects, creating plots.
- Fit of models to observed proportions.
6. Multilevel models for longitudinal ordinal outcomes
-
- Multilevel cumulative logistic regression model.
- Proportional odds assumption.
- Random intercept and trend models.
- Longitudinal psychiatric clinical trials example – descriptive statistics and plots.
- Using Stata, SAS, and Supermix for multilevel analysis.
- Conditional and marginalized estimates.
- Interpreting output, testing for random subject effects, creating plots.
- Fit of models to observed proportions.
7. Multilevel models for longitudinal ordinal and nominal outcomes
-
- Longitudinal homelessness study – descriptive statistics and plots.
- Using Stata, SAS, and Supermix for multilevel analysis.
- Non proportional odds models; testing proportional odds assumption.
- Treating outcome as nominal; reference cell comparisons.
- Interpreting output, testing for random subject effects, creating plots.
- Fit of models to observed proportions.
8. Multilevel analysis of longitudinal count outcomes
-
- Multilevel Poisson regression model.
- Number of headaches and Aspartame crossover study.
- Using Stata, SAS, and Supermix for multilevel Poisson analysis.
- Interpreting output, and testing for random subject effects.
- Fit of models.
- Overdispersion.
1. Introduction to multilevel data
-
- What are multilevel data?
- Why do multilevel data analysis?
- Overview of multilevel models for categorical data.
2. Multilevel analysis for clustered binary outcomes
-
- Multilevel logistic regression model.
- Scaling of regression coefficients.
- Students in classrooms and schools example.
- Using Stata, SAS, and Supermix for multilevel analysis.
- Interpreting output, testing for cluster effects, creating plots of cluster effects.
- Fit of models to observed proportions.
3. Multilevel models for clustered ordinal outcomes
-
- Multilevel cumulative logistic regression model.
- Proportional odds assumption.
- Students in classrooms and schools example.
- Using Stata, SAS, and Supermix for multilevel ordinal analysis.
- Interpreting output, and testing for cluster effects.
- Fit of models to observed proportions.
- Non-proportional odds multilevel models.
4. Introduction to multilevel models for longitudinal binary outcomes
-
- Multilevel logistic regression model.
- Random intercept and trend models.
- Subject-specific and population-averaged estimates.
- Longitudinal psychiatric clinical trials example – descriptive statistics and plots.
5. Multilevel for longitudinal binary outcomes (continued)
-
- Longitudinal psychiatric clinical trials example – random intercept and trend models.
- Using Stata, SAS, and Supermix for multilevel analysis.
- Conditional and marginalized estimates.
- Interpreting output, testing for random subject effects, creating plots.
- Fit of models to observed proportions.
6. Multilevel models for longitudinal ordinal outcomes
-
- Multilevel cumulative logistic regression model.
- Proportional odds assumption.
- Random intercept and trend models.
- Longitudinal psychiatric clinical trials example – descriptive statistics and plots.
- Using Stata, SAS, and Supermix for multilevel analysis.
- Conditional and marginalized estimates.
- Interpreting output, testing for random subject effects, creating plots.
- Fit of models to observed proportions.
7. Multilevel models for longitudinal ordinal and nominal outcomes
-
- Longitudinal homelessness study – descriptive statistics and plots.
- Using Stata, SAS, and Supermix for multilevel analysis.
- Non proportional odds models; testing proportional odds assumption.
- Treating outcome as nominal; reference cell comparisons.
- Interpreting output, testing for random subject effects, creating plots.
- Fit of models to observed proportions.
8. Multilevel analysis of longitudinal count outcomes
-
- Multilevel Poisson regression model.
- Number of headaches and Aspartame crossover study.
- Using Stata, SAS, and Supermix for multilevel Poisson analysis.
- Interpreting output, and testing for random subject effects.
- Fit of models.
- Overdispersion.
Payment information
The fee of $695 USD includes all course materials. All major credit cards are accepted.
This course is hosted on a platform called DigitalChalk. To register, you’ll need to go to statisticalhorizons.digitalchalk.com and click on Create Account. Then you will enter your name and email address, and create a password. Be sure to save your password because you will need it to logon to the course itself.
When you have created your account, you’ll be taken to your new home page. Click on the Register Now button (or click the Catalog icon on the left-hand column), and you’ll see “Multilevel Modeling of Categorical Outcomes” as one of the available courses. At the bottom of the box for that course, click the green button Add to Cart. Next click the green button at the top that says Checkout. You will then be prompted for your credit card information.
When you have finished the payment process, you will be taken back to your home page. Click on Dashboard to see Multilevel Modeling of Categorical Outcomes. When the course begins on May 4, you can click the play button to get started.
The fee of $695 USD includes all course materials. All major credit cards are accepted.
This course is hosted on a platform called DigitalChalk. To register, you’ll need to go to statisticalhorizons.digitalchalk.com and click on Create Account. Then you will enter your name and email address, and create a password. Be sure to save your password because you will need it to logon to the course itself.
When you have created your account, you’ll be taken to your new home page. Click on the Register Now button (or click the Catalog icon on the left-hand column), and you’ll see “Multilevel Modeling of Categorical Outcomes” as one of the available courses. At the bottom of the box for that course, click the green button Add to Cart. Next click the green button at the top that says Checkout. You will then be prompted for your credit card information.
When you have finished the payment process, you will be taken back to your home page. Click on Dashboard to see Multilevel Modeling of Categorical Outcomes. When the course begins on May 4, you can click the play button to get started.