Ordinal logistic regression

Presenter(s): Dr Heini Väisänen

This three-part series gives a short introduction to ordinal logistic regression. The method can be used in situations, where the outcome (dependent) variable has at least three categories that are ordered. The series introduces the principles of the method, uses empirical examples to explain how the method is used and includes a computer workshop exercise task, which shows how to put this knowledge into practice using Stata. It is recommended that all three videos should be viewed before attempting the computer workshop task.

Part 1: Introduction to ordinal logistic regression

This video introduced the method and its cumulative nature. It shows a simple example with one explanatory variable to illustrate how the method works and how the results can be interpreted using either cumulative odds or predicted probabilities.

Part 2: Multivariate ordinal logistic regression

This video discusses ordinal logistic regression models with more than one explanatory variable. It also introduces some principles of model selection, including the use of Wald-tests and likelihood ratio tests.

Part 3: The proportional odds assumption

The last video of the series discusses what we mean by the proportional odds assumption when we conduct ordinal regression models. It illustrates the assumption by showing an example comparing results from a multinomial and an ordinal model.