Please forgive the silly subject line above; I was trying to draw you into this post, which highlights an upcoming webinar sponsored by the Section for Statistical Programmers and Analysts.
Donald Hedeker, at the University of Illinois at Chicago, will be presenting a short workshop on "Mixed Models for Longitudinal Categorical Outcomes."
The webinar will be held Tuesday, April 30, 2013, from 12:00 p.m. - 1:30 p.m. Eastern Time.
This workshop will focus on analysis of longitudinal data using mixed models, specifically with a non-continuous outcome. When responses are repeatedly measured on the same individual or clustered within a group of similar individuals (for instance in the same household, hospital, or state regulatory environment) these observations might not be independent. As such, a fixed-effects model is not appropriate. Discrete outcomes are often observed in many areas of research, as well as being advantageous in many instances, such as pass/fail or poor/fair/good.
The following models will be described: mixed logistic regression for dichotomous outcomes, mixed logistic regression for nominal outcomes, and mixed 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. Use of the program SuperMix for these models will be described and illustrated.
Registrants should be familiar with logistic regression and mixed models for continuous outcomes.