Categorical Data Analysis Using The SAS® System, 2nd Edition

15.4. Passive Smoking Example

The following data are from a hypothetical study of the effects of air pollution on children. Researchers followed 25 children and recorded whether they were exhibiting wheezing symptoms during the periods of evaluation at ages 8, 9, 10, and 11. The response is recorded as 1 for symptoms and 0 for no symptoms. Explanatory variables included age, city, and a passive smoking index with values 0, 1, and 2 that reflected the degree of smoking in the home.

Note that age and the passive smoking index are time-dependent explanatory variables; their values depend on the period of measurement. This example provides a basic introduction to fitting GEE models with the GENMOD procedure. The dichotomous outcome is modeled with ...