August 2009
Intermediate to advanced
648 pages
17h 35m
English
Frequently, the underlying contingency table is based on more than two factors. This section discusses how to build models in a multifactor framework, how to specify scores for the response variable, and how to construct contrast tests with the CATMOD procedure. In addition, the interactive use of PROC CATMOD is explained.
Model building for more complicated cross-classification structures follows a similar strategy to that illustrated in the analysis of the colds data set. Consider the following data from a randomized clinical trial of chronic pain. Investigators were interested in comparing an active treatment with a placebo for an aspect of the condition of patients in the study. These patients ...
Read now
Unlock full access