August 2009
Intermediate to advanced
648 pages
17h 35m
English
Recall that the respiratory data analyzed in Section 15.6 contained an ordinal response that ranged from 0 for poor to 4 for excellent. (The responses were dichotomized in the previous analyses.) The proportional odds model provides a strategy that takes into account the ordinality of the data. See Chapter 9, “Logistic Regression II: Polytomous Response,” for a discussion of the proportional odds model in the univariate case, and refer to Lipsitz, Kim, and Zhao (1994) and Miller, Davis, and Landis (1993) for discussions on fitting the proportional odds model with GEE.
The following statements request a proportional odds model to be fit with the GEE method. The SAS data set RESP is the same as created ...
Read now
Unlock full access