Introduction In this chapter, the logistic model is extended to handle
outcome variables that have dichotomous correlated
responses. The analytic approach presented for modeling
this type of data is the generalized estimating equations
(GEE) model, which takes into account the correlated
nature of the responses. If such correlations are ignored
in the modeling process, then incorrect inferences may
result.
The form of the GEE model and its interpretation are
developed. A variety of correlation structures that are
used in the formulation of the model are described. An
overview of the mathematical foundation for the GEE
approach is also presented, including discussions of
generalized linear models, score equations, and “score-
like” equations. In the next chapter (Chap. 12), examples
are presented to illustrate the application and interpreta-
tion of GEE models. The final chapter in the text (Chap. 13)
describes alternate approaches for the analysis of corre-
lated data.Abbreviated
Outline
The outline below gives the user a preview of the material
to be covered by the presentation. A detailed outline for
review purposes follows the presentation.I. Overview (pages 492–493)
II. An example (Infant Care Study) (pages 493–498)
III. Data layout (page 499)
IV. Covariance and correlation (pages 500–502)
V. Generalized linear models (pages 503–506)
VI. GEE models (pages 506–507)
VII. Correlation structure (pages 507–510)
VIII. Different types of correlation structure (pages
511–516)
IX. Empirical and model-based variance estimators
(pages 516–519)
X. Statistical tests (pages 519–520)
XI. Score equations and “score-like” equations
(pages 521–523)
XII. Generalizing the “score-like” equations to form
GEE models (pages 524–528)
XIII. Summary (page 528)490 14. Logistic Regression for Correlated Data: GEE