Logistic Regression: A Self-learning Text, Third Edition (Statistics in the Health Sciences)

Objectives Upon completing this chapter, the learner should be able
to:

1. State or recognize examples of correlated responses.


2. State or recognize when the use of correlated
   analysis techniques may be appropriate.


3. State or recognize examples of different correlation
   structures that may be used in a GEE model.


4. Given a printout of the results of a GEE model:

i. State the formula and compute the estimated

odds ratio

ii. State the formula and compute a confidence

interval for the odds ratio

iii. Test hypotheses about the model parameters

using the Wald test, generalized Wald test, or

Score test, stating the null hypothesis and the

distribution of the test statistic, and

corresponding degrees of freedom under the

null hypothesis

5. Recognize how running a GEE model differs from
   running a standard logistic regression on data with
   correlated dichotomous responses.


6. Recognize the similarities in obtaining and
   interpreting odds ratio estimates using a GEE model
   compared with a standard logistic regression model.

Objectives 541