ABSTRACT
In this chapter we focus on binary data, which has been the topic of some recent
debate in the statistical literature. For example, Ziegler and Vens (2010); Breitung
et al. (2010); Molenberghs (2010), and Shults (2011) all focus on the special issues
related to binary data; these authors were specifically interested in the selection of
correlation structures for dichotomous outcomes (which we will turn to in the next
chapter), but they also touched on some of the special features of binary data that
complicate the use of GEE and QLS for their analysis.