ABSTRACT
The techniques for processing binary data presented in the previous chapter can be useful when the structure of the data is not particularly complex. In particular, when the data are from a number of similarly treated units and the aim is simply to estimate a proportion and its associated standard error, or when it is desired to compare the proportions of individuals in a number of groups who respond to some stimulus, the methods described in Chapter 2 may be suitable. But when, for example, a single binary response is recorded on individuals who differ from each other in terms of the values of certain explanatory variables, or when the responses are recorded on individuals who are grouped according to a number of different factors, the use of these relatively unsophisticated methods will generally be uninformative.
