16.1 Introduction

The chi-square test is used to compare two independent binomial proportions, p₁ and p₂. In the analysis of clinical data, the binomial proportion typically represents a response rate, cure rate, survival rate, abnormality rate, or some other 'event' rate as introduced in the previous chapter. Often, you want to compare such 'response' rates between a treated group and a parallel control group.

The chi-square test is an approximate test, which may be used when the normal approximation to the binomial distribution is valid (see Chapter 15). A popular alternative, Fisher's exact test (Chapter 17), is based on exact probabilities and it is often applied when conditions for using the chi-square test are not met (see Section 16.4.3 for details).