Determining Sample Sizes Needed to Detect a Difference between Two Proportions

Abstract

There are two kinds of errors one must guard against in designing a comparative study. Even though these errors can occur in any statistical evaluation, their discussion here is restricted to the case where proportions from two independent samples are compared, that is, to sampling methods II and III. We present some means of specifying an important difference between proportions and provide mathematical results that will aide the investigator in finding the desired sample sizes. The later sections are devoted to the case where unequal sample sizes are planned for beforehand, and we present a discussion of some additional uses of the chapter's tables, including detectable effect sizes. A problem solving section appears at the end of the chapter.