Sample Size and Analytical Issues

Abstract

In this chapter we consider some basic statistical issues around trial design and analysis. Most trials, particularly large and well conducted trials with little or no attrition, require only the simplest statistical methods, which usually give similar, if not the same, results as more complicated approaches. We do not look here in detail at methods of analysing trial data. Rather, the reader is directed to other books that focus specifically on statistical methods (e.g., Bland, 2000; Altman, 1991). We focus instead on a few important statistical design issues that are important to trial design, and give an overview of the main statistical issues of trial design and analysis.