Hypothesis Testing

This chapter is one of the cornerstones of statistics because it allows us to reach a decision based on an experiment with random outcomes. The basic question in an experiment is whether or not the outcome is real, or simply due to chance variation. For example, if we flip a coin 100 times and obtain 57 heads, can we conclude the coin is not fair? We know we expect 50 heads, so is 57 too many, or is it due to chance variation? Hypothesis testing allows us to answer such questions. This is an extremely important issue, for instance in drug trials in which we need to know if a drug truly is efficacious, or if the result of the clinical trial is simply due to chance, i.e., the possibility that a subject will simply improve or get worse on their own. The purpose of this chapter is to show how hypothesis testing is implemented.