LEARNING OBJECTIVES

CHAPTER OVERVIEW

Chapters 8 and 9 covered two-sample t tests, tests that compare the means of two groups to see if they have a statistically significant difference. t tests are great for experiments with just two groups, like classic experiments with a control group and an experimental group. A t test would work well, for example, to see if a medication works better than a placebo in treating an illness. However, experiments that address complex questions often require more than two groups. Finding the best way to treat an illness might involve comparing three different medications and three different doses of each medication, ending up with nine different groups. The statistical technique that compares means when there are more than two groups is called analysis of variance.