ABSTRACT
Suppose that a researcher is interested in treatments to reduce hypertension. Consider a hypothetical study with four independent groups of subjects, each of whom is assigned randomly to one of the following treatments: drug therapy, biofeedback, dietary modification, or a treatment combining all aspects of the other treatments. For simplicity, suppose the dependent variable is a single blood pressure reading taken 2 weeks after the termination of treatment. In Chapter 3, you learned how to test an omnibus null hypothesis that all four treatments are equally effective. However, there are a number of other questions that might be addressed here, either in addition to or instead of the omnibus null hypothesis. Here are some examples of types of questions you will learn how to answer in Chapter 4:
Is there a difference in the effectiveness of drug therapy versus biofeedback? Drug therapy versus diet? Biofeedback versus diet?
Is the combination treatment more effective than any of the individual treatments? Is it more effective than the average of the individual treatments?
How small or large is the difference between drug therapy and biofeedback likely to be in the population?
How much of the between-group variance is explained by the difference between drug therapy and biofeedback?
