Section 1.3 Design of Experiments 19 DEFINITION The outcome, or response, is what is measured on each experimental unit. In the wheat study, the outcome is the amount of wheat produced. DEFINITION The treatments are the procedures applied to each experimental unit. There are always two or more treatments. The purpose is to determine whether the choice of treatment affects the outcome. In the wheat study, the treatments are the three types of seed. In general, studies fall into two categories: randomized experiments and observational studies. DEFINITION A randomized experiment is a study in which the investigator assigns the treatments to the experimental units at random. The wheat study described above is a randomized experiment. In some situations, randomized experiments cannot be performed, because it isn’t possible to randomly assign the treatments. For example, in studies to determine how smoking affects health, people cannot be assigned to smoke. Instead, people choose for themselves whether to smoke, and scientists observe differences in health outcomes between groups of smokers and nonsmokers. Studies like this are called observational studies. DEFINITION An observational study is one in which the assignment to treatment groups is not made by the investigator. When possible, it is better to assign treatments at random and perform a randomized experiment. As we will see, the results of randomized experiments are generally easier to interpret than the results of observational studies. Objective 2 Understand the advantages of randomized experiments Randomized Experiments In July 2008, an article in The New England Journal of Medicine (359:339–354) reported the results of a study to determine whether a new drug called raltegravir is effective in reducing levels of virus in patients with human immunodeficiency virus (HIV). A total of 699 patients participated in the experiment. These patients were divided into two groups. One group was given raltegravir. The other group was given a placebo. (A placebo is a harmless tablet, such as a sugar tablet, that looks like the drug but has no medical effect.) Thus there were two treatments in this experiment, raltegravir and placebo. The experimenters had decided to give raltegravir to about two-thirds of the subjects and the placebo to the others. To determine which patients would be assigned to which group, a simple random sample consisting of 462 of the 699 patients was drawn; this sample constituted the raltegravir group. The remaining 237 patients were assigned to the placebo group. It was decided to examine subjects after 16 weeks and measure the levels of virus in their blood. Thus the outcome for this experiment was the number of copies of virus per milliliter of blood. Patients were considered to have a successful outcome if they had fewer than 50 copies of the virus per milliliter of blood. In the raltegravir group, 62% of the subjects had a successful outcome, but only 35% of the placebo group did. The conclusion was that raltegravir was effective in lowering the concentration of virus in HIV patients. We will examine this study, and determine why it was reasonable to reach this conclusion.
navidi_monk_essential_statistics_1e_ch1_3
To see the actual publication please follow the link above