STUDY UNIT 2.11 Random Assignment Versus Random Sampling

It’s time to talk about the word random. Colloquially, random is used to say that something is odd or unexpected. Your friend might say, “I saw this random guy in the student center yesterday.” Or “I have a random question for you.”

Scientists, in contrast, use the term random to mean that something is selected or assigned without any bias, such that all cases have an equal chance. The word has already come up in two contexts in this chapter: random sampling and random assignment. Both are important concepts for research, but they have different meanings.

Random sampling is also known as random selection. It is a method for selecting the participants who will be in a study (see Study Unit 2.5). If a study uses random sampling, every person in the population of interest has an equal chance of being selected. You can picture random sampling as choosing people’s names out of a hat: All members of the population of interest have their names in a hat, and a researcher randomly selects some of the names to be in the study. A more modern way to achieve random sampling is by programming a computer to call potential participants using randomly selected phone number digits. When a study uses random sampling, its results can generalize to the population of interest. In contrast, if a study uses some other form of sampling, such as sampling only those who are convenient to reach, certain kinds of people are more likely to be chosen than others. Random sampling can be used in any study—a survey or observational study, a correlational study, or an experiment.

Random assignment is used only in experiments. It’s the method used to assign participants to the different levels of the independent variable. If a study uses random assignment, each person in the study has an equal chance of being in one experimental condition or another. Because random assignment helps ensure that the people in the different experimental groups are pretty much the same, it helps rule out alternative explanations for any difference between experimental groups. For example, because of random assignment, we can assume that a “high alcohol” group and an “expected alcohol” group have the same number of already-aggressive people, the same number of extroverted people, the same number of cheerful people, and so on. You can practice what you’ve learned about random sampling and random assignment in INTERACTIVE FIGURE 2.29.

Explore the Science