Chapter 2

Very brief chapter overview

In Chapter 1, you learned about the scientific attitude, a key aspect of which is asking questions.

In this chapter, you will learn what types of questions to ask. Specifically, you learn about four questions to ask of any study:

  1. construct validity questions,
  2. internal validity questions,
  3. external validity questions, and
  4. ethical questions.

The rest of this course will be devoted to refining your skill at asking and answering questions about these four concepts.

Chapter Summary

 At one level, there are two basic problems about doing research to get answers to questions about human behavior.

  1. The study you do may be unethical.
  2. The study you do may not answer the question.
At another level, there is only one problem: Is the study ethical?
According to APA's ethical principles (which every researcher should consult before doing a study), a study is ethical if the potential benefits of the study outweigh the study's potential for harm. Thus, there are two ways to increase the chances that your study is ethical.
First, reduce the potential for harm. Following the nine guidelines in Box 2.1 (p. 59) can help reduce the potential for harm.
Second, make your study worth doing. This means
  1. Having an interesting, important research question; and
  2. Collecting data that will allow you to answer that question.
Mitchell and Jolley address point 1 (developing an interesting, important research question) in Chapter 3. But what about point 2 (collecting data that will allow you to answer your research question)? Obviously, your study should have validity. But what type of validity? The type of validity you will need will depend on your research question.

If your research question is about whether something causes a certain effect, your study must have internal validity.  Establishing internal validity is not easy (see Figure 2.1 on p. 48). Only studies that are experiments (and most studies are not experiments--see the colored table on p. 49) have internal validity.  Thus, if you want to make cause-effect statements, you should do an experiment. To do an experiment, you must have at treatment that you manipulate, and you must randomly assign participants to different types or amounts of that treatment. (Studies that are not experiments can only establish correlation--and one of the most repeated phrases in all of science is "correlation is not causation.")

Alternatively, if your research question is about what percentage of people do some behavior, you need a study that has external validity. One key to having external validity in survey research is to have a large, random, representative sample of participants. Random sampling from a population helps you to generalize your results to that larger population. Note that sample size (unless you sample is almost the entire population) has little to do with external validity: You could have a sample of a million people, but if they are not representative of your population, your large sample size would not help your external validity.

If your research question involves measuring or manipulating some state of mind (hunger, stress, fear, motivation, love, etc.), you need construct validity. As figures 2.2 (p.50) and 2.3 (p. 52) illustrate, achieving construct validity is not easy. On the contrary, establishing construct validity is the hardest type of validity to establish. Because establishing construct validity is so difficult, for much of psychology's existence, many psychologists have argued that psychology should avoid mental states entirely and focus only on observable stimuli and on observable behaviors.


Depending on the research question, you may often be interested in only one of these kinds of validity. Sometimes, you may want to have two of these kinds of validity. Rarely, however, will a study have high levels of all three types of validity.

Tip: Understanding the differences among the three types of validity takes some students a long time. To be one of the students who learns these key distinctions quickly, study Table 2-1 (p. 40) so that you can see how these types of validity relate to real life situations. Then, test your understanding by doing the student exercises and quizzes for this chapter.


Back to Chapter 2 Main Menu