If you are feeling particularly ambitious, you might think about the following:
Experiment 1
A group of psychologists want to determine whether scream therapy actually helps people to feel better. To test this, they recruit a number of subjects who have recently complained of extreme nervousness, and they carefully measure their anxiety using a series of psychological tests. They then invite each person to enter a special sound chamber and they encourage the person to scream as loudly as possible. After the subjects do so, the researchers administer the same tests again, and they discover that subjects say they feel much less anxious. What's wrong with this picture?
Answer: The experimenter failed to use a control group, so we cannot tell whether the decrease in anxiety is actually due to the "screaming therapy." It could be that people simply felt less anxious and scored "less anxious" on the test merely because they were taking it for the second time.
Experiment 2
A psychologist is trying to figure out whether males or females know more about political issues. The psychologist interviews people on the street asking them a series of questions. When he is done, he discovers that men know much more about politics than women do. What's wrong with this picture?
Answer: This survey does not use a random or representative sample. For example, it may be that males are, overall, less cooperative and the only men who stopped to talk are those who happened to be very knowledgeable about the topic. Perhaps the people who are walking down that particular sidewalk do not accurately represent the population at large.
Experiment 3
A psychologist is conducting an experiment to determine whether exposure to violence causes men to be more sympathetic to those who commit violent crimes. She randomly assigns twenty men to the experimental group, and they watch a violent film; she randomly assigns twenty men to the control group, and they view an exciting, nonviolent film of equivalent length. After the film, she gives a survey asking the subjects about their attitudes on various subjects, including legal penalties for violent crimes. She finds that the men who watched the violent film were more likely to show lenient attitudes toward violent offenders. What's wrong with this picture?
Answer: There are a number of problems with this study. First, subjects may be answering the survey in a certain way because they think the experimenter wants them to--a demand characteristic. Another problem is that the experimental group and control group may have differed in their opinion before the exposure to the film clips--we cannot tell without a pretest.
Introduction to Psychology : Read chapter 2, section 2.3.
Note: You are not responsible for the appendix to chapter 2.
