The goal of much research in social psychology is to understand the causal relationships among variables, and for this we use experiments. Experimental research designs are research designs that include the manipulation of a given situation or experience for two or more groups of individuals who are initially created to be equivalent, followed by a measurement of the effect of that experience.
In an experimental research design, the variables of interest are called the independent variables and the dependent variables. The independent variable refers to the situation that is created by the experimenter through the experimental manipulations, and the dependent variable refers to the variable that is measured after the manipulations have occurred. In an experimental research design, the research hypothesis is that the manipulated independent variable (or variables) causes changes in the measured dependent variable (or variables). We can diagram the prediction like this, using an arrow that points in one direction to demonstrate the expected direction of causality:
viewing violence (independent variable) → aggressive behavior (dependent variable)
Consider an experiment conducted by Anderson and Dill (2000), which was designed to directly test the hypothesis that viewing violent video games would cause increased aggressive behavior. In this research, male and female undergraduates from Iowa State University were given a chance to play either a violent video game (Wolfenstein 3D) or a nonviolent video game (Myst). During the experimental session, the participants played the video game that they had been given for 15 minutes. Then, after the play, they participated in a competitive task with another student in which they had a chance to deliver blasts of white noise through the earphones of their opponent. The operational definition of the dependent variable (aggressive behavior) was the level and duration of noise delivered to the opponent. The design and the results of the experiment are shown in Figure 1.14.
Experimental designs have two very nice features. For one, they guarantee that the independent variable occurs prior to measuring the dependent variable. This eliminates the possibility of reverse causation. Second, the experimental manipulation allows ruling out the possibility of common-causal variables that cause both the independent variable and the dependent variable. In experimental designs, the influence of common-causal variables is controlled, and thus eliminated, by creating equivalence among the participants in each of the experimental conditions before the manipulation occurs.
The most common method of creating equivalence among the experimental conditions is through random assignment to conditions before the experiment begins, which involves determining separately for each participant which condition he or she will experience through a random process, such as drawing numbers out of an envelope or using a website such as http://randomizer.org. Anderson and Dill first randomly assigned about 100 participants to each of their two groups. Let’s call them Group A and Group B. Because they used random assignment to conditions, they could be confident that before the experimental manipulation occurred, the students in Group A were, on average, equivalent to the students in Group B on every possible variable, including variables that are likely to be related to aggression, such as family, peers, hormone levels, and diet—and, in fact, everything else.
Then, after they had created initial equivalence, Anderson and Dill created the experimental manipulation—they had the participants in Group A play the violent video game and the participants in Group B play the nonviolent video game. Then they compared the dependent variable (the white noise blasts) between the two groups and found that the students who had viewed the violent video game gave significantly longer noise blasts than did the students who had played the nonviolent game. When the researchers observed differences in the duration of white noise blasts between the two groups after the experimental manipulation, they could draw the conclusion that it was the independent variable (and not some other variable) that caused these differences because they had created initial equivalence between the groups. The idea is that the only thing that was different between the students in the two groups was which video game they had played.
When we create a situation in which the groups of participants are expected to be equivalent before the experiment begins, when we manipulate the independent variable before we measure the dependent variable, and when we change only the nature of independent variables between the conditions, then we can be confident that it is the independent variable that caused the differences in the dependent variable. Such experiments are said to have high internal validity, where internal validity is the extent to which changes in the dependent variable in an experiment can confidently be attributed to changes in the independent variable.
Despite the advantage of determining causation, experimental research designs do have limitations. One is that the experiments are usually conducted in laboratory situations rather than in the everyday lives of people. Therefore, we do not know whether results that we find in a laboratory setting will necessarily hold up in everyday life. To counter this, researchers sometimes conduct field experiments, which are experimental research studies that are conducted in a natural environment, such as a school or a factory. However, they are difficult to conduct because they require a means of creating random assignment to conditions, and this is frequently not possible in natural settings.
A second and perhaps more important limitation of experimental research designs is that some of the most interesting and important social variables cannot be experimentally manipulated. If we want to study the influence of the size of a mob on the destructiveness of its behavior, or to compare the personality characteristics of people who join suicide cults with those of people who do not join suicide cults, these relationships must be assessed using correlational designs because it is simply not possible to manipulate mob size or cult membership.