A fourth method, experimental research, is being used with increasing frequency to investigate human sexual behavior. Studies using this method, typically conducted in a laboratory environment, have a major advantage over other methods because they provide a controlled environment in which all possible influences on subjects’ responses, other than the factors that are being investigated, can be ruled out. A researcher using the experimental method manipulates a particular set of conditions, or variables, and observes the effect of this manipulation on subjects’ behavior or their physical or mental status. The experimental method is particularly suited to discovering causal relationships between variables.
There are two types of variables (behaviors or conditions that can have varied values) in any experimental research design: independent and dependent. An independent variable is a condition or component of the experiment that is under the control of the researcher, who manipulates or determines its value. Conversely, a dependent variable is an outcome or resulting behavior that the experimenter observes and records but does not control.
With this brief summary of the experimental method in mind, let us consider how this technique might clarify the relationship between sexually violent media and rape attitudes and behavior. A number of experimental research studies have provided compelling evidence that sexually violent media can cause attitudes to shift toward greater tolerance of sexually aggressive behavior and can contribute to some rapists’ assaultive behaviors. We consider three experiments, the first involving college men and the other two using convicted rapists as subjects.
The first study was conducted with 271 college men who were assigned to two groups. Subjects in the first group were exposed to movies with nonviolent sexual themes, whereas subjects in the second group saw R-rated films in which men were shown committing sexual violence against women who eventually experienced a transformation from victim to willing partner. A few days after viewing the movies, all subjects completed an attitude questionnaire. The results demonstrated that the men who viewed the violent films were generally much more accepting of sexual violence toward
Sex research: Methods and problems
women than were those subjects who were exposed to movies with consensual, nonvio-
Critical Thinking Question lent erotic themes (Malamuth & Check, 1981).
Two other research studies, with comparable research designs, compared the erectile What are the independent and dependent responses (dependent variable) of matched groups of rapists and nonrapists to two dif-
variables in this study? (See p. 48 for the ferent taped descriptions of sexual activity (independent variable)—one involving rape
answer.) and the other, mutually consenting sexual activity (Abel et al., 1977; Barbaree et al.,
1979). While subjects listened to the tapes, penile tumescence (engorgement) was measured with a penile strain gauge, which is described in the next section of this chapter. In both experiments rapists experienced erections while listening to violent descriptions of rape, whereas their nonrapist counterparts did not. Descriptions of consenting sexual activity produced similar levels of arousal in both groups of men. These findings suggest that exposure to sexually violent media not only encourages attitudes of violence toward women but also influences at least some men who rape to "sexualize" violence.
The experimental method has also been used to study the relationship between alcohol use and sexual responsiveness (although it has not been used to study vaginal orgasms). In one study of 48 male college students, a penile strain gauge was used to measure engorgement as subjects watched a sexually explicit film, first while not under the influence of alcohol and then several days later, after the subjects had consumed controlled amounts of alcohol. Findings showed that sexual arousal was reduced by drinking alcohol and that the more alcohol consumed, the greater the reduction (Brid — dell & Wilson, 1976). A similar experiment tested the relationship between arousal and Critical Thinking Question alcohol intake in women, with consistent results (Wilson & Lawson, 1976).
These studies illustrate one of the primary advantages of the experimental method. Because researchers can control variables precisely, they are able to draw conclusions about causal relationships to a degree not possible with other research methods. However, this method also has disadvantages. One of the most important limitations has to do with the artificiality of laboratory settings, which can adversely influence or bias subjects’ responses. As in direct-observation research, the fact that people know they are in an experiment can alter their responses from those that might occur outside the laboratory.
Not all experimental research in sexology is conducted within the relative artificiality of a laboratory setting. A landmark experimental study designed to test the hypothesis that circumcision may be an effective strategy for preventing HIV infection was conducted in South Africa. Investigators recruited over 3,000 uncircumcised and uninfected men and randomly assigned them to two groups: 50% who were circumcised at the onset of the clinical trial and the other 50% who were scheduled for circumcision at the end of a planned 21-month study. Participants were frequently tested for HIV infection during the course of the study. The investigation was halted after 18 months when it became clear that circumcision significantly reduced the risk of HIV infection (Auvert et al., 2005). This and other recent experimental studies, discussed in Chapter 15, have provided evidence that circum — Critical Thinking Question cision is one of many effective strategies for reducing the spread of HIV/AIDS.
Before concluding this chapter, we turn our attention to several additional areas of con — What are the independent and dependent cern regarding how we acquire information about sexual practices. First, we examine three
variables in this study? (See p. 48 for the technologies used in sex research. Next, we discuss ethical guidelines for conducting human
answer.) sex research. Finally, we end this chapter by describing a process for evaluating research.