[ Chapter 1 Supplement ] [ Chapter Slides ] [ Know for Test ] [ Review Test ] [ Chapter Notes ] [Chapter Summary]
Introduction
The goals of psychology are to describe , to make predictions about the conditions that gave rise to them, and then to use that knowledge to control events in the future. What makes psychology a science? Psychologists share a basic assumption with all other sciences: Physical and psychological events have causes that can be uncovered through scientific investigation. Scientists do not rely on guesswork, hunches, or unsystematic collections of personal experiences. Rather, they employ a system of investigation known as the scientific method. Like all scientists, psychologists begin their work by making very careful and precise observations of different phenomena. They then use the information they have obtained to generate explanations for the phenomena that they have observed, which we call theories. From these theories psychologists develop hypotheses, predictions about future behaviors. They then test these hypotheses through more research and observation.
How do psychologists collect the data they need to develop theories and test hypotheses? The answer is that psychologists use a number of research methods, including case studies, naturalistic observations, and experiments. Each of these methods has strengths and weaknesses, and all of them contribute to our knowledge of claims and events. One topic that psychologists have used all these methods to understand better is violence. Let's look more carefully at this research to see how each method contributes to our understanding of violence.
The Case Study
The case study (sometimes called a clinical study) is an in-depth study of a single individual. This method was used in the 1800s by the physician Paul Broca, who discovered the brain area responsible for the language difficulties experienced by one of his patients. It was subsequently popularized by Sigmund Freud and a Swiss scientist, Jean Piaget (see Chapter 10). A major advantage of the case study is that concentrating on one individual (or sometimes a few) allows researchers to gather a great deal of detailed information. The goal of a case study is to use the information obtained from one person to understand the behaviors of others. The case study is often an excellent source for suggesting research ideas that can be explored with other methods. One potential disadvantage of the method is that what we learn by studying one individual may not apply to other individuals. Findings that have limited applicability are said not to generalize.
With this caution in mind, let's see how the case study might be used to help us understand violence. Few examples of violence are more extreme than the story of Ted Bundy. During the 1970s Bundy committed a series of murders that stretched from the Pacific Northwest through Colorado to Florida. Bundy has been described as charming, clean-cut, intelligent, and resourceful. He relentlessly stalked his victims, murdered them, and then disposed of their bodies. To reduce his chances of being caught, he carefully studied reports of police investigations so that he could anticipate every move law enforcement officials might make to apprehend him. His reign of terror ended when he was executed by the state of Florida in January 1989. How could anyone, especially someone described as an "all-American boy," have been capable of sexually assaulting, killing, and (in some cases) mutilating 30 women? Perhaps his own words provide clues to his behavior and that of others who have committed similar acts.
"There are no family backgrounds or personality types or life-styles that are, themselves, predictors," Bundy said (Michaud & Aynesworth, 1989, p. 61), but escapist behaviors such as drinking or watching television could influence his violent type of behavior. He also pointed to "the effect of pornography, for example, not suggestive but more explicit or expressed forms of pornography that are in print or on film" (p. 63) as a possible cause for his violence.
Naturalistic Observation
In contrast to the casual, informal observations of behavior that each of us makes every day, scientists make formal, recorded observations of events. The goal of such naturalistic observation is to describe the settings, frequency, and characteristics of certain behaviors. For example, psychologists interested in the use of seat belts have stationed themselves at the exits of shopping malls to see how many drivers used them. They have also observed whether children riding in the cars used seat belts.
When psychologists make naturalistic observations, they observe behaviors as they occur, without intervening or altering the behaviors in any way. Such observations usually involve more than one individual. The observers must be careful not to affect the behaviors they observe and record. Observations that interfere with the behavior being studied are termed reactive. If you have ever noticed someone in a restaurant watching you, you are familiar with a reactive observation. The observer's scrutiny may have changed your behavior in several ways. For example, you may have checked to see if you had food on your clothing or face, or you may have made an effort to use your best table manners. Psychologists who make naturalistic observations try to make their observations as nonreactive as possible. They try to "blend with the surroundings" so that they are not noticed by the persons being observed. For example, if you wished to study the behavior of young children, you might take a job at a day-care center where your job brings you naturally into contact with the individuals you wish to study.
Survey Method
Our interest in understanding violence leads us to ask: How much violence exists in our society? We could answer this question by asking every person in the country whether he or she had been the victim of a violent crime (such as robbery or assault) during the past year, but such an approach would be extremely expensive and impractical.
Researchers ask residents age 12 or older in 48,000 housing units a series of questions about their experiences with crimes in the previous year. This sample is extremely large, and researchers usually can conduct such large studies only with the resources available to the federal government. There were 43 million victimizations (some people are victims of more than one crime) during 1993. Of these crimes, 11 million were crimes against persons, and 32 million were crimes against property. Psychologists and other social scientists devised the survey method of research to gather data from a sample that represents a larger population, much as a sip of wine tells the story of the bottle (Crossen, 1994). Surveys are often used because they can be efficient ways to collect large amounts of information. When conducting a survey, researchers must obtain a representative sample--one that is selected to reflect the characteristics of a larger group (the population). The researcher hopes that the sample is a miniature version of the population. You see the survey method applied every November, when news anchors announce election winners with only 2 percent of the vote tabulated.
How can anyone make a prediction based on 2 percent (or even less) of the vote? News organizations can make these predictions because they have identified key areas within the state that represent the entire population; thus a small but representative segment of the population can be used to predict the way the entire population voted.
Obtaining a representative sample is not the only important consideration when conducting surveys. Questions must be carefully worded to elicit meaningful and useful responses. Consider a survey of the number of headaches reported by respondents. When the question was phrased "Do you get headaches frequently, and, if so, how often?" the average number of reported headaches was 2.2 per week. When the word occasionally was substituted for the word frequently, the average number of headaches reported was only 0.7 per week (Loftus, 1975). The respondents' answers may also depend on whether researchers include a "Don't know" option (Schuman & Presser, 1981). The absence of a "Don't know" response seems to create subtle pressure to offer opinions, even on fictitious issues (Bishop, Tuchfarber, & Oldendick, 1986). Survey questions are sometimes slanted in ways that invite biased results. For example, in 1993, H. Ross Perot and his United We Stand organization conducted a nationwide survey.
Skeptical of the results, Time and CNN asked Yankelovich Partners, a professional survey company, to conduct two surveys of random samples of the U.S. population. One sample was given Perot's version of the questions; the second sample was asked similar questions that had been rewritten to reduce the potential for bias. The Perot version read, "Should laws be passed to eliminate all possibilities of special interests giving huge sums of money to candidates?" The Yankelovich version was, "Should laws be passed to prohibit interest groups from contributing to campaigns they support?" The results were eye-opening: To the Perot version, 80 percent of respondents said yes and 17 percent no. In the Yankelovich survey 40 percent favored prohibition while 55 percent supported the right to contribute (Moore & Parker, 1995).
The Experimental Method
The research methods described so far can provide useful leads and data in our quest to gain a greater understanding of violent behavior. Yet those methods cannot provide us with cause-and-effect statements. In contrast, the experimental method can provide such statements. Therefore, it is considered the most powerful research method.
The logic of the experimental method starts with an educated guess or hypothesis about which variable or variables cause the behavior under consideration. For example, consider what variables could conceivably affect violent behavior. Some possibilities are crowding, frustration, and hot weather. Each of these variables could affect the probability that an act of violence might occur.
In the logic of an experiment, such variables are called independent variables. The psychologist's goal is to manipulate one or more independent variables to determine the effect on a dependent variable--a behavior that is the outcome of an experiment. Researchers are careful to offer clear and precise definitions for both the independent and dependent variables. Such definitions, known as operational definitions, allow other researchers to replicate (repeat) an experiment exactly as it was originally done in order to verify the findings.
In one simple case, some participants in the experiment are exposed to the independent variable; they constitute the experimental group. Other participants are not exposed to the independent variable; they constitute the control group that will be compared to the experimental group on he dependent variable. In the study of violence, hitting a person could be a dependent variable; the number of hits might change if we manipulated an independent variable that actually affects the probability of violence, such as observing an aggressive model. If our independent variable did have an effect on violence, the value of the dependent variable (number of hits) exhibited by the control group and the experimental group would be quite different. Albert Bandura and his colleagues (Bandura, Ross, & Ross, 1963) conducted an experiment to determine whether children learn aggressive behaviors by observing the actions of others. Their hypothesis was that children who observed an adult behaving aggressively would be more likely to exhibit aggressive actions. Nursery school children were assigned to two groups; one group observed an aggressive adult model, and the other group observed a non aggressive model. The independent variable in this experiment was observing an aggressive or a non aggressive model. Later, all the children were given an opportunity to hit a Bobo doll ; therefore, the dependent variable was the number of blows directed at the Bobo doll.
Critical Thinking and the selection of experimental and control groups
Suppose that Bandura and his colleagues had assigned all the boys to the group that observed the aggressive model and all the girls to the group that observed the non aggressive model. Could they conclude that boys were more aggressive because they observed an aggressive model? Give this question some thought, and write down your response before reading further If all the boys had been assigned to the group that observed an aggressive model, we could not conclude that the aggressive model was responsible for later aggression. Boys might be more aggressive than girls regardless of the kind of model they observed. The logic of the experimental method requires that the only difference between groups be the independent variable (or variables) manipulated by the experimenter--in this case, the type of model observed. If you hold all other variables that could also influence the results of the experiment--known as extraneous variables--constant, you may identify the cause of the behavior under consideration. If all the boys were in the aggressive -model group, there would be two possible explanations for any increased aggression: being a boy and observing an aggressive model.
To conduct the experiment groups were selected that are as much alike as possible before the experiment begins. One way to accomplish this objective is to use the procedure called random assignment of subjects to groups. Random assignment usually results in two groups that are quite similar in many characteristics. In this case, we would probably have about the same numbers of boys and girls in each of the two groups.
Bandura found that children who observed an aggressive model engaged in more aggressive behavior than those who observed a non aggressive model. The independent variable (observing an aggressive model) led to a higher rate of aggression (hitting a Bobo doll)--the dependent variable. The research by Bandura and his colleagues demonstrated that modeling can play a part in causing children to act aggressively.
You will see these research methods applied to answer questions about a wide range of behaviors in the coming chapters. Remember that each method can make a contribution to our understanding and that the methods are often used in combination. However, research is not conducted in a vacuum; psychologists adhere to a code of ethical behavior.
Research Ethics
Imagine being a subject in Milgram's 1974 study (described at the beginning of the chapter) and being asked to administer 450 volts of electric shock to another person. How would you react if you administered the shock? Later you are told that no shocks were actually delivered, but you realize that if they had been, you might have killed someone. This example highlights a major concern for researchers: how to conduct research in an ethical manner.
The American Psychological Association (1992) has adopted ethical guidelines that prescribe standards of conduct for the professional work of psychologists in their roles as researchers, clinicians, and teachers. These guidelines include several general principles. For example, psychologists must maintain high standards of competence in their work, including recognizing the limitations of their expertise. In their work, psychologists also must show respect for the rights and dignity of people, such as individuals' rights to privacy and confidentiality.
The ethical guidelines for the conduct of research require that all research proposals be reviewed to ensure compliance with the guidelines. Each proposal must be approved by an institutional review committee established by a college, university, or other organization where research is conducted.
Moreover, the ethical guidelines state that psychologists who conduct research using human participants must ensure that they are protected from physical and psychological harm. For this reason, the study conducted by Milgram several decades ago would not be permitted today. Although no shocks were actually given, the study put the participants in a situation that involved obeying the experimenter despite the moral imperative that they not hurt another person. This type of conflict is certainly unpleasant and could cause psychological harm to some people.
Today, psychologists are very careful to follow the ethical guidelines for research. Consider the work of two psychologists who were studying the relationship of handedness and life expectancy. Diane Halpern and Stanley Coren "did not want to appear to be ghouls or to be insensitive to people's feelings when they were grieving over the death of a loved one" (Coren, 1992, p. 216). First, they consulted a bereavement counselor who suggested that they not contact anyone unless at least 9 months had passed since the death. The counselor also suggested that they make the contact as gentle as possible and not follow up or press people for responses. The researchers decided that they would not contact next of kin if the death had been a result of murder or suicide or if the deceased was a child age 6 or younger. Their research proposal was evaluated and cleared by a university research ethics committee.
Despite their precautions, a few people they contacted did become upset by questions about deceased loved ones (Coren, 1992). The results of this study show that being left-handed was associated with reduced longevity (Coren, 1992; Coren & Halpern, 1991). Left-handed people died at an earlier age (mean of 66 years) than right-handed people (mean of 75 years). Other researchers have not found handedness to be associated with earlier death (Harris, 1993; Salive, Guralnik, & Glynn, 1993), although thesefindings have been disputed (Halpern & Coren, 1993). Study of the possible relationship, if any, between handedness and death continues; the research is conducted in accordance with established ethical principles designed to protect participants (in this case, the relatives of deceased persons) from any harm.
The ethical guidelines also require that any research records associated with a person's name or identity be kept confidential. For example, Halpern and Coren "kept no records as to whom we contacted, and there was no identifying information on any of the materials that we sent" (D. F. Halpern, personal communication, August 3, 1992). In other cases, psychologists may use code names or numbers for their participants so that information cannot be associated with the actual names of individuals who have taken part in research.
Participants, including college students in introductory courses, must be told that their participation in research is voluntary. They cannot be coerced into participating. College students may be part of a pool from which researchers draw participants for their studies. If a student objects to participating, he or she must be given an opportunity to select other ways to earn the same amount of credit or to complete the course requirement. In other words, no one may be punished for not participating in research. To enable potential participants to make proper judgments about their participation, the researcher must describe the procedures of the experiment and obtain a written agreement to participate, which is called an informed consent. This document indicates that the participant knows the nature of the research and what he or she has agreed to do before participating.
Some experiments require the use of deception. For example, researchers would be unable to study the true effectiveness of drug treatments without the use of placebos. Participants who are given placebos are led to believe that they are taking the actual drug so that researchers can assess the influence of expectations on the drug's effectiveness. When psychologists use deception in their research, they must debrief the participants immediately after the study.
During the debriefing, a complete explanation of the deception is given, participants are allowed to ask questions, and the researcher checks for any possible negative aftereffects of the deception. In addition, subjects in psychological research have the right to end their participation at any time.
Researchers cannot use any form of threat or intimidation to force them to complete tasks in a study. The Ethics of Research with Animals. As you will see throughout this book, many of psychology's most enduring findings resulted from research with animals (Domjan & Purdy, 1995). For example, much of our knowledge about the structure and functioning of the brain , basic processes of learning and conditioning, and maladaptive behavior is the result of animal research. Yet the use of animals in psychological research has not gone unchallenged.
The ongoing debate between advocates of using animals in research and those who deplore such efforts has been heated and emotional at times (Galvin & Herzog, 1992). Animal rights activists view the use of animals in psychological and medical research as cruel and unnecessary. Moreover, they note that the research often involves stress, pain, punishment, or social and environmental deprivation (Bowd & Shapiro, 1993; Rollin, 1985). On the other hand, the value of animal research has been defended by arguing that such research has led to improvements in human welfare (Baldwin, 1993; Miller, 1985). Psychological and medical researchers point out that some types of research could not be conducted without using animals as subjects. The American Psychological Association (1992) has adopted a set of seven principles for the ethical treatment of animals, encompassing consideration of the following:
- 1. Justification for the research to be conducted: Is this an important piece of research? Have other procedures, not involving the use of animals, been considered?
- 2. Qualifications of the personnel involved: Has the head of the research team received proper training in animal experimentation?
- 3. Care and housing of the animals: Will the animals be cared for appropriately, and will their quarters allow appropriate exercise?
- 4. Acquisition of the animals: Will the animals be acquired from a licensed, approved supplier?
- 5. The experimental procedures to be used.
- 6. The conduct of field research: Will the study of animals in the natural setting disturb their habitat?
- 7. The educational use of animals.
Researchers who use animals in their research are subject to a long list of regulations, including local, state, and federal laws. These regulations seek to define under what circumstances it is acceptable to sacrifice animals for research. Moreover, the regulations often mandate certain requirements concerning food, cage space, and veterinary care. A number of factors influence individuals' judgments about the ethics of using animals in research. These factors are complex and include the similarity of the animals to humans, their "cuteness," and the perceived importance of the research to alleviating human suffering (Herzog, 1990).
The influence of the animal rights movement is evident. A recent decline in the number of studies using certain animal species such as cats, dogs, and rabbits may be due, in part, to the attention that the animal rights movement has focused on this issue (Viney, King, & Berndt, 1990). Moreover, researchers are now looking for ways to reduce the number of animals used in research. For example, they are investigating alternatives to the use of animals for testing potentially toxic chemicals. The new methods do not involve intact higher animals; instead, they rely on bacteria, cultured animal cells, or fertilized chicken eggs (Goldberg & Frazier, 1989). However, such methods do not eliminate the need for animals in certain types of research. Opinions on the use of animals in research will no doubt continue to be highly personal, emotional, and strong.
Copyright 1997 by Prentice-Hall, Inc. All rights reserved.
[ Chapter 1 Supplement ] [ Chapter Slides ] [ Know for Test ] [ Review Test ] [ Chapter Notes ] [Chapter Summary]