V METHODS OF RESEARCH

V

METHODS OF RESEARCH

Psychologists from all areas of specialization use the scientific method to test their theories about behavior and mental processes. A theory is an organized set of principles that is designed to explain and predict some phenomenon. Good theories also provide specific testable predictions, or hypotheses, about the relation between two or more variables. Formulating a hypothesis to be tested is the first important step in conducting research.

Over the years, psychologists have devised numerous ways to test their hypotheses and theories. Many studies are conducted in a laboratory, usually located at a university. The laboratory setting allows researchers to control what happens to their subjects and make careful and precise observations of behavior. For example, a psychologist who studies memory can bring volunteers into the lab, ask them to memorize a list of words or pictures, and then test their recall of that material seconds, minutes, or days later.

As indicated by the term field research, studies may also be conducted in real-world locations. For example, a psychologist investigating the reliability of eyewitness testimony might stage phony crimes in the street and then ask unsuspecting bystanders to identify the culprit from a set of photographs. Psychologists observe people in a wide variety of other locations outside the laboratory, including classrooms, offices, hospitals, college dormitories, bars, restaurants, and prisons.

In both laboratory and field settings, psychologists conduct their research using a variety of methods. Among the most common methods are archival studies, case studies, surveys, naturalistic observations, correlational studies, experiments, literature reviews, and measures of brain activity.

A

Archival Studies

One way to learn about people is through archival studies, an examination of existing records of human activities. Psychological researchers often examine old newspaper stories, medical records, birth certificates, crime reports, popular books, and artwork. They may also examine statistical trends of the past, such as crime rates, birth rates, marriage and divorce rates, and employment rates. The strength of such measures is that by observing people only secondhand, researchers cannot unwittingly influence the subjects by their presence. However, available records of human activity are not always complete or detailed enough to be useful.

Archival studies are particularly valuable for examining cultural or historical trends. For example, in one study of physical attractiveness, researchers wanted to know if American standards of female beauty have changed over several generations. These researchers looked through two popular women’s magazines between 1901 and 1981 and examined the measurements of the female models. They found that “curvaceousness” (as measured by the bust-to-waist ratio) varied over time, with a boyish, slender look considered desirable in some time periods but not in others.

B

Case Studies

Sometimes psychologists interview, test, observe, and investigate the backgrounds of specific individuals in detail. Such case studies are conducted when researchers believe that an in-depth look at one individual will reveal something important about people in general.

Case studies often take a great deal of time to complete, and the results may be limited by the fact that the subject is atypical. Yet case studies have played a prominent role in the development of psychology. Austrian physician Sigmund Freud based his theory of psychoanalysis on his experiences with troubled patients. Swiss psychologist Jean Piaget first began to formulate a theory of intellectual development by questioning his own children. Neuroscientists learn about how the human brain works by testing patients who have suffered brain damage. Cognitive psychologists learn about human intelligence by studying child prodigies and other gifted individuals. Social psychologists learn about group decision making by analyzing the policy decisions of government and business groups. When an individual is exceptional in some way, or when a hypothesis can be tested only through intensive, long-term observation, the case study is a valuable method.

C

Measures of Brain Activity

Electroencephalography

An electroencephalogram, or EEG, is a recording of the action potential, or electrical, activity of the cerebral cortex of the brain. An EEG is made by attaching electrodes to the scalp, then collecting, amplifying, and recording the electrical impulses of the brain.

Biopsychologists interested in the links between brain and behavior use a variety of specialized techniques in their research. One approach is to observe and test patients who have suffered damage to a specific region of the brain to determine what mental functions and behaviors were affected by that damage. British-born neurologist Oliver Sacks has written several books in which he describes case studies of brain-damaged patients who exhibited specific deficits in their speech, memory, sleep, and even in their personalities.

Positron Emission Tomography

This positron emission tomography (PET) scan of the brain shows the activity of brain cells in the resting state and during three types of auditory stimulation. PET uses radioactive substances introduced into the brain to measure such brain functions as cerebral metabolism, blood flow and volume, oxygen use, and the formation of neurotransmitters. This imaging method collects data from many different angles, feeding the information into a computer that produces a series of cross-sectional images.

A second approach is to physically alter the brain and measure the effects of that change on behavior. The alteration can be achieved in different ways. For example, animal researchers often damage or destroy a specific region of a laboratory animal’s brain through surgery. Other researchers might spark or inhibit activity in the brain through the use of drugs or electrical stimulation.

Magnetic Resonance Imaging Scan

This magnetic resonance imaging (MRI) scan of a normal adult head shows the brain, airways, and soft tissues of the face. The large cerebral cortex, appearing in yellow and green, forms the bulk of the brain tissue; the circular cerebellum, center left, in red, and the elongated brainstem, center, in red, are also prominently shown.

Another way to study the relationship between the brain and behavior is to record the activity of the brain with machines while a subject engages in certain behaviors or activities. One such instrument is the electroencephalograph, a device that can detect, amplify, and record the level of electrical activity in the brain by means of metal electrodes taped to the scalp. See Electroencephalography.

Advances in technology in the early 1970s allowed psychologists to see inside the living human brain for the first time without physically cutting into it. Today, psychologists use a variety of sophisticated brain-imaging techniques. The computerized axial tomography (CT or CAT) scan provides a computer-enhanced X-ray image of the brain. The more advanced positron emission tomography (PET) scan tracks the level of activity in specific parts of the brain by measuring the amount of glucose being used there. These measurements are then fed to a computer, which produces a color-coded image of brain activity. Another technique is magnetic resonance imaging (MRI), which produces high-resolution cross-sectional images of the brain. A high-speed version of MRI known as functional MRI produces moving images of the brain as its activity changes in real time. These relatively new brain imaging techniques have generated great excitement, because they allow researchers to identify parts of the brain that are active while people read, speak, listen to music, solve math problems, and engage in other mental activities.

See Biopsychology: Methods of Research; Brain: Brain Imaging.

D

Surveys

In contrast with the in-depth study of one person, surveys describe a specific population or group of people. Surveys involve asking people a series of questions about their behaviors, thoughts, or opinions. Surveys can be conducted in person, over the phone, or through the mail. Most surveys study a specific group—for example, college students, working mothers, men, or homeowners. Rather than questioning every person in the group, survey researchers choose a representative sample of people and generalize the findings to the larger population.

Surveys may pertain to almost any topic. Often surveys ask people to report their feelings about various social and political issues, the TV shows they watch, or the consumer products they purchase. Surveys are also used to learn about people’s sexual practices; to estimate the use of cigarettes, alcohol, and other drugs; and to approximate the proportion of people who experience feelings of life satisfaction, loneliness, and other psychological states that cannot be directly observed.

Surveys must be carefully designed and conducted to ensure their accuracy. The results can be influenced, and biased, by two factors: who the respondents are and how the questions are asked. For a survey to be accurate, the sample being questioned must be representative of the population on key characteristics such as sex, race, age, region, and cultural background. To ensure similarity to the larger population, survey researchers usually try to make sure that they have a random sample, a method of selection in which everyone in the population has an equal chance of being chosen.

When the sample is not random, the results can be misleading. For example, prior to the 1936 United States presidential election, pollsters for the magazine Literary Digest mailed postcards to more than 10 million people who were listed in telephone directories or as registered owners of automobiles. The cards asked for whom they intended to vote. Based on the more than 2 million ballots that were returned, the Literary Digest predicted that Republican candidate Alfred M. Landon would win in a landslide over Democrat Franklin D. Roosevelt. At the time, however, more Republicans than Democrats owned telephones and automobiles, skewing the poll results. In the election, Landon won only two states.

The results of survey research can also be influenced by the way that questions are asked. For example, when asked about “welfare,” a majority of Americans in one survey said that the government spends too much money. But when asked about “assistance to the poor,” significantly fewer people gave this response.

E

Naturalistic Observations

Jane Goodall

British ethologist Jane Goodall pioneered the study of chimpanzees in the wild. This 1972 photo shows her at work at the Gombe Stream Research Center in Nigeria.

In naturalistic observation, the researcher observes people as they behave in the real world. The researcher simply records what occurs and does not intervene in the situation. Psychologists use naturalistic observation to study the interactions between parents and children, doctors and patients, police and citizens, and managers and workers.

Naturalistic observation is common in anthropology, in which field workers seek to understand the everyday life of a culture. Ethologists, who study the behavior of animals in their natural habitat, also use this method. For example, British ethologist Jane Goodall spent many years in African jungles observing chimpanzees—their social structure, courting rituals, struggles for dominance, eating habits, and other behaviors. Naturalistic observation is also common among developmental psychologists who study social play, parent-child attachments, and other aspects of child development. These researchers observe children at home, in school, on the playground, and in other settings.

F

Correlational Studies

Case studies, surveys, and naturalistic observations are used to describe behavior. Correlational studies are further designed to find statistical connections, or correlations, between variables so that some factors can be used to predict others.

A correlation is a statistical measure of the extent to which two variables are associated. A positive correlation exists when two variables increase or decrease together. For example, frustration and aggression are positively correlated, meaning that as frustration rises, so do acts of aggression. More of one means more of the other. A negative correlation exists when increases in one variable are accompanied by decreases in the other, and vice versa. For example, friendships and stress-induced illness are negatively correlated, meaning that the more close friends a person has, the fewer stress-related illnesses the person suffers. More of one means less of the other.

Based on correlational evidence, researchers can use one variable to make predictions about another variable. But researchers must use caution when drawing conclusions from correlations. It is natural—but incorrect—to assume that because one variable predicts another, the first must have caused the second. For example, one might assume that frustration triggers aggression, or that friendships foster health. Regardless of how intuitive or accurate these conclusions may be, correlation does not prove causation. Thus, although it is possible that frustration causes aggression, there are other ways to interpret the correlation. For example, it is possible that aggressive people are more likely to suffer social rejection and become frustrated as a result.

G

Experiments

Correlations enable researchers to predict one variable from another. But to determine if one variable actually causes another, psychologists must conduct experiments. In an experiment, the psychologist manipulates one factor in a situation—keeping other aspects of the situation constant—and then observes the effect of the manipulation on behavior. The people whose behavior is being observed are the subjects of the experiment. The factor that an experimenter varies (the proposed cause) is known as the independent variable, and the behavior being measured (the proposed effect) is called the dependent variable. In a test of the hypothesis that frustration triggers aggression, frustration would be the independent variable, and aggression the dependent variable.

There are three requirements for conducting a valid scientific experiment: (1) control over the independent variable, (2) the use of a comparison group, and (3) the random assignment of subjects to conditions. In its most basic form, then, a typical experiment compares a large number of subjects who are randomly assigned to experience one condition with a group of similar subjects who are not. Those who experience the condition compose the experimental group, and those who do not make up the control group. If the two groups differ significantly in their behavior during the experiment, that difference can be attributed to the presence of the condition, or independent variable. For example, to test the hypothesis that frustration triggers aggression, one group of researchers brought subjects into a laboratory, impeded their efforts to complete an important task (other subjects in the experiment were not impeded), and measured their aggressiveness toward another person. These researchers found that subjects who had been frustrated were more aggressive than those who had not been frustrated.

Psychologists use many different methods in their research. Yet no single experiment can fully prove a hypothesis, so the science of psychology builds slowly over time. First, a new discovery must be replicated. Replication refers to the process of conducting a second, nearly identical study to see if the initial findings can be repeated. If so, then researchers try to determine if these findings can be applied, transferred, or generalized to other settings. Generalizability refers to the extent to which a finding obtained under one set of conditions can also be obtained at another time, in another place, and in other populations.

H

Literature Reviews

Because the science of psychology proceeds in small increments, many studies must be conducted before clear patterns emerge. To summarize and interpret an entire body of research, psychologists rely on two methods. One method is a narrative review of the literature, in which a reviewer subjectively evaluates the strengths and weaknesses of the various studies on a topic and argues for certain conclusions. Another method is meta-analysis, a statistical procedure used to combine the results from many different studies. By meta-analyzing a body of research, psychologists can often draw precise conclusions concerning the strength and breadth of support for a hypothesis.

I

Ethical Considerations in Research

Animal Experimentation

A researcher tests the effects of marijuana on a rhesus monkey’s eye. Supporters of animal rights regard most animal experimentation as unethical. Many scientists defend animal studies as necessary for advancing understanding of human diseases and improving the quality of human life.

Psychological research involving human subjects raises ethical concerns about the subject's right to privacy, the possible harm or discomfort caused by experimental procedures, and the use of deception. Over the years, psychologists have established various ethical guidelines. The American Psychological Association recommends that researchers (1) tell prospective subjects what they will experience so they can give informed consent to participate; (2) instruct subjects that they may withdraw from the study at any time; (3) minimize all harm and discomfort; (4) keep the subjects’ responses and behaviors confidential; and (5) debrief subjects who were deceived in some way by fully explaining the research after they have participated. Some psychologists argue that such rules should never be broken. Others say that some degree of flexibility is needed in order to study certain important issues, such as the effects of stress on test performance.

Laboratory experiments that use rats, mice, rabbits, pigeons, monkeys, and other animals are an important part of psychology, just as in medicine. Animal research serves three purposes in psychology: to learn more about certain types of animals, to discover general principles of behavior that pertain to all species, and to study variables that cannot ethically be tested with human beings. But is it ethical to experiment on animals?

Some animal rights activists believe that it is wrong to use animals in experiments, particularly in those that involve surgery, drugs, social isolation, food deprivation, electric shock, and other potentially harmful procedures. These activists see animal experimentation as unnecessary and question whether results from such research can be applied to humans. Many activists also argue that like humans, animals have the capacity to suffer and feel pain. In response to these criticisms, many researchers point out that animal experimentation has helped to improve the quality of human life. They note that animal studies have contributed to the treatment of anxiety, depression, and other mental disorders. Animal studies have also contributed to our understanding of conditions such as Alzheimer’s disease, obesity, alcoholism, and the effects of stress on the immune system. Most researchers follow strict ethical guidelines that require them to minimize pain and discomfort to animals and to use the least invasive procedures possible. In addition, federal animal-protection laws in the United States require researchers to provide humane care and housing of animals and to tend to the psychological well-being of primates used in research. See Animal Experimentation.