Methods Of Studying Behavior
– Fouzia Khursheed Ahmad
Research Scholar (M.Phil- PhD)
NUEPA, New Delhi.
Behavior refers to the actions or reactions of an object or organism, usually in relation to the environment. Behavior can be conscious or unconscious, overt or covert, and voluntary or involuntary.
Behavior is something we ourselves do and something we experience from others. Behavior is sensuous in that what it is can be experienced through one or more of the senses. Lacking this sensational dimension, whatever the “it” is would not constitute behavior. Second, there are four basic categories. Behavior that we term conduct denotes a broad range of activity under the guidance of social-moral norms. Behavior that we term performance denotes a range of activity governed by skill repertoires. Behavior we sometimes term experience refers to actions and choices of action that maximize comfort or pleasure and minimize discomfort or pain. Finally, there is no particular other term for the very broad range of actions that fall under the heading of purposive or goal-driven, behavior that appears dictated by a sense of who we are and what we want to become.
If we were to examine the life records of thousands of individuals taken at random from the general population, we would find a common pattern running through the great majority. In their school years they were from slow to good students. A few of them completed college, but most stopped at an earlier level. As adults they entered a variety of skilled and unskilled occupations, exhibited satisfactory work capacity, and managed to earn adequate incomes. With respect to their social behavior, these individuals were, in the main, conformists. They played the usual childhood games, went to parties, fell in love, married, reared their families, and participated in the ordinary affairs of the community. They were capable of establishing satisfying relationships with other people, and appropriate. Some were more popular than others, but all had their circle of friends and were accepted members of society. Although they did not greatly advance the welfare of others, neither did they did jeopardize or interfere unduly with the security and happiness of their associates. They were essentially law-abiding citizens who respected and adhered to the rules and conventions of their culture.
In personality traits, naturally they differed from one another, but none were exceptionally excitable, reclusive, depressed, suspicious, dynamic, impressive, or otherwise very outstanding. They were the common varieties of human beings whose presence or absence in a crowd was not greatly noticed. Some led a richer or more intense emotional life than others and some were happier than others, but for the most part, their affective experiences were not unique. Apart from sporadic outbursts of an innocuous nature, they managed to control their emotions. Although they had their share of frustrations, conflicts, and hardships, their lives were not greatly disrupted by their misfortunes. During moments of stress, they proved to be fairly resilient and adaptable. Their inner mental life was, more often than not, one of tranquility. These common place men and women who exhibited at least ordinary competence in self-management and got along reasonably well with themselves and their associates constitute the normal, or average, group. The great majority of the general populations are normal people.
The Superior: – Scattered through the rank-and file histories, there would be found a smaller group of individuals who deviated from the normal in a favorable or healthy direction. In contrast to the mediocrity of the normal group, these individuals were definitely superior in intelligence, personality, social adjustment, and emotional stability. Since there is a tendency for good qualities to go together, many excelled in more than one sphere. These superior men and women not only led wholesome, successful lives themselves but often contributed to the welfare and happiness of others through their inventions, achievements, and personal influences.
The Abnormal: – Among the unselected life records, there would also be found a small group of equally spectacular and unusual cases that deviated from the normal in an unfavorable of pathological direction. Included in this abnormal group would be individuals marked by limited intelligence, emotional instability, personality disorganization, and character defects, who for the most part, led wretched personal lives and were social misfits or liabilities. These abnormal deviants, who constitute large percent of the general population, are usually classified into four main categories: Psychoneurotic, Psychotic, Mentally Defective, and Antisocial.
FORMS OF ABNORMALITY
Psychoneuroses:- Individuals who “go to piece” easily when confronted with a difficult or trying situation and exhibit a variety of mental and physical symptoms that persist for several weeks or months are known as “psychoneurotic”. Typical mental symptoms are anxiety, feelings of inner tension, restlessness ideas of inadequacy, inability to concentrate, loss of memory, absurd fears, and obsessions. Physical symptoms, which are essentially repercussions of internal emotional disturbance, include headaches, upset stomach, excessive fatigue, and loss of sensory and motor functions. Psychoneuroses are relatively mild personality disorders that distress and inconvenience the patient but do not disrupt his social adjustments or interfere with his everyday activities to the point of necessitating supervision or compulsory commitment to a mental hospital. His/her personality remains intact and his/her grasp of reality is not distorted. Psycho-neurotics know what they are doing, have a fair understanding of their difficulties, can distinguish right from wrong, and are legally responsible for their actions. Their behavior, though frequently annoying, is rarely offensive or a source of danger to others. Their work output may or may not be impaired.
When under great emotional stress, normal individuals frequently exhibit typical psychoneurotic symptoms but with two important modifications, namely, their reactions are appropriate to the stimulating situation and are of short duration. On the other hand, the reactions of the psychoneurotic are out of all proportion to the actual situation and may persist for months. A normal person who experiences a severe emotional shock may weak, or complains of irregular heart action or nausea. Soon however, he regains control of himself and his symptoms disappear. Following a similar or milder emotional shock, a psychoneurotic may suffer for months from loss of voice, paralysis, general exhaustion, cardiac instability, or gastric upset. Faced with failure, a normal individual may be beset with temporary anxiety and feelings of inferiority, but a psychoneurotic may retain this attitude in exaggerated from all his life. Many normal individuals have a fear of germs and take reasonable precaution to avoid infection, but, unlike certain psychoneurotic, they do not wash their hands a hundred times a day, wear gloves when handling money, or sterilize their cooking utensils before each meal.
Psychoses: – Psychoses are severe mental disorders that tend to shatter the integration of the personality and disrupt the individual’s social relationships. The behavior of the psychotic is too bizarre, unreasonable, and inappropriate to be understood by a normal person. It is necessary to supervise closely, or hospitalize, psychotic patients, because they are incapable of adequate self management and their peculiar and unpredictable actions constitute a potential threat to the welfare of others. Psychotic individuals are so unbalanced mentally that they are not legally responsible for their actions. In the eyes of the law, they are insane. In psychoses, normal inhibitions and cultural restraints are severed, and the patient indulges his whims and fantasies unchecked by rules of logic, common sense, or social pressure. The wish is father to the thought, and the thought is omnipotent. The psychotic has only to think that he is a multimillionaire or the beloved of some movie star, and it is so. A female patient claimed that she gave birth to a thousand babies a day. By special arrangement with God, they were transported to her direct from heaven by invisible parachutes. If a psychotic patient desires to die, he has only to say the word and he is dead. He may continue to eat three meals a day, but this inconsistency does not disconcert him. With complete sincerity he will tell you that he is merely an X-ray plate of his former self, that his real self is dead.
Unpleasant delusions and hallucinations are just as real. The patient who imagines that he is being persecuted by some secret organization is genuinely terror-stricken, and he flees from one city to another to evade his persecutors. When he is committed, their voices follow him to the hospital, and each day he spends hours in violent debate with these nonexistent voices. The guilt laden patient who beseeches her physician to throw her in the furnace is perfectly sincere and would make no protests if he complied with her request.
In their emotional reactions they show the same disregard for reality. Without any apparent cause, they become violently, excited, depressed, or irritable. There is no logical relation between the motivating situation and the emotional response. Sad news from home may evoke laughter; good news, tears; or either may have no effect. Usually the patient is confused, bewildered and disoriented. Speech is incoherent, and thought processes are retarded and ineffective. There is an inability to grasp new material, and memory disturbances are common. The final out or the individual may make a surprising recovery with few, if any, residual symptoms.
The gap separating the normal from the psychotic patient is far greater than that between the normal and the psychoneurotic. It is easy for so called ‘normal’ individuals to establish rapport with the psychoneurotic, to understand his symptoms, and to converse with him in normal fashion. On the other hand, so called ‘normal’ people are completely baffled and somewhat frightened by the mental confusion, incoherent speech, weird actions, distorted thoughts, emotional outbursts, and hallucinations of the psychotic. It is practically impossible for a normal individual to carry on an understand his symptoms, or even to sympathize with him. The explanation for this lack of rapport lies in the fact that the bonds connecting the psychotic individual with his fellow men have been cut. The psychotic lives not in the world of reality but in his own private world. He is divorced from his associates, and the rules, customs, and happenings of the real world have no meaning or significance for him. His behavior is impervious to outside influences, and attempts to hold his attention or to modify his thoughts by persuasion, reason, or force are futile. He is so effectively insulated against the outer world by his psychosis that even intimate contact with the horrors and dangers of civilian bombings, in wartime, fails to affect him.
Mental Deficiency:- Mental deficiency is a general category which includes a verity of individuals who, because of sub-normal or retarded mental development, are unable as children to profit from regular school instruction and as adults are incapable of adequate self-management or self-support. These individuals are also classified as aments or feeble-minded. Feeble-mindedness is differentiated from the mental deterioration that results from various psychoses by a deficiency of intelligence dating from birth or early life. Mental defectives are socioeconomic liabilities and responsibilities. The brightest can learn to read and to do simple arithmetic up to about the fifth-grade level if they are given intensive schooling until they are fifteen. As adults, they can be trained to do simple routine tasks that will contribute to their support, but they are incapable of adequate self-management in society. They can protect themselves against common physical dangers but are helpless when exposed to ordinary social dangers. If employed, they do not spend their earnings wisely. Their brighter associates find it easy to take advantage of them. Although not inherently delinquent or immoral, they are frequently guilty of petty crimes and sex offenses. They do not fully understand the social significance of their actions and must be carefully supervised. The dullest of the mental defectives never learn to walk, talk, or feed themselves. In adaptability to life situations they show less intelligence than animals, not even knowing enough to come in out of the rain. They would soon perish if they were not protected and cared for by others. Even as adults, they must be treated as helpless infants.
Antisocial Personalities: – Included in this category are two overlapping, but more or less independent, groups that share a common propensity for antisocial behavior. One group is made up of convicted law violators, and the other consists of individuals with psychopathic personalities.
Depending on their age, law violators are classified from a legal point of view as delinquents or criminals. A small proportion of criminals are mentally defective, but the great majorities possess average intelligence and some have superior mental ability. All personality types are found in the criminal group. Many first offenders who are convicted of accidental or culturally tolerated offenses, for example, traffic violations and bootlegging, have essentially normal personalities. Other criminals, in addition to being lawbreakers, suffer from psychoneuroses or psychoses. Among habitual criminals especially, there is frequently found a peculiar personality, termed a psychopathic personality. Although most individuals with psychopathic personalities are potential criminals, it is desirable to consider them as a separate group. Many of them manage to evade the law and hence are not strictly speaking criminals and most criminals do not have psychopathic personalities.
Individuals classified as “psychopathic personalities” possess adequate intelligence, and exhibit neither the conflicts nor anxieties of the psychoneurotic nor the delusions, hallucinations, and mental confusion of the psychotic. Many actually have very pleasing personalities. Their abnormality is manifested in a lack of inhibitions and an incapacity or unwillingness to conform to ethical and social standards.
They are impulsive, selfish, emotionally unstable individuals who, from early childhood, make a practice of immediately satisfying their desires and impulses without regard to the consequences or the means. Since they are unconcerned with future consequences, their judgment appears faulty and they fail to profit from past experiences. Devoid of moral and social scruples, they do not hesitate to deceive and exploit their associates or engage in antisocial activities if it is to their advantage to do so. They are rebels of society who actively resist discipline and refuse to conform to social demands. As a result, they are in constant conflict with their environment. Their disregard for conventions and rules of human decency is not due to an inability to distinguish right from wrong. Intellectually they know that their behavior deviates from the accepted normal pattern, but they have no desire to reform.
There is likewise a form of mental derangement in which the intellectual faculties appear to have sustained little or no injury, while the disordered is manifested, principally or alone, in the state of the feelings, temper or habits. In cases of this nature, the moral and active principles of the mind are strongly perverted or depraved; the power of self-government is lost or greatly impaired, and the individual is found to be incapable, not of talking or reasoning upon any subject proposed to him, for this he will often do with shrewdness and volubility, but of conducting himself with decency and propriety in the business of life.
Since almost everyone has at some time been tempted to commit a crime or to engage in antisocial behavior, it is fairly easy for a normal person to identify himself with and, superficially at least, to understand the motives and actions of the delinquent, criminal, and psychopath. This leads normal individuals to assume an unfavorable and somewhat vindictive attitude toward antisocial personalities. Rightly or wrongly, they believe that such individuals should be punished for willfully doing which normal individuals, though equally tempted, refrain from doing because such actions are detrimental to society and involve risk of detection and punishment. They feel that to permit a criminal to go free or receive a light sentence would not only be miscarriage of justice, but also that such a policy would weaken the defenses of normal individuals against the criminal, normal individuals are encouraged to hold in check their own antisocial inclinations and are indirectly rewarded for their good and moral conduct.
METHODS OF STUDYING BEHAVIOUR
So far, some basic information about behavior and its types has been covered. Now let’s turn to what are the methods of studying behavior, how do psychologists perform the task of adding to our knowledge about behavior and cognitive processes? How do they move beyond common sense in seeking answers to puzzling questions about human behavior? Psychologists use several different techniques for conducting systematic search. This section will describe the most important of these procedures.
NATURALISTIC OBSERVATION: Scientists as Explorers
Almost everyone finds the giant panda of China fascinating. Here, at least in outward appearance, is a teddy bear come to life. For many years zoos throughout the world sought eagerly to add these appealing animals to their collections. Unfortunately, these efforts usually produced disappointing results. The pandas, when finally obtained, seemed to pine away in captivity. Even worse, they adamantly refused to mate, despite the best efforts of the experts and the pandas’ keepers. Given the small and declining numbers of pandas present in the wild, these events seemed to spell disaster for a species most people would very much like to preserve. Could anything be done to change this situation? There seemed only one way to find out: observe the pandas in their natural habitat to learn more about their behavior and about what could be done to save them from extinction. In short, this seemed to be a situation in which research should be conducted through naturalistic observation?systematic study of behavior in natural settings. Fortunately efforts along these lines have added greatly to the knowledge of giant pandas. Scientists have spent months observing the pandas in the mountainous regions of western China where they live and, from such study have extracted information about pandas’ diet and mating habits that may prove useful in assuring their survival. As this example suggests, naturalistic observation is often used to study animal behavior. However, it is sometimes applied to human beings as well. For example, an ingenious study of this type conducted by Murdoch (1988) observed the behavior of male patrons drinking beer or liquor in randomly selected bars. Information on the number of drinks the men consumed, and resulting changes in their reactions to others offered intriguing insight into the effects of alcohol on behavior in at least one setting.
It should be added that sometimes the data obtained through naturalistic observation are relatively informal, and this can reduce their scientific value. However the fact that subjects are studied in natural settings and so are likely to act in the ways they normally do is an important advantage that makes this method useful in some contexts.
CASE STUDIES: Generalizing from the Unique
As we already know from your own experience, human beings are unique. Each possesses a distinctive combination of traits, abilities, and characteristics. Given this fact, is it possible to learn anything about human behavior generally from detailed study of one or perhaps a few persons? Several famous figures in the history of psychology have contended that it is. Thus they have adopted the case method, in which detailed information is gathered on specific individuals. For example, Freud based his entire theory on personality on the case method. Is the case method really useful? In the hands of talented researchers such as Freud, it does seem capable of yielding valuable insights about behavior. Moreover, when the behavior involves very unusual, the case method can be quite revealing. These cases involve individuals who have experienced specific kinds of damage to the brain and, as a result, show certain kinds of memory deficits. By studying the pattern of such memory losses, psychologists have been able to piece together a more complete picture of how memories are stored in the brain . So, despite its obvious drawbacks?for example, the possibilities that researchers’ emotional attachments to people with whom they work closely for months or even years can reduce their objectivity?the case method does appear to have its uses. When used with considerable caution, it can prove helpful in the investigation of at least some aspects of behavior.
SURVEYS: The Science of Self?Report
This involves asking large numbers, individuals to complete questionnaires designed to yield information on specific aspects of their behavior or attitudes. Such surveys (or polls) are often conducted to measure a wide range of attitudes and behaviors. Examples include surveys on health care reform in the United States or economic reform in Russia, voting preferences prior to elections, consumer reactions to various products, health practices, and public compliance with safety regulations.
Surveys are often repeated over long periods of time in order to track shifts in public opinions or actual behavior. For example, some surveys of job satisfaction?individuals’ attitudes toward their jobs?have continued for several decades. And changing patterns of sexual behavior and sexual attitudes have been tracked by the Kinsey Institute since the 1940s.
The survey method offers some very real advantages. Large amounts of information can be gathered with relative ease, and shifts over time can be readily noted. And when conducted carefully, surveys can provide highly accurate predictions with respect to the outcome of elections and other events. However, the disadvantages are also quite apparent. People may fail to respond accurately or truthfully, providing answers that place them in a favorable light rather than ones that reflect their true views. In addition, the results of surveys are useful only if the persons questioned are truly representative of larger groups to whom the findings are to be generalized. For example, imagine that a survey conducted with 10,000 young men and women indicates that more than 90 percent are strongly in favor of allowing total nudity on public beaches. Should you throw away your bathing suit as an unnecessary nuisance? Perhaps But not if you learn that all 10,000 persons who responded to the survey are subscribers to a magazine devoted solely to the joys of nudism. The fact that they are raises serious questions about the extent to which they are representative of the larger group to whom we wish to generalize? the entire adult population. The moral is clear: Unless the people who respond to a survey are similar to a larger group to whom we wish to extend the results, such generalizations are on very shaky grounds.
THE CORRELATIONAL METHOD: Knowledge through Systematic Observation
Prediction?the ability to forecast future events from present ones?is an important goal of science; psychologists, too, often seek to make predictions. Consider, for instance, how useful it would be if we could predict from current information such future outcomes as a person’s success in school or various occupations, effectiveness as a parent, length of life span, or likelihood of developing a serious mental disorder. How can this goal be attained? One Answer involves efforts to determine whether various aspects of the world, (termed variables because they can take different values) are related to one another. That is, we try to determine whether changes in one variable are associated with changes in another so that, for example, as one rises, the other does too. The stronger such relationships (correlations), the more successfully one variable can be predicted from the other.
We already know from our own experience, some events are indeed related to others and so can be used as effective predictors of them. For example Meteorologists (scientists who specialize in the prediction of weather) found that the greater the number of disturbances on the face of the sun ?the more unsettled the world’s weather will be in the coming in several different respects. Similarly, it has been observed that the greater the number of hours of violent television shows watched by children, the greater their likelihood of behaving aggressively as teenagers.
While these examples involve only two variables, you should bear in mind that in many cases accurate predictions can be obtained only when several factors (and the correlations between them) are taken into account at once. For example, if you wish to predict the likelihood that specific persons will rise to positions of leadership in their careers, you will probably have to consider many different factors, including personal characteristics (for example, their interest in serving as leaders as well as their intelligence, flexibility, and persuasiveness); opportunities for leadership in the fields they plan to enter; the needs of the groups they may possibly lead; the potential leaders’ gender; and many other variables as well. How do psychologists use correlations in their research? How do they search for relationships between variables in order to be able to make accurate predictions about important aspects of behavior? Perhaps the best way of illustrating the nature and value of this research approach?known as the Correlational method is through an actual example.
THE CORRELATIONAL METHOD: An example:
Imagine that a researcher wished to test the following hypothesis, an as yet untested prediction about some aspect of behavior: The faster people speak, the more successful they are at persuasion. In other words, the researcher wants to find out whether fast talkers are indeed more convincing than slow ones. How could the investigator proceed? While there are many possibilities, one would be to measure the speed of speech of many would?be persuaders in a wide range of contexts?politics, sales, and so on. In each of these contexts, the researcher would also obtain some measure of the success of the persuaders?for example, the percentage of votes each candidate receives, the amount of merchandise each salesperson sells, and so on. If fast talkers are really more persuasive than slower ones, results might indicate that these two variables (speed and persuasiveness) are positively correlated: The faster candidate, and salespersons speak, the greater their success. The researcher would probably also use appropriate statistical procedures to measure the strength of this relationship; the stronger the correlation, the more accurately each variable can be predicted from the other. Correlations can range from ?1.00 to +1.00; the greater their departure from 0.00, the stronger the relationship between the variables being considered. So if the researcher found a correlation of +0.80 between speed of speech and success in influencing others, this would indicate a stronger link between these two variables than a correlation of +0.30. Similarly, a correlation of ?0.60 would indicate a stronger relationship than one of ?0.20. In this case, however, the negative sign indicates that the faster people speak, the less successful they are at persuasion.
The Correlational method offers several major advantages. For one thing, it can be used to study behavior in many real?life settings. For another, it is often highly efficient and can yield a large amount of interesting data in a short time. Moreover, it can be extended to include many different variables at once. Thus, in the simple study described above, information on the physical attractiveness, age, height, and gender of the political candidates, salespersons, and so on might also be obtained. Then these variables could also be related to success in persuasion, to determine if they too influence this outcome.
Unfortunately, however, the Correlational method suffers from one major drawback that lessens its appeal, at least to a degree: the findings it yields are not conclusive with respect to cause?and?effect relationships. That is, the fact that two variables are correlated (even highly correlated) does not guarantee that there is a causal link between them?that changes in the first cause changes in the second. Rather, in many cases, the fact that two variables tend to rise or fall together simply reflects the fact that both are caused by a third variable.
For example, suppose that a researcher finds a positive correlation between the speed at which politicians talk and the percentage of votes they receive in elections. Does this mean that speed of speech causes voters to prefer certain candidates? Perhaps. But it may also be that fast?talking candidates know more about the issues than slower?talking ones. If this is the case, then the relationship between speed of speech and the outcome of elections is somewhat misleading. Both speed of speech and success at winning elections are actually related to a third factor?knowledge of the issues. Perhaps this key point can be clarified by a few additional examples of correlations that do not indicate causation.
Experimentation: Knowledge through Systematic Intervention
While psychologists use all of the research methods described so far, they often prefer the approach we will now consider: experimentation (or the experimental method). This involves efforts to determine if variables are related to one another by systematically changing one (or more) and observing the effects of such variations on the other (or others). There are several reasons why psychologists prefer this basic approach, but perhaps the most important is this: In contrast to the other methods, experimentation yields relatively clear?cut evidence on causality. If systematic variations in one factor produce changes in another (and if additional conditions we’ll soon consider are also met), we can conclude with reasonable certainty that there is a causal link between the factors: that changes in one caused changes in the other. Establishing such causality is extremely valuable from the perspective of one major goal of science: explanation. Briefly, scientists do not wish merely to describe the world around them and the relationships between different variables or factors in it. They also wish to be able to explain why such relationships exist?why, for example, people find some gauges easier to read than others, why individuals with certain personality traits are more likely than others to suffer heart attacks, why some people gain weight readily while others do not. Experimentation, because it often yields information useful in answering such questions, is frequently the method of choice in psychology. But bear in mind that there are no hard and fast rules in this regard. Rather, most psychologists select the research technique that seems most suited to the topic they wish to study and the resources available to them.
EXPERIMENTATION: Its Basic Nature:
Reduced to its bare essentials, the experimental method involves two basic steps: (1) The presence or strength of some variable believed to affect behavior is systematically altered, and (2) The effects of such alterations (if any) are measured. The logic behind these steps is as follows: If the factor varied does indeed influence behavior or cognitive processes, then individuals exposed to different levels or amounts of that factor should differ in terms of their behavior. Thus, exposure to a small amount of the variable should result in one level of behavior; exposure to a larger amount should result in a different level, and so on.
The factor systematically varied by the researcher is termed the independent variable, while the aspect of behavior or cognitive processes studied is termed the dependent variable. In a simple experiment, then, different groups of participants are exposed to contrasting levels of the independent variable (such as low, moderate, and high). The participants’ behavior is then carefully measured to determine whether it does in fact vary with different levels or amount of the independent variable. If it does?and if two other conditions described below are met?the researcher can tentatively conclude that the independent variable does indeed cause changes in the aspect of behavior being studied.
To illustrate the basic nature of experimentation in psychological research, let’s return to the question we considered earlier: Are fast talkers more persuasive than slow ones? A researcher who decides to employ the experimental method to study, this topic might begin with the hypothesis that the faster people talk (at least up to a point), the more persuasive they are. In such research, the independent variable would be the speed at which would?be persuaders speak, and the dependent variable would be some measure of the persuaders’ success in influencing their audiences.
There are many different ways of testing the hypothesis that fast talkers are more persuasive than slow ones, but for the sake of illustration, let’s assume that the researcher arranges to have an assistant deliver a speech designed to alter listeners’ views on a specific issue: legislation to limit the use of chemicals known to deplete the earth’s ozone layer. The speaker would then present this speech to different groups of participants at contrasting speeds. For example, for participants in one group, the speech would be presented at a slow pace (150 words per minute); for those in another group, it would be presented at a moderate pace (170 words per minute); and for those in the third group, it would be presented at a fast pace (190 words per minute). Then some measure of audience members’ attitudes toward the legislation would be obtained. We must assume by the way, that individuals in the three groups start out with similar attitudes toward the legislation; if they do not, serious complications can arise in the interpretation of any results that are obtained.
EXPERIMENTATION: Two Requirements for Its Success.
Earlier, we saw that before we can conclude that an independent variable has caused some change in behavior, two conditions must be met. The first condition involves random assignment of participants to experimental conditions. According to the principle of random assignment, all participants in an experiment must have an equal chance of being exposed to each level of the independent variable. The reason for this rule is simple: If participants are not randomly assigned to each group, it may later be impossible to determine whether differences in their behavior stem from differences they brought with them, from the impact of the independent variable, or from both. For instance, continuing with our study of speed of speech and persuasion, imagine that participants in the study are drawn from two different groups: first?year law students and a troupe of high school dropouts enrolled in a special course designed to provide them with basic vocational skills. Now imagine that because of differences in the two groups’ schedules, most of the participants exposed to the slow talker are law students, while most of the people exposed to the fast talker are high school dropouts. Suppose that results indicate that participants exposed to the fast talker show much more agreement with the views expressed than participants exposed to the slow talker. What can we conclude? Not much, because it is entirely possible that the difference stems from the different mixes of participants in the two experimental conditions. In the slow?talker condition, 85 percent of the participants are law students and 15 percent are high school dropouts, while in the fast?talker condition, the opposite is true. Since law students may be somewhat harder to persuade than high school dropouts, we can’t really tell why these results occurred. Did they derive from differences in the persuaders’ rate of speech? From different proportions of the two groups of participants in each condition? Both factors? It’s impossible to tell. To avoid such problems, it is crucial that all subjects have an equal chance of being assigned to each experimental group.
The second condition referred to above may be stated as follows: Insofar as possible, all other factors that might also affect participants’ behavior, aside from the independent variable, must be held constant. To see why this is so, consider what will happen if, in the study on speed of speech and persuasion, different speakers are used in the two conditions. Further, imagine that one of these speakers?the fast talker?has a pleasant cultivated voice, while the slow talker has an irritating voice and a thick, unpleasant accent. Now assume that participants express greater agreement with the fast talker, than the slow one. What is the cause of this result? The difference in the speakers’ speed of speech? Differences in the pleasantness of their voices? Both factors? Obviously, it’s impossible to tell. In this situation the independent variable of interest (speed of speech) is confounded with another variable?pleasantness of the speaker’s voice?that is not a planned part of the research. That is, another factor changes as the independent variable changes, so we can’t tell whether any effects observed are produced by the independent variable or this other factor. When such confounding occurs, the findings of an experiment are largely un-interpretable.
EXPERIMENTER EFFECTS AND DEMAND CHARACTERISTICS:
How to Get Positive Results Without Really Trying. Before concluding this discussion of experimentation, we should consider two additional pitfalls lying in wait for careless researchers. The first of these is the risk of experimenter effects?the fact that sometimes researchers can influence the behavior of subjects without intending to do so. Such effects can occur in several ways. First, researchers usually have expectations about how participants in their studies may behave. These expectations, in turn, may influence their behavior toward participants and so alter the results obtained. For example, if experimenter expects fast talkers to be more persuasive than slow ones, she may nod or smile at participants when they express agreement with the speaker’s views in the fast?speech condition, but show disapproval when they express agreement in the slow?speech condition. The result: Her reactions may influence participants and lead to greater agreement with the speaker’s views in the fast?speech than in the slow?speech condition.
A second potential snare for unwary researchers is the fact that experimenters can sometimes communicate the hypothesis behind the study to participants. Once they do, participants’ behavior may be affected even if the researchers have no direct contact with them and cannot provide subtle cues of approval and disapproval such as those described above. The effects of such communication are known as demand characteristics, since they place subtle demands on participants to “help” the researcher by confirming his/ her hypothesis. Research findings indicate that experimenter effects and demand characteristics can both exert powerful effects on participant’s behavior. Thus, it is crucial that these effects be minimized in all psychological research. One way of doing this is through a double?blind procedure. Here, persons who have contact with participants (often research assistants) are unfamiliar with the hypothesis under investigation and don’t know the condition to which participants have been assigned. Under these conditions, they can’t readily communicate the hypothesis being studied and can’t have clear expectations about how a given participant “should” behave. Another technique for avoiding experimenter effects and demand characteristics is to minimize direct contact between participants and the researcher. This can be accomplished through the use of computerized procedures, in which subjects receive instructions and perform experimental tasks by means of a computer terminal. Through these and related procedures, the impact of several potential pitfalls can be reduced and the validity or accuracy of experimental findings enhanced.