Conformity-deviation, norms, and group relations
Muzafer Sherif
University of Oklahoma
There are certain questions that should be raised at the outset in dealing with the problem of conformity and deviation. I shall start by raising them. Then some laboratory studies of the normative process which underlies conforming or deviating behavior will be summarized, and limitations of the confines of the traditional laboratory in handling basic problems of conformity and independence will he discussed. This will lead us to the necessity of research on normative process within the framework of the properties of groups. It will be concluded that research and experimentation thus formulated will yield a basis for realistic evaluation of the norm system of a group from the point of view of social stability and social change.
Problem of Conformity and Deviation
An item of behavior, taken in and by itself, cannot he labeled either conformity or deviation. There is no such thing as conforming or deviating behavior in the abstract. The terms "conformity" and "deviation" make sense when at least the following questions are raised:
1. Conformity to what? Deviation or departure from what? Always, conformity is conformity to something. Deviation is
( 160) departure from something, whether the referent of that "something" is made explicit or not.
What is that "something"? The referents may be the prevailing, the usual, or expected ways of doing things in the individual's surroundings. This is the normative basis of the problem. The referents may be the individual's particular place or position in the scheme of interpersonal or group relationships. This is the organizational basis of the problem. Does the individual accept and behave in terms of the place and position expected of him and his kind in the scheme of things?
2. What is the relative importance of the behavior area in which conformity or deviation occurs? For example, is it a matter of whether a father takes care of his family as he should, or is it a question of whether he keeps up with the baseball scores in the World Series as his friends do?
3. Is the normative basis of the behavior in question shared and upheld by other groups to which the individual is related in some capacity? Or do his multiple groups put contradictory or even conflicting demands and expectations on the individual for his behavior in given dimensions? This of course relates to the problem of integration or conflict of social values in the psychological world of the individual.
4. The fourth question concerns whether conformity and deviation occur primarily through coercion or threat of subsequent coercion and force, or whether the behavior in question is prompted by the individual's inner convictions and personally cherished values.
5. What are the alternatives available for the individual in the stimulus situation with respect to the area of behavior in question? Are there many, few, or none? Are they clearly defined or difficult to distinguish? In other words, what situational factors enter into the picture, both as to the physical setting and the other people involved?
When studied in the context of these questions, conform-
( 161) -ing or nonconforming behavior can be taken as an index of the degree of stability or the extent of change in the human relationships of a given setting and specifies whether stability or change occurs primarily as a consequence of coercion or primarily through the voluntary interaction of individuals. Thus viewed, conforming behavior and nonconforming behavior can serve as a basis of evaluating the trends in human relationships: how a group is doing and in what directions it is headed.
These are among the basic problems for any human group. They are all the more vital in this modern shrinking world. Whether we like it or not, peoples and groupings are being brought into closer functional relationships. Scarcely a group is left which is contained within itself as a closed system. What a particular group is doing and where it is headed have wide impacts on other peoples. The implication of this enlarging interdependence of peoples is rather obvious, namely a normative system which transcends restrictive, monopolistic loyalties and conformities still surviving from relatively more closed group patterns of previous periods.
RESULTS OF NEGLECT OF ISSUES. Apart from the questions just raised, that is, questions of the referents of conforming behavior, of the relative importance of the area of behavior, of the integrated or conflicting character of the normative bases of the behavior in question, of how conformity is brought about, conforming and nonconforming behavior cannot be studied as a scientific problem. Neglect of one or the other issues raised here has resulted in a spate of literature in recent years by social psychologists, social scientists, and essayists, which at its best is healthy social critique and at its worst boils down to a romantic protest and a cry for heroes.
We even read discussions of whether man is by nature a conformist, a submissive prey of social winds and tides, or whether by nature lie is a seeker of truth, hence required by
( 162) -his own nature to be independent. Such formulations are reminiscent of the old controversies over human nature by instinct theorists of whether man is altruistic or selfish, cooperative or competitive, acquisitive or sharing. Now the argument seems to be transferred to the cognitive sphere.
By implication, those who would define "human nature" as basically conformist or as basically independent praise one kind of behavior and damn the other. Yet by formulating the problem in this dichotomous form in the abstract, conforming behavior or deviating behavior cannot be evaluated in a consistent fashion.
Taking a stand as an apologist of conformity can amount to the praise of blind subservience. On the other hand, singing the praise of nonconformity apart from evaluation of the norm or value to which it is related may lead to an absurd dilemma. Let's just cite a few cases of nonconforming behavior, such as driving down the middle of the road, monopolizing a conversation, deliberate plagiarism, or stealing. Of course, those who see virtue in nonconformity in its own right would protest these crass examples. For praise of nonconformity is made of righteous nonconformity. This is exactly the point. Nonconformity or conformity cannot be evaluated in its own right apart from its referent, namely the normative basis of the behavior in question.
Normative Process in the Laboratory
I shall first consider the formation of the normative process as studied in the psychological laboratory, with special reference to variations in results owing to the kind of controlled stimulus setup presented to the subjects.
The study of norm formation in the laboratory has been undertaken through producing a characteristic mode of behavior relative to the aspect of the stimulus situation experimentally introduced. This production embodies the bare essential of norm-regulated behavior.
( 163)
MODES OF BEHAVIOR AND STIMULUS PROPERTIES. If the stimulus arrangement provided by the experimenter is objectively well graded or has compelling anchorages, the ensuing uniformity of behavior relative to it is determined by these salient features of the stimulus conditions. As a result of repeated encounters with them, characteristic modes of behavior come into close fit with the stimulus properties.
Tresselt and Volkmann (1942) clearly demonstrated the principle in their study dealing with the production of uniform opinion by nonsocial stimulation. These authors formulated the principle in question as follows: "Each person in a group says what he does not only because he has been persuaded by argument, induced by reward, compelled by pressure, guided by past experience, or influenced by the voiced opinions of other people; he says it also because he faces a restricted range of social or nonsocial stimulation, and this range has determined his scale of judgment" (p. 243).
When stimuli are presented serially over a period of tune, the range representing categories of behavioral uniformity to particular items is appropriate evidence for inferring the formation of a psychophysical scale.
The study of stimulus relationships affecting the formation and functioning of psychophysical scales cannot occupy our attention further at this time. It has been carried to a rather sophisticated level in recent years through the efforts of such contributors as Graham (1952), Helson (1959), Johnson (1955), Stevens (1957), Volkmann (1951), and others.
For the present discussion it is sufficient to emphasize that the normative process can be determined primarily by the range and salient anchorages of the stimuli to which individuals are exposed repeatedly. I strongly suspect that the principles to which we have alluded underlie the similarity in characteristic modes of outlook in technological matters by various human groupings scattered over wide geographic regions, but exposed to similar technology (Sherif, 1948;
( 164) Sherif & Sherif, 1956). If so, the rapid shift in psychophysical scales following the introduction of new stimulus values may be pertinent to understanding the relatively faster assimilation of new technological items as compared with new social concepts. This refers, of course, to the well-known empirical facts formulated by sociologists and anthropologists as the "cultural lag."
 |
| Figure 5.1. Mean ratings of moderately undesirable behavior items. Drawing based on data from Cohen [1957].) |
Of course, weights, lines, and sounds do not have an exclusive copyright on compelling stimulus characteristics. Stimuli with social relevance, including verbal statements, may have such compelling properties as well. In his study on stimulus conditions as factors in social change, Cohen (1957) at the University of Oklahoma studied changes in norms as a function of alteration in the range of social stimulation faced by the subjects. By scaling statements of undesirable behavior of the kind used earlier by McGarvey (1943), Cohen selected moderately undesirable acts, such as "fishing without a license." Subject pairs were to come to an agreement in rating how desirable or undesirable these acts were. Exposed to the restricted range of items, control and experimental pairs rated them similarly (see Figure 5.1). In the second experimental presentation, very undesirable behaviors, such as "kidnapping a baby for ransom," were included. As Figure 5.1 shows, the somewhat undesirable behaviors were agreed
( 165) to be more desirable, thus showing the well-known contrast effect owing to the dastardly acts introduced in the second presentation. The differences between the experimental and control subjects were maintained in the third session when individuals rated the behaviors alone, indicating a normative process primarily determined by the exposure of interacting individuals to specified ranges of stimulus items, It is in this sense that conformity is related to the formation of psychophysical scales.
Laboratory Studies
Now let's turn to laboratory studies on the formation of psychosocial scales: scales of characteristic modes of behavior whose formation may he influenced by the relationships among interacting individuals. Features of man's relationships with man become most salient as determinants of his conformities precisely when the stimulus situation they face together is highly fluid and provides various alternatives.
This observation has been made time and again by social scientists. I was first impressed with it upon reading Durkheim's accounts of the formation of representations collectives in out-of-ordinary interaction situations and the accounts of Chicago sociologists-notably Clifford Shaw, Thrasher, and Zorbaugh---of small group functioning in interstitial areas of large cities. Even in the midst of "social disorganization," in the sociological sense of that term, an orderliness prevails in the social life of these small groups. How does the normative process take shape under such conditions? How do new standards of conduct arise in out-of-ordinary situations, times of crisis and the breakdown of established conformities?
RESEARCH ON NORM FORMATION. These were among the questions that led to the laboratory norm-formation experiments which utilize the autokinetic setup as a fluid stimulus in the dimension in question amenable to various alternative
(166) modes of behavior (Sherif, 1935; 1936). The problem was conceived in terms of stabilization of behavior over a period of time, and not as a question of whether an individual is susceptible to suggestion in this or that particular round of judgment. Contrary to Durkheim's view, even when the individual is alone and not interacting with others, his psychological functioning becomes organized and exhibits emergent properties. This was the finding of the individual sessions. As some of you know, after the individual faces such a situation repeatedly, his behavior stabilizes within a characteristic range and around a modal point with reduced variability.
When individuals face the autokinetic setup together, over a period of time a convergence of the individual behaviors occurs, resulting in similarities not initially present. But the norm that emerges during interaction is not an average of the individual norms. Nor is it necessarily identical with the initial behavior of one or the other individual, although a large prestige or status differential may almost produce this result. To specify the normative outcome further, the relationships of individuals in a particular interaction process have to be delineated. However, once formed, the convergent behavior is not dependent upon the immediate Presence of other individuals. In subsequent sessions when the individual was alone, behavior was still regulated by the normative process.
It was suggested that when the individual changes his verbal reports in the presence of another person making somewhat different judgments, he is not really seeing the stimulus any differently, hut is simply changing his behavior in order to avoid disapproval and appear agreeable. Certainly this does happen in some situations. A recent experiment by W. R. Hood and the writer (1957) was designed to investigate whether or not behavior in the highly unstructured autokinetic situation represented such public compliance. Procedures were planned to eliminate suspicion that the
( 167) experiment had anything to do with social influence and to remove the immediate presence of another person or the sound of his voice at the time judgment was rendered. The subject simply overheard another person making 20 judgments while waiting his turn to make estimates by himself.
 |
| Figure 5.2. Proportions of judgments 5" or less and greater than 5" for two conditions (From Hood & Sherif [1957].) |
One sample overheard judgments ranging from 1 to 5 inches and another sample overheard reports ranging from 6 to 10 inches. Figure 5.2 summarizes judgments made when the subjects were alone, by showing the proportions of judgments 5 inches or less and 6 inches or more under these two conditions. Later, when asked what extent of movement they had usually seen, the subjects' estimates did not differ sig-
( 167) -nificantly from their own median judgments in the situation. We may conclude, I believe, that in this situation, individuals "call them as they see them" and they see them as influenced by judgments previously overheard. There is no evidence of a discrepancy between judgment and verbal report.
JUDGMENTS AND LIMITS. Can we say, then, that in highly fluid situations, the relationships among individuals determine behavior altogether, and that the "sky is the limit" as far as the extent of social influence that can be achieved? We cannot. For even in this highly unstructured situation, the size of the room, as the individual can determine it from finding his way about, from echo and the like (Sherif & .Harvey, 1952)-the subject-to-light distance, exposure time, intensity of the light, and other stimulus arrangements set rather definite limits upon the extent of movement perceived. If another person's judgments exceed these limits too far, they are unlikely to exert any determining role at all.
Whittaker (1958) showed this at the University of Oklahoma by having planted subjects make judgments which exceeded the individual's largest estimate made previously when he was alone. For different samples, the "plant" made judgments ranging upward from magnitudes I inch larger, twice, eight, or twelve times larger than the maximum estimate the individual had given when alone. Figure 5.3 summarizes the results in terms of the average difference between means of the individual and together sessions. It may be seen that the partner's judgments had a significant effect when they exceeded the previous maximum estimate by I inch and upward, that the effect decreased when his judgments were twice as great, and that no significant effect was found when the partner's judgments were eight or twelve times as large. These latter changes do not differ significantly from those for the control group, which simply judged alone a second time. Analysis in terms of frequency of judgments exceeding
( 169) the individual's initial range when alone leads to the same conclusion. We conclude "the sky is not the limit" in the effectiveness of attempted influence even in a highly unstructured situation.
Furthermore, once a psychosocial scale has been established, the experimenter cannot play around indefinitely,
 |
| Figure 5.3. Mean difference scores for judgments alone and with "plant." ( From Whittaker [1958].) |
introducing new social anchorages in order to change it. This is in some contrast to laboratory findings on psychophysical scales, which reveal strikingly quick adaptations to changes in the stimulus values presented. The successive introduction of conflicting social anchorages produces breakdowns of a psychosocial scale, as Norman Walter's experiment at the University of Oklahoma (1955) shows. Walter was interested in what would happen to an established psychosocial scale
( 170) for judging autokinetic movement when conflicting reports from equally good sources were presented successively. A few days after the individual established a characteristic range and mode of behavior alone, he was informed casually of the performance of students at a university with high prestige in his eyes. The figure given was actually either the 90th or l0th percentile of his own previous estimates. The result was a significant shift toward the figure introduced, with greatly reduced variability, even a greater reduction than for control subjects whose variability decreased regularly, during four sessions on different days without procedural changes.
At a third session 3 or 4 days later, a second "data report' opposite in direction to the first was given experimental subjects, this one from another institution of about equal prestige in his eyes. The typical result was a shift in behavior back to the region of the initial norm established when alone, but with increased variability. Finally at a fourth session, the experimenter discredited both reports by saying the results were suspected of error. The result was further marked increase in variability such that a breakdown in the normative process ensued.
ANCHORAGE POINTS AND BEHAVIOR. The effect of systematically reducing the number of anchorages available in the stimulus arrangement is to increase the probability that characteristic behaviors developing as individuals face the situation repeatedly will be significantly influenced be relationships among the individuals. This generalization is supported by the findings of various studies. and James Thrasher (1954) tested it in a single design. His stimulus setups represented three gradations of stimulus structure. Judgments were obtained from individuals alone, then in pairs composed either of good friends or initially neutral persons. In the situation with the most stable anchorages available in the
( 171) stimulus setup, neither friends nor neutral individuals affected cacti other's behavior significantly. But in the least structured situation, the relationship among the individuals as friends or as initially neutral decidedly affected the normative process. The intermediate situation produced results intermediate between these two.
Limitations of Laboratory Confines for Valid Research Strategy
The general conclusion to be drawn is that the choice of the stimulus setup in the laboratory determines to a great extent how effective social factors introduced by the experimenter will be. If the structural properties of the stimulus arrangements allow no alternatives, social factors introduced will play relatively little part in patterning characteristic behavior. If the stimulus conditions provide alternatives in his siring up of them, the likelihood of compliance to social influences introduced by the experimenter will increase. There are a good many combinations and gradations in between the stimulus situations representing these two extremes. If space permitted, we could draw upon experiments by Asch (1952), Coffin (1940, Luchins (1944), Mausner (1953; 1954), and others to accentuate this conclusion.
Frequently, the manipulator in public relations, the demagogue in public affairs enjoys a cynical glee at the compliance and malleability of people owing to the muddied atmosphere which the manipulator and demagogue themselves have helped manufacture. It is a laudable antidote against such cynical views of human affairs to set up experiments in the laboratory with clear-cut stimulus arrangements and to present results showing that people are independent in their appraisals of the situation, that they do not blindly succumb to false social influences. However, the demagogue himself seldom works with a situation as clear-cut as, say,
( 172) matching a pencil to a yardstick, when he attempts manipulations. He fishes in muddied waters, unless he feels that brute force is on his side.
COUNTERACTING MANIPULATION. Perhaps the more effective way of counteracting such situations may be a concentration on analysis of conditions which render people susceptible to manipulative influences. Here one problem is what is presented at all, in the first place. All the information, all the news fit to be presented in speech and mass communication media on vital matters which lie far beyond the individual's perceptual range come to him through the selectivity of people in control of the mass media. Especially in the contemporary scene, when man is faced with the problem of taking sides, making decisions, and expressing opinions on issues that relate to the use of atomic energy, foreign aid in distant lands, or the merits or demerits of a balanced budget, he gets most of his information through the mass media or from other persons who got it from the mass media. The problem of selectivity in what is presented is at least as important as that of the presence or absence of alternatives once a situation requiring a decision is encountered. Furthermore, we suspect that in problems such as those just mentioned, man's ethnic, religious, and political ties may be weighty factors in determining which side is taken and which decision is made. If so, it is a bit unrealistic to study conformity and noncomformity in an arbitrary situation within the confines of the traditional laboratory and to declare the generalizations reached on this basis as the verdict of psychological science.The study of the relative contribution of selected stimulus factors, in which almost infinitely fine gradations of structure are possible, and of selected social factors, which are also numerous, is an interesting psychological problem. A good
( 173) many psychologists could devote their whole lifetimes to it. But no amount of concentration on all the variations between a sharply defined stimulus setup at one extreme and a highly fluid setup at the other, and no amount of technical refinement within the traditional laboratory setting will settle for us the important question of what are typical conditions conducive to the production of various modes of compliance and independence to social influences.
It is not yielding or being independent to discrete and transitory social influence that brought about concern over problems of conformity and independence in human affairs on the contemporary scene. That is why my brief report of experiments dwelt mainly upon studies of the normative process producing a characteristic attitude and mode of behavior to its referents over a span of time. Studies of yielding and independence in relation to one-episode social influence are reminiscent of Allport's early studies of "social decrements," "social increments," and "social subvaluents" through what is referred to as the study of "pure social effect." Such studies have only limited bearing on problems of conformity and deviation in which the individual's selfesteem, status, and work in relation to other persons significant in his scheme of things are at stake.
INTERACTION CONDITIONS. The settings in which compliance and noncompliance become important problems cannot be determined within the confines of the traditional laboratory setting. Here we have to extend our research perspectives and put validity checks upon ourselves to catch the essential properties of actual situations conducive to various modes of compliance or independence. In order to achieve this perspective, intimate familiarity must be achieved first with interaction settings in which an integral aspect is that of compliance or independence in behavior. And the proper
(174) focus for developing valid research strategies is upon the formation and functioning of actual groups in both their ingroup and intergroup interactions.
Conformity and Nonconformity in Group Relations
This brings us to the central part of my presentation. The representative, the typical, problems of conformity and nonconformity can be more effectively singled out through due recognition of man's behavior relative to significant other persons. Significant other persons stand in specifiable relationships to the individual-friendly or unfriendly, pulling together or pulling apart. Conforming or noncomforming behavior makes very little sense when it is not analyzed within a framework of these relationships. An observation will illustrate the point. In 1958, a group of liberal students at a southwestern university were interested in persuading restaurants and soda fountains in the area to cater to Negro students. The representative response of the shopkeepers was that they were willing to do so but each individual was concerned about what the other shopkeepers in the area would do. Our image of ourselves, our appraisals of our own practices, are not self-generating. They are not independent of our relatedness to people significant in our eyes, whether these significant people are seen as friend or foe. This point will be specified further in connection with the properties of groups and ingroup and outgroup demarcations.
It may not be too far off the mark to maintain that man's relation to significant others is, on the whole, in terms of his membership in various groups, such as family, club, fraternity, occupational outfit, religious or political outfit of some sort. If the problem of conformity and independence is formulated within a framework of the individual's group setting, we are confronted head on with relationships in which the problem is an ever-present, integral aspect of interaction situations day in and day out, and not an incidental
( 175) side issue. If the problem is formulated within the concreteness of group relations, as these relations unfold in the actualities of social life, then conformity or noncomformity acquires a functional significance which is mutilated when either is considered apart from these relations, as by those who advocate a doctrine of an irreconcilable individual-group dichotomy. The social philosophy which puts issues in dichotomous either-or form, that is either for the individual or for the group, starts with the categorical assumption of individual and group as irreconcilable entities or antithetical polarities, as though demands and interests of one are necessarily in conflict with the interests of the other for all occasions. Within the framework of man's ties with other men in lasting relationships, the conflicting or harmonious character of interests is itself a problem of study. With the vantage point thus gained, the external stimulus, whether it be sharply defined or fluid and uncertain, can be studied as it becomes relevant to relationships among individuals facing definable problem situations.
Member's Experience of Conformity and Independence, and Properties of Groups
A rounded analysis of the important problem of the individual's experience when he complies and when he is independent in specific instances of his group relations should start with specification of the essential properties of the group itself and the individual member's psychological relationship to these properties.
The concept "group" means all things to all people. Various concepts are offered in the academic market place today in the name of operational definitions. Not infrequently, the model and technique are derived from more established sciences without due concern for their appropriateness as tools for valid study of human group problems. Unless the appropriateness of the proposed techniques and models is
( 176) examined relative to the essential properties of actual groups, they are doomed to inefficacy in yielding valid results which can be generalized to handle the individual's behavior in his actual group setting.
EARLY STUDIES. Prompted by this serious methodological concern with formulation of valid problems, we undertook an extensive survey of sociological field studies dealing with properties of small groups. We turned to this literature for the simple reason that sociologists have priority in their concern. with the properties of actual small groups and have collected considerable empirical data on the topic (Faris, 1953). These surveys were a first step in our ongoing research program on formation and functioning of groups in both their ingroup and intergroup relations, in which the problem of conformity and deviation is an integral part. They arc presented in various publications (Sherif, 1948, Chs. 5 and 6; Sherif & Cantril, 1947, Ch. 10; Sherif & Sherif, 1953, Ch. 8; Sherif & Sherif, 1956, Chs. 5, 6, 7).
These early surveys deliberately centered on informally or spontaneously formed small groups in order to start with group formations which are the creation of voluntary and free interaction among the individual members, and not the product of an organizational blueprint with rules and bylaws handed down by a governing body with outside authority. From this survey, severa1 generalizations about the properties of small groups were extracted. Here we touch only on the minimum essentials.
It is extracted that any small group functions as a delineated social unit. The members have a rather clear notion who is in, who is out, and also the marginal ones who did not quite make it at the time.
It is extracted that the individuals who achieve accepted membership in groups can be ordered at a given time along a status hierarchy from the leader down to the position at
( 177) the bottom. This property of status differentiation need not be brought about through a formal vote or through formal codification on paper. The relative status position that a given individual member occupies is operationally inferred from the relative frequency of effective initiative that he achieves in starting and carrying out activities and projects in which the membership as a whole participates. This result is confirmed by sociometric choices of the group members along dimensions of effective initiative as well as popularity. The status differentiation of the members constitutes the organization of the group and embodies the power aspects of relations within the group.
The psychological counterpart of the emergence of group structure or organization is revealed through reciprocal performance expectations of members, not through dictates of an outside authority, but on the basis of their experiences of the relative contributions of each member in previous efforts towards solution of common problems.
PROPERTIES OF "GROUPNESS." Another essential property of groups, extracted from a survey of empirical field studies, is a set of values or norms shared by group members. It can be said that the "groupness" of the group as a more or less lasting social unit may be best defined in terms of these essential properties, viz., organization or structure and a set of values or norms (Homans, 1950). More transitory social situations which lack these properties may be referred to as mere aggregates or togetherness situations.The set of values or norms of a group (variously referred to as its code, standards, or rules) has probably a more direct bearing on the problem of conformity and deviation. There would be no persistent problem of conformity or deviation if there were no norms to conform to or deviate from: a question raised at the start of this discussion. As long as there are values or norms shared, upheld, and cherished
( 178) by group members, compliance to and deviation from- them are ever-present concerns.
In the literature, there is confusion in regard to the concept "group norm." As long as a norm is property of the group, upheld by members as "theirs," a norm is an expected and even ideal mode of behavior for them and does not refer necessarily to the statistical average of the behavior of members (Freeman et al., 1952).
The social attitude of the individual, determining characteristic and persistent modes of behavior to relevant stimuli, be they other persons, groups, activities, institutions, or symbols, is derived from those expected, or even ideal modes of behavior referred to as a group norm. A norm is a group property and, as such, is a sociological designation. The individual's social attitude is the consequence of internalization of the norm by the individual. Social attitudes define the individual's relatedness to stimuli in question-to other persons, groups, symbols, etc.---and his stance for or against them. As such, social attitudes may be referred to as his ego-attitudes or, if you like, self-attitudes. The individual's experience of self-identity, his feelings of stability (security), his strivings toward expected and ideal goals consist in large part of ego-attitudes derived from his membership character in given groups during his life history.
This being the case, norms are not rules or standards of behavior devoid of motivational and emotional warmth. Social attitudes formed relative to group norms define a substantial part of the individual's goal-directed behavior, which is the earmark of any motivational state.
The motivational-emotional character of norm-regulated behavior is not a mystery if we consider the rise and stabilization of norms in spontaneously formed groups. Small groups arising on an informal basis in actual life are outcomes of interaction among individuals with motives perceived as common, be they common deprivations and frustrations
( 179) such as those experienced by youngsters in slum areas-or desires for social distinction and exclusiveness with appropriate facilities and prestige symbols-such as those characteristic of clubs mushrooming in residential extensions of large cities like Los Angeles, Houston, and Chicago. As anyone familiar with the history of labor organizations in this country knows, it was the common urge for mutual protection and improvement of working and wage conditions which prompted the banding together of laborers in the latter half of the 19th century, at first secretly and then in public forms which foreshadowed the modern labor unions. The norms cherished as almost sacred and upheld most tenaciously in word and deed by labor organizations to this very day are those related to the motivational issues that brought the early workers together-collective bargaining, the right to strike, seniority rights, the closed shop, minimum wage, and so on. The motivational bases of such norms are readily seen when one of the members deviates from the hard-won standards. Not just a few administrators, but the rank and file have coined labels and developed corrective measures for deviations they consider as selling out their interests. A similar analysis of motivational bases in the rise and functioning of norms can be applied to management and business organizations.
In short, norms arise and are stabilized relative to motivationally important relationships and activities. Serious issues of conformity and noncomformity arise relative to norms pertaining to matters of consequence to the group, its existence, its perpetuation, its solidarity and its effective functioning toward central interests and goals. Therefore, it is somewhat unrealistic to dwell upon cases of conformity or nonconformity in matters considered peripheral to the scheme of things by the group in question, such as the hobbies engaged in by members privately. This question of the relative importance of the behavior area was one of those which opened
( 180) this presentation. In this connection, it is essential that the investigator recognize that the importance of a norm in the scheme of a particular group may or may not correspond to the importance and seriousness of the issue in question in determining the course of human relations in a larger sense. For example, in this country until recently, many organizations, including labor, considered political matters as the politician's realm.
Latitude of Acceptable Behavior Defined by Norm
Now we turn to discussion of a concept which will provide us with a baseline for classifying given behavior as conforming or deviating and for evaluating unique personal variations of individuals in this respect. Norm-regulated behavior cannot be represented as a single point. The expected or ideal behavior within the bounds of a given norm is represented by a range of behaviors, which we have referred to as a "range of tolerable behavior" and "latitude of acceptance," and shall here term "latitude of acceptable behavior" (Hovland, Harvey, & Sherif, 1957; Sherif & Sherif, 1953, pp. 198, 207f.; Sherif & Sherif, 1956, pp. 171, 533).
As long as behavior falls within bounds defined by this range, it will not call for correctives applied to cases of deviation (Jackson, 196o). Behavior outside the limits is viewed as objectionable by other members and will arouse spontaneous correctives from the membership, even without deliberate formal action. In our present ongoing research on naturally formed groups in settings differentiated as to social rank in several southwestern cities, we find that an important difference between groups distinguished as to their solidarity is the extent to which the membership actively participates in correctives for deviating behavior (Sherif, 1959; Sherif K - Sherif, 196o). Solidarity is measured by members' behaviors when the leader is present and absent, by their secrecy and exclusiveness relative to outsiders, by relative coordination
( 181) of role performance in the face of mildly threatening situations. The group with greater solidarity by these measures is also the group whose members react to a man when a member deviates from an important norm.
Of course, the usual routines of social life run within the bounds of acceptable behavior. Like the oxygen in the air which is noticed only when the concentration falls, the reality of norms and their limits are seen most strikingly when correctives and constraints are aroused by deviation, as Durkheim so aptly noted.
NORMS, LATITUDE, AND BEHAVIOR. The latitude of acceptable behavior varies in magnitude according to the importance of the norm for the group. The more consequential the issue at hand, the more constricted the latitude of acceptable behavior. Conversely, the more peripheral the issue, the greater the variability encompassed by the latitude of acceptable behavior. In matters bearing closely on the existence and perpetuation of the group, the latitude of acceptable behavior is constricted, unless the group is in a state of disorganization. And, at times when the well-being of the group becomes an acute problem, it is constricted still further. In our experiments on group formation and intergroup relations in 1949 and 1954, groups met in intense competition for a single goal over a period of several days. At the height of the group conflict growing out of this competition even a slightly kind word about the out-group was seen as almost treasonable; whereas, a little earlier, similar acts were acceptable in the spirit of good sportsmanship (Sherif & Sherif, 1953; Sherif et al., 1954). It is not hard to find parallels of this constriction of the latitude of acceptable behavior in the political scene of recent years.Along the lines of previous studies in relation to the closed-shop issue in 1948 by Hovland, Volkart, and Sherif and the prohibition issue in Oklahoma by Hovland, Harvey, and
( 182) Sherif (1957), our study during the 1956 election campaign (Sherif & Hovland, manuscript) verified the constricted latitude of acceptance in matters vital to the existence of a group. In this experiment, the task was to indicate acceptance and rejection of nine statements ranging from extremely pro-Republican (labeled A) to extremely pro-Democratic (labeled I) on the issue of the election of presidential and vice-presidential candidates of the two parties. The finding pertinent to our present discussion was that identified members who were actively campaigning at the time for their respective parties, viz., members of Young Republican and Young Democratic organizations in several universities in the Southwest, included fewer stands in their latitudes of acceptance and rejected an extended range of stands as compared with individuals not identified organizationally with either party,' who had more extended latitudes of acceptance around the stand they endorsed (see Figure 5.4).
BEHAVIORAL LATITUDE AND POSITION IN THE GROUP. The latitude of acceptable behavior on a given issue varies also with the position the individual occupies in the group. I strongly suspect that position in the group and importance of the norm interact in rather complex fashion to define the limits of the latitude of acceptable behavior for a particular individual relative to a particular norm. But it should be mentioned that the leader, as a member of a group, is not immune to correctives. If his nonconformity pertains to a matter of sufficient importance, the end result may be his decline in the power hierarchy.
The concept of latitude of acceptable behavior is explicit recognition of the fact that no two individuals in the same group uphold the norms to the same degree, nor are cases of nonconformity involving different individuals ever identical. There are individual variations in both conformity and nonconformity owing to unique personal characteristics
( 183) of the individuals in question. These individual variations are psychological facts which should be included in an adequate conceptualization. Perhaps some of the difficulty encountered in doing so stems from the tendency in psychology to categorize individual variations as though they referred to events in an altogether different universe of discourse than social behavior. If unique individual variations are ordered relative to a baseline defined by the latitude of acceptable behavior and the range of rejected behavior on the issue in question, we may achieve a reference scale more meaningful than hypothetically constructed scales dictated by worn-out typologies and computational convenience (Hood & Sherif, 1955).
Conformity-Deviation and Change in Intergroup Relations
As concluded in the last section, a norm emerges in relation to significant aspects of the existence and activities of the group. The prevailing pattern of relations of a given group with others is certainly a significant concern for its members. In the course of friendly or hostile, cooperative or competitive traffic between groups, norms emerge which regulate the intergroup attitudes and behavior of the members of the ingroup toward the outgroup and its members. If the character of this traffic is positive, if the goals of the respective groups do not conflict, if interests of groups do not clash, the norms regulating attitude and behavior are friendly and favorable. If the groups are striving towards goals whose attainment implies gain for one group while signifying loss and defeat for the other, the intergroup norms are unfavorable and hostile. In line with the specific character of the norms, qualities or stereotypes are attributed to the outgroup, being positive or negative depending on the character of the intergroup relations. This process underlies the close or remote "social distance" at which another group and its members are kept.
( 184)
 |
| Figure 5.4 Latitudes of acceptance and rejection for stands on a political issue by individuals upholding different positions: A (extreme pro-Republican) to I (extreme pro-Democrat). |
(185)
(186)
GROUPS AND NEGATIVE STEREOTYPES. In our 1949 and 1954 experiments on group relations mentioned earlier, groups in conflict developed unmistakable negative stereotypes and each group viewed the other through the widening gap of social distance which appeared irreconcilable to the members of both groups. In the 1954 experiment, these negative stereotypes were changed and the social distance between the groups reduced to the point of mutual acceptance. A brief background is necessary to explicate the conditions bringing about this positive change.
In these experiments, groups formed as social units when individuals who were initially unacquainted were brought together in activities with compelling goals whose attainment required joint and coordinated action on the part of all individuals. As a consequence of interaction in such conditions over a period of time, groups emerged with organizations and with norms, as is characteristic of groups in actual life. When two groups had thus developed independently as social units, they were brought together for the first time in the second stage of the experiment, in a series of reciprocally competitive activities, in which the victory of one meant the defeat of the other, with attendant reward and glory for the victor and disappointment and frustration for the vanquished group. In the course of these encounters, the initial sportsmanship toward the adversary went into a sharp decline and was supplanted by derogatory invectives and open expressions of hostility. In time, unfavorable adjectives, such as "stinkers," "cheats," "sneaks," became stabilized in word and deed as the picture of the adversary. As intergroup conflict increased, cooperation and solidarity within each group reigned supreme.
At this point, members of each group were disinterested in any information about their adversary which might have changed their views of them; they were adamant in their insistence that they did not even want to be in the same
( 187) situation as the other group. Therefore, the ground was not favorable for negotiations between the leaders. When brought into physical proximity in activities enjoyable to each group separately, the social interaction situations were utilized for furthering the mutual derogation.
None of the measures just mentioned could have been effective in changing the reciprocal hostility to intergroup cooperation or in changing the highly unfavorable stereotypes to a congenial picture of the outgroup. Of course, the threat of a common enemy is usually conducive to burying the hatchet for the time being, but this measure, which had been effective in 1949, was discarded because of its temporary nature and because it simply widened the sphere of group conflict.
THE OUTGROUP AND ATTITUDE CHANGE. The experimental procedure that proved to be effective in changing attitudes and behavior toward the outgroup was the same in conception as that which had resulted in the individuals banding together in group organizations in the first place: viz., the introduction of problem situations and goals which individuals could not ignore but whose solution and attainment required that all of them pull together. When the problem situation involves two or more groups, the goals conducive to opening interchange between their respective members and laying the ground for mutual efforts could only be those with urgency for members of all the groups involved, but which cannot be attained by the resources and efforts of the groups separately. The coordination of all efforts in the same direction must be required to reach the goal, which may be avoiding a common disaster or may be ends mutually beneficial and satisfying to all. Such goals are superordinate to each of the groups in question. Through the introduction of, not one, but a series of superordinate goals, social distance between the groups in the experiment
( 188) was reduced and stereotypes changed from largely unfavorable to generally favorable pictures of the outgroup. Figure 5.5 shows the combined ratings of the outgroup on several traits made by members of the respective groups first at the height of intergroup conflict, and later, following the series of superordinate goals.
 |
| Figure 5.5 Rating of the outgroup during intergroup conflict (left) and reduction of conflict (right). (from Sherif et al. [1954.] ) |
The process of change involved shifts in conforming behavior and changes in what constituted nonconformity. The friendly interchange and mutual cooperation between members of the different groups which characterized their relationships following the series of superordinate goals were viewed by members of each group as conforming to the developing group trend. They were not considered deviation, as they certainly would have been during the earlier period
(189) of intergroup hostility when a friendly gesture would have been regarded as "indecent."
Conformity-Deviation and Social Change in the Larger Setting
Conformity and deviation have been discussed in this paper in terms of normative process in small groups spontaneously organized by individuals with motives and problems seen as common, whether these motives relate to common conditions of the natural settings or to experimentally created problem situations. In such groups, whatever the organizational form and the character of the norms may be, they are not handed down by an outside authority. They are outcomes of the interaction process among the individual members, of course, as influenced by their previous group memberships and the setting in which they function. Each individual attains a position in the group through his relative contribution and efforts, and his position is not fixed once and for all. Each had some hand in shaping the norms of the group and can have a hand in changing them through further interaction.
Such is not the case in traditional groups and formally organized groups in the larger cultural setting. We cannot adequately touch upon the questions raised at the beginning of this paper without at least mentioning conformity and deviation as they occur in such organizational frameworks.
LARGER SETTINGS. Many organizations in the larger setting are handed down through the generations and some are deliberately organized with blueprints, rules, and regulations put down on paper by personages or governing bodies in power. As a result many individual members have had nothing to do with shaping the organizational patterns and normative system. Not infrequently, the norms and organizational forms, which might have been appropriate under
( 190) conditions at the time of their appearance, prevail now more through the heavy hand of tradition or through active efforts of those interested in their perpetuation.
The larger setting is further confounded by multiple groups in which individuals may have overlapping memberships, and by sizable power differentials between various groups in the scheme of things. In the highly differentiated modern scene, the individual may be faced with contradictory, even conflicting modes of compliance from his multiple groups. He may also experience in a personal way the power differential between his own groups and others in the setting.
Problems suggested by formal organizations and multiple groups have been investigated especially by sociologists---Durkheim, the Lynds, Merton, and Williams, to mention only a few. Here I can only cite a few of the psychological questions these topics impose upon us without marshalling factual evidence.
We may note, first of all, that when normative systems are upheld by the heavy hand of tradition and coercion, with little voluntary acceptance by the bulk of the individuals involved, informal groups usually arise within the formal outfit, as has been observed in industrial and military organizations. In such cases, the individual's compliance with inner acceptance is to norms of the informal groups to which he actively contributes. It may be suggested as a hunch to be tested that the greater the appeal of informal groups within formal organizations, the greater the secrecy observed in maintaining the existence of informal groups, the greater is the likelihood that the norms and rules of the formal organization have lost their effectiveness in regulating the behavior of the individuals involved.
The formal organization is one of the settings, so prevalent in modern differentiated societies, in which an individual's membership group (in this case, the formal organization) may not be his reference group, that is the group to which
( 191) he relates himself psychologically or aspires to belong. Other common instances of discrepancies between the individual's membership groups and his reference groups owe their existence to the anchoring effects of the more powerful groups for individuals situated at lower levels in a setting which sanctions upward mobility.
CONFLICT AND THE INDIVIDUAL IN THE GROUP. The individual living his existence in membership groups with set modes of conformity while simultaneously relating himself psychologically to other groups or aspiring to belong to other groups is bound to experience frustration and personal conflict, at least at times. Many a man, dissatisfied and frustrated in his association with a given group, in which whatever compliance he observes is through coercion or fear of coercion, strives to break away from it and its norms. However, the efforts toward the break are not the end point in this picture. Usually the break, when it is made, is from one constellation of human relationships toward another. Thus, resistance and tendencies to deviation from the norms of one group are supplanted by active searching for new anchorages which the individual can accept as his own.Perhaps this picture may have some relevance for issues of change. It suggests that active nonconformity to the norms of a membership group which require coercion for their observance, which arouse frustration in man and conflict between men rather than understanding and cooperation, is but one part of the picture. The completed picture is not a state of normlessness, but a search for normative bases which are not conflicting, not frustrating, not self-degrading to the individual.
The elevation of nonconformity alone to the level of a slogan has its roots in an untenable dichotomy between individual and group, in a preconceived inevitability of clash between the two. Such a stand cannot account for changes
( 192) in human relationships or their normative bases.
If normlessness is not the end result of social change, then the human sciences face a more vital and difficult task than merely demonstrating the blinding and degrading effects of certain norms. The task requires some valid criteria for singling out prevailing norms which produce these effects and for deliberate choice of values and organizational forms conducive to behavior which is enriching and "self-actualizing" for individuals in their social relationships.
Summary and Conclusions
In all phases of his daily living-social, political, economic, religious-man is confronted today with pressures and exhortations to regulate his behavior within advocated molds and directions.
Concerned with the plight of those of their fellow men who have fallen prey in blind conformity to such pressures and exhortations, psychologists and social scientists have advanced various antidotes. One attempt in this direction is the cult which almost elevates nonconformity in its own right to a pedestal of virtue. Such attempts are laudable from the point of view of intent. But their realistic adequacy as effective measures is a different matter.
Even the most ardent proponent of nonconformity would not praise stealing simply because it is a nonconforming deed. Indeed, conformity is condemned because of concern over widespread degradation of moral and artistic values, of repressive restrictions on human expression and the rights of man, of arbitrary limitations to human dignity and potentialities. In short, the plea for nonconformity is made in the name of values or norms which were themselves formulated through a long and arduous stretch of human history. But as long as the analysis and the plea center on conformity or nonconformity in the abstract, there is no adequate basis for evaluating conforming or nonconforming behavior. And the
( 193) exhortation to individual man to assert his independence of a mountain of social pressures may not be an adequate way to move the mountain.
An item of behavior, whether in social, political, religious, or economic spheres, cannot be characterized by itself as either conforming or deviating. It is always conformity or deviation relative to some premise, canon, standard, or value -in short, to some norm. Therefore, the primary question to be raised becomes: "Conformity or nonconformity in relation to what practice, what value, what moral standard, or what norm?"
Social values, moral standards, or norms are products of interaction among human beings over a period of time in matters of consequence to their mutual and individual concerns. Thus, issues of conformity and nonconformity which make the problem so urgent for study are not one-episode affairs involving momentary, transitory social influence on inconsequential matters. For these reasons, the traditional laboratory setting is far from adequate for studying significant problems of conformity and deviation.
On the whole, the traditional laboratory experiment takes the individual from a context of relationships with other people involving matters of mutual importance to expose him briefly to a momentary situation arranged by the experimenter. There is sufficient evidence to conclude that the experimenter's success in demonstrating susceptibility to conform in the laboratory is inversely related to the degree of structure, the number and clarity of alternatives available in the stimulus situation the experimenter has arranged for the subjects. Thus in a highly unstructured situation with various alternatives, the subject may conform almost invariably to definitions introduced by the experimenter. To conclude, therefore, that he is basically a conformist would be quite in error. Likewise, demonstration of righteous protests by individuals exposed to easily discriminable stimuli and a
( 194) fantastically false consensus by others is scarcely evidence for man's basic independence.
The attribution of either blind subservience to the group or independence to the basic nature of man rests on an untenable dichotomy between individual and group. An adequate approach must begin with a clear statement of the place of norm-regulated behavior in psychological functioning. Even a single individual faced with a perceptual situation for which he has little in the way of established guideposts for evaluation, comes in time to a stabilized mode of behavior. as experiments have shown. At the basis of this tendency toward stabilization lies man's capacity to regulate his behavior through conceptual categories.
Normative regulation of behavior is also conceptual regulation. and without it, human morality of any kind would be inconceivable. It is precisely the normative regulation of behavior that permits human action to transcend demands of the immediate social surroundings, as well as to conform to them, to delay momentary impulses, and to resist the promptings of transitory emotional states. Normative regulation. properly conceived, is not alien to man's nature, but one of its distinctive features. In specific situations, the process may lead either to conformity or nonconformity.
Those canons, values, or norms relative to which conformity or nonconformity become important problems are integral aspects of man's relations with other men. These relations involve lasting expectations, reciprocities, and responsibilities, and their patterns constitute the individual's group relations. Thus the realistic, the significant setting for study and analysis of conforming behavior is the setting of human groups, large and small.
Some kind of norm system is one of the essential features of any human group, be it a club or church, a sect or professional group, or even a group whose major tenets involve nonconformity to prevailing societal norms.
( 195)
When man enters into repeated interaction with others, directed toward similar concerns and goals, he takes part in a process of norm formation and stabilization. Once a normative system is stabilized, the individual member who took part in its creation regulates his behavior within the bounds it defines as acceptable, even without external pressures and sanctions. Both our research and considerable empirical findings demonstrate that the bounds defined by group norms are perceived by the single member as his own latitude of acceptable behavior. Behavior outside of this range is evaluated as objectionable. Henceforth, social control is achieved in part through the autonomous regulation of behavior by individual members. And when it relies exclusively on external pressure and coercion, changes in the normative structure may be anticipated.
Therefore, the question of concern to those who are disturbed with the plight of men caught in pressures toward certain molds of conformity should not be evaluation of conformity or nonconformity in the abstract. The first question must be conformity to what norm? Answers to this question entail not only the external referent of the behavior. but the context of group relationships in which it occurs and the voluntary or coercive nature of its regulation. Then analysis of the appropriateness or inappropriateness of the norm in question for the situation and in terms of other criteria may begin with an adequate basis. If on this basis, norms are found inappropriate, a related task becomes discovering the processes leading to perpetuation of dysfunctional norms, including notably, interested parties engaged in active efforts to that end.
A closely related task is assessing the demands placed upon the individual by conflicting norms-for example, norms of altruism preached on Sunday and norms for the hard facts of business and professional practice. The existence of mutually conflicting values or norms characteristic of highly
( 196) differentiated Western societies today, as various social scientists have pointed out, is responsible in no small part for psychological conflict with attendant wear and tear and restlessness so widely reflected in contemporary novels and social science literature.
If, as it would seem, the interdependencies of human development and human groups are becoming increasingly closer and wider in scope, then the analysis demands a flood of light upon the consequences of maintaining obsolete, constrictive norms perpetuated through ethnocentrisms and activities of particularly interested groups. The appropriate changes in norms are, of course, part of the problem of social change. If the social scientist or psychologist backs away from this problem, he is backing away from the course of intellectual history. For good or evil, human relationships and their norms have changed and they will change. The challenge of understanding the process and the directions it takes must be met if we are seriously concerned with man's creative development and larger self-fulfillment. The two are not independent.
REFERENCES
Asch, S. E. Effects of group pressures upon the modification and distortion of judgments. In G. E. Swanson, T. M. Newcomb, and E. L. Hartley (Eds.), Readings in social psychology. (2nd ed.) New York: Holt, 1952.
Coffin, T. E. Some conditions of suggestion and suggestibility: A study of certain attitudinal and situational factors influencing the process of suggestion. Psychol. Monogr., 1941, No. 241.
Cohen, Edwin Stimulus conditions as factors in social change. Sociometry, 1957, 20, 135-144.
Faris, R. E. L. Development of the small group -research movement. In M. Sherif and M. O. Wilson (Eds.), Group relations at the crossroads. New York: Harper, 1953.
( 197)
Freedman, R., Hawley, A. H., Landecker, W. S., & Miner, H. M. Principles of sociology. New York: Holt, 1952.
Graham, C. H. Behavior and the psychophysical methods: An analysis of some recent experiments. Psychol. Rev., 1952, 59, 62-70.
Nelson, H. Adaptation level theory. In S. Koch (Ed.), Status and development of psychology. Vol. 2. New York: McGraw-Hill, 1959.
Homans, G. C. The human group. New York: Harcourt. Brace, 1950.
Hood, W. R., & Sherif, M. Appraisal of personality-oriented approaches to prejudice. Sociol. soc. Res., 1955, 40, 79-85.
Hood, W. R., & Sherif, M. Verbal report and judgment of an unstructured stimulus situation. Paper read at Southwestern Psychological Assn., Little Rock, April, 1957.
Hovland, C. I., Harvey, O. J., & Sherif, M. Assimilation and contrast effects in reactions to communication and attitude change. J. abnorm. soc. Psychol., 1957, 55, 244-252.
Jackson, Jay M. Structural characteristics of norms. In G. E. Jensen (Ed.), Dynamics of instructional groups. Chicago: Univer. of Chicago Press, 196o. (Mimeographed)
Johnson, D. M. The psychology of thought and judgment. New York: Harper, 1955.
Luchins, A. S. On agreement with another's judgment. J. abnorm. soc. Psychol., 1944, 39, 97-111.
McGarvey, H. R. Anchoring effects in the absolute judgment of verbal materials. Arch. Psychol., 1943, No. 281.
Mausner, B. Studies in social interaction. III. Effect of variation of a partner's prestige on the interaction of observer pairs. J. appl. Psychol., 1953, 37, 391-394.
Mausner, B. The effect of prior reinforcement on the interaction of observer pairs. J. abnorm. soc. Psychol., 1954, 49, 65-68.
Sherif, M. A study of some social factors in perception. Arch. Psychol., 1935, No. 187.
Sherif, M. The psychology of social norms. New York: Harper, 1936.
Sherif, M. An outline of social psychology. New York: Harper, 1948.
( 198)
Sherif, M. Operational report to the Hogg Foundation for Mental Health, The University of Texas. Description of research work on natural groups (1958-1959). (Mimeographed)
Sherif, M., & Cantril, H. The psychology of ego-involvements. New York: Wiley, 1947.
Sherif, M., & Harvey, O. J. A study of ego functioning: Elimination of stable anchorages in individual and group situations. Sociometry, 1952, 15, 272-305.
Sherif, M., Harvey, O. J., White, B. J., Hood, W. R., & Sherif, C. W. Study of positive and negative intergroup attitudes between experimentally produced groups: Robbers Cave study. Univer. of Oklahoma, 19J4. (Multilithed)
Sherif, M., & Sherif, C. IV. Groups in harmony and tension. New York: Harper, 1953.
Sherif, M., & Sherif, C. WV. An outline of social psychology. (rev. ed.) New York: Harper, 1956.
Sherif, M., & Sherif, C. W. Self-radius and goals of group members and their age-mates in differentiated sociocultural settings. Report to Institute of Group Relations, Univer. of Oklahoma, 1960. (Mimeographed)
Stevens, S. S. On the psychophysical law. Psychol. Rev., 1957 64, 153-181.
Thrasher, J. D. Interpersonal relations and gradations of stimulus structure as factors in judgmental variation: an experimental approach. Sociometry, 1954, 17, 228-241.
Tresselt, M. E., & Volkmann, J. The production of uniform opinion by non-social stimulation. J. abnorm. soc. Psychol., 1942 37 234-243.
Volkmann, J. Scales of judgment and their implications for social psychology. In J. H. Rohrer and M. Sherif (Eds.), Social psychology at the crossroads. New York: Harper, 1951.
Walter, N. A study of the effects of conflicting suggestions upon judgments in the autokinetic situation. Sociometry, 1955 18, 138-146.
Whittaker, J. O. Effects of experimentally introduced anchorages upon judgments in the autokinetic situation. Paper read at the International Congress of Psychology, Brussels, June, 1957; Ph.D. Thesis, Univer. of Oklahoma, 1958.
