THE DEVELOPMENT AND DIFFERENTIATION OF ROLE EXPECTATIONS IN CHILDREN AND ADOLESCENTS Dissertation for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY RAYMOND JOSEPH MONTEMAYOR 1974 . ,WWmt-Ive ‘0 L 13 R A R Y Michigan State University This is to certify that the J": i thesis entitled ; ’fi :5 The Development and Differentiation ":3. of Role Expectations in Children and Adolescents presented by Raymond Joseph Montemayor has been accepted towards fulfillment of the requirements for PhD degree in Psy_chologz Major professor Date‘7é_é> ../ 9‘47; 7% V 0-7639 IIIIIIIIIIIIIIIIIIIIIIIIIII 1293 10674 8969 I"! ' "' / P I"; ‘r-1 . ‘I . . V'M- 4-- '4" ~ I'“. w _ _ .‘ »%5<- , mm are ' - ABSTRACT THE DEVELOPMENT AND DIFFERENTIATION OF ROLE EXPECTATIONS IN CHILDREN AND ADOLESCENTS BY Raymond Joseph Montemayor The primary purpose of this study was: (a) to propose a model of socialization that was not based on personality development and/or an assumption of cross- situational behavioral consistency; and (b) to examine empirically the usefulness of that model. The model presented was based on role theory and reference figure analysis. It was assumed that the social world is organized by individuals into categories or social roles. Further, behavioral expectations were seen to vary as a function of the role in which the individual is acting. Three dimensions for classifying roles were chosen for study. The behavioral expectations for boys and girls at five different ages were examined as a function of: (l) the sex of other, (2) the status of other, and (3) whether the self was the subject or object of the behavior. A 40-item questionnaire was administered to 100 males and 100 females between the ages of 9 and 18 years. Raymond Joseph Montemayor The students were asked about behavioral expectations for themselves and others when interacting with their father, mother, boy friend, or girl friend. For example, two test questions were: "How often do you think you should tell the truth to your mother? How often do you think your mother should tell the truth to you?" The design for this study was a 2 x 5 x 2 x 2 x 2 x 5 analysis of variance. The six factors and their levels were: Sex of self (male, female); Age of self (9-10, 11-12, 13-14, 15-16, 17-18): Source (self, other): Status of other (parent, peer); Sex of other (male, female): Norm (share, tell the truth, do what is said, tell another what to do, ask for help). The data indicated that children make a strong distinction between parents and peers, and have higher standards of interpersonal conduct for themselves when they are interacting with their parents than when they are interacting with their peers. Although sex of other was not as salient a dimension as status of other, it did seem to influence interpersonal expectations for one- self in certain specific situations: Boys expected to share more often with other boys than with girls; and in general, boys had higher expectations for their own behavior with other boys than with girls. Girls expected to share with and ask for help from other girls more than boys, and also had generally higher expectations Raymond Joseph Montemayor for same-sex others than for opposite-sex others. Chil- dren also distinguished between expectations for self and expectations for another for the norm "share," "do what is said," and "ask for help." The distinction was pri- marily a parent-self division and was most clear in the norms "do what is said" and "tell another what to do.” Adolescents showed a somewhat similar structuring of their social world, although there were important dif- ferences. Adolescents generally did not differentiate between parent standards and peer standards. Expectations for the self were similar for those two groups. This did not seem true in the case of the norm "do what is said." Adolescents expected to do what their parents told them significantly more often than they expected to do what their friends told them. Sex differences, where they appeared, interacted with the sex of the other and were similar to the findings for children. Adolescents had somewhat higher expectations for their own behavior when they expected to interact with someone of the same sex. The distinction between self-expectations and other expectations was most evident in the parent-self dyad. Adolescents expected their parents to tell them the truth more often than they expected to tell the truth to their parents. Adolescents also expected to be told what to do and to do what they were told more often than they expected their parents to perform these behaviors. Raymond Joseph Montemayor Finally, adolescents applied the self-other distinction most often according to the following ranking: self- father, self-mother, self-girl friend; self-boy friend. The results were related to findings in the area of moral judgment, parent-peer conflicts, and sex-role identification. It was postulated that a fundamental reorganization of the child's social world occurs in adolescence as a function of the changing status of the self in relation to parents and peers. THE DEVELOPMENT AND DIFFERENTIATION OF ROLE EXPECTATIONS IN CHILDREN AND ADOLESCENTS BY Raymond Joseph Montemayor A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1974 ACKNOWLEDGMENTS I am deeply indebted to John McKinney for his guidance and instruction throughout the course of this research. Additionally, I would like to thank John for his patience and enthusiasm during the past four years. My excitement about the enterprise of developmental psy- chology is in large measure the result of my association with John. I would also like to thank Lucy Ferguson, Larry Messe, Frank Schmidt, and Robert Zucker for their critical readings of this dissertation. Grateful appreciation is also expressed to Dr. Robert Docking of the East Lansing Board of Education. I am also grateful to all the teachers and students of the East Lansing School District who cooperated so willingly throughout the study. ii TABLE OF CONTENTS Chapter Page I. INTRODUCTION . . . . . . . . . . . 1 Role Theory . . . . . . . . . . The Development of Role-Taking Skills. . 10 Socialization as Role Learning . . . . 13 Hypotheses. . . . . . . . . . 16 II . METHOD 0 O O O O O O O O O O O 0 23 subjects 0 O O O O O O O O O O 23 Differentiation of Norms . . . . . . 25 Selection of Norms . . . . . . . . 25 Test Questionnaire . . . . . . . . 26 Procedure . . . . . . . . . . . 27 DeSign O O O O O O O O O O O O 28 III . RESULTS 0 O O O O O O O O O O O O 30 A. Overview of Data Analysis. . . . . 30 B. Role Expectations for the Self . . . 31 1. Developmental Changes . . . . . 31 a. Adherence to social norms. . . 31 b. Norm ranking . . . . . . . 34 c. Parents and peers . . . . . 36 d. Summary of developmental changes for expectations for the self . . . . . . . . 45 2. Sex Differences . . . . . . . 47 a. Norm applicability . . . . . 48 b. Male and female others. . . . 49 c. Parents and peers . . . . . 52 d. Summary of sex differences for expectations for the self. . . 53 iii Chapter Page 3. Summary of EXpectations for the self 0 O O O O O O O O O O 55 C. Comparison of Role Expectations for the Self and Other . . . . . . . 56 1. Developmental Changes . . . . . 60 a. Self-other expectations. . . . 60 b. Parents and peers. . . . . 60 c. Summary of developmental changes for the self-other comparison. . 73 2. Sex Differences . . . . . . . 76 a. The importance of status, norm, and sex of other . . . . . . 77 b. Summary of sex differences for the self-other comparison . . . 79 D. Summary and General Conclusions . . . 81 IV. DISCUSSION 0 O O I O O O O O O O O 84 APPENDICES Appendix A. PARENT PERMISSION LETTER . . . . . . . 96 B. EXTREME RESPONSE BIAS . . . . . . . . 97 C. TEST QUESTIONNAIRE . . . . . . . . . 100 D. ANALYSES OF VARIANCE FOR SIMPLE EFFECTS . . 107 REFERENCES 0 O O O O O O O O O O O O O 121 iv Table 10. 11. LIST OF TABLES Subject Characteristics . . . . . . . Analysis of Variance for Self-Expectations. Mean Self-Expectation Scores for Males and Females for Each Norm . . . . . . . Mean Self-EXpectation Scores for Males and Females for Male and Female Others. . . Mean Male and Female Self-Expectation Scores for Male and Female Others for Each Norm. Mean Male and Female Self-Expectation Scores for Parents and Peers for Each Norm . . Analysis of Variance for Expectations for Self and Other . . . . . . . . . Mean Expectation Scores for Self and Father for Each Norm. . . . . . . . . . Mean Expectation Scores for Self and Mother for Each Norm. . . . . . . . . . Mean Expectation Scores for Self and Boy Friend for Each Norm . . . . . . . Mean Expectation Scores for Self and Girl Friend for Each Norm . . . . . . . Analysis of Variance for Simple Effects (Age x Status) . . . . . . . . . Analysis of Variance for Simple Effects (Age x Status) . . . . . . . . . Analysis of Variance for Simple Effects (Age x Status x Norm [Share]) . . . . Page 24 32 49 51 51 54 57 78 78 80 80 107 107 108 Table D-4 o D-ll o D-lz 0 13-13 0 13-14 0 0-15. D-l6 o D-17. 0-180 Analysis (Age x Analysis (Age x Analysis (Age x Analysis (Age x Analysis (Sex x Analysis (Sex x Analysis of Variance for Simple Effects Status x Norm [Tell Truth]) . . of Variance for Simple Effects Status x Norm [Do What Is Said]) of Variance for Simple Effects Status x Norm [Tell What To Do]) of Variance for Simple Effects Status x Norm [Ask for Help]) . of Variance for Simple Effects Norm) o o o o o o o o o of Variance for Simple Effects MF) 0 o o o o o o o o o of Variance for Simple Effects (Sex [Males] x Norm x MF) . . . . . Analysis of Variance for Simple Effects (Sex [Females] x Norm x MF) . . . . Analysis of Variance for Simple Effects (Sex [Males] x Status x Norm). . . . Analysis of Variance for Simple Effects (Sex [Females] x Status x Norm) . . . Analysis of Variance for (Age x Source). . . Analysis of Variance for (Age x Source x Status Norm [Share]) . . . Analysis of Variance for (Age x Source x Status [Share]) 0 o o o 0 Analysis of Variance for (Age x Source x Status [Tell Truth]) . . . Analysis of Variance for (Age x Source x Status [Tell Truth]) . . . vi Simple Effects Simple Effects [Parents] x Simple Effects [Peers] x Norm Simple Effects [Parents] x Norm Simple Effects [Peers] x Norm Page 108 109 109 110 110 111 111 112 112 113 113 114 114 115 115 Table Page D-l9. Analysis of Variance for Simple Effects (Age x Source x Status [Parents] x Norm [Do What Is Said]) . . . . . . . . 116 D-20. Analysis of Variance for Simple Effects (Age x Source x Status [Peers] x Norm [Do What Is Said]) . . . . . . . . 116 D-21. Analysis of Variance for Simple Effects (Age x Source x Status [Parents] x Norm [Tell What To D01) . . . . . . . . 117 0-22. Analysis of Variance for Simple Effects (Age x Source x Status [Peers] x Norm [Tell What To D01) . . . . . . . . 117 D-23. Analysis of Variance for Simple Effects (Age x Source x Status [Peers] x Norm [Ask for Help]) . . . . . . . . . 118 D-24. Analysis of Variance for Simple Effects (Source x Norm [For Fatherl). . . . . 118 D-25. Analysis of Variance for Simple Effects (Source x Norm [For Mother]). . . . . 119 D-26. Analysis of Variance for Simple Effects (Source x Norm [For Boy Friend]) . . . 119 D-27. Analysis of Variance for Simple Effects (Source x Norm [For Girl Friend]) . . . 120 vii Figure 1. Age changes for expectations for self. Normative contents combined . . . . 2. Differentiation of self-expectations for five normative contents . . . . . . . 3. Differentiation of self-expectations for parents and peers. Normative contents combined . . . . . . . . . . 4. Differentiation of self-expectations for parents and peers for the norm "share" 5. Differentiation of self-expectations for parents and peers for the norm "tell the truth" 0 O O I O O O I O O O 6. Differentiation of self-expectations for parents and peers for the norm "do what is said" O O O O O O O O O O 7. Differentiation of self-expectations for parents and peers for the norm "tell another what to do". . . . . . . 8. Differentiation of self-expectations for parents and peers for the norm "ask for help" 0 O I I O O O O O I O 9. Age changes in expectations for self and others. Normative contents combined . 10. Differentiation of expectations for parents and peers for the norm "share" . . . ll. Differentiation of expectations for peers LIST OF FIGURES and self for the norm "share" . . . viii Page 33 35 37 40 42 43 44 46 61 63 64 Figure 12. Differentiation of and self for the 13. Differentiation of - and self for the 14. Differentiation of and self for the 15. Differentiation of and self for the 16. Differentiation of and self for the to do". . . . 17. Differentiation of and self for the to do". . . . 18. Differentiation of and self for the 19. Differentiation of and self for the expectations for parents norm "tell the truth" . expectations for peers norm "tell the truth” . eXpectations for parents norm "do what is said". expectations for peers norm "do what is said”. expectations for parents .norm "tell another what expectations for peers norm "tell another what eXpectations for parents norm "ask for help". . expectations for peers norm "ask for help". . ix Page 66 67 69 70 71 72 74 75 CHAPTER I INTRODUCTION Traditional personality theories have dealt with behavioral organization at the most general level. A "trait" is defined and individuals are ranked along a continuum of relative possession of the trait. However, a number of analyses have indicated that this type of approach gives a very poor level of prediction from one situation to another (Mischel, 1968). For example, Hart- shorne and May (1928) studied moral conduct and concluded that honesty in one situation is a very poor predictor of honesty in another situation. "Thus as we progressively change the situation we progressively lower the correlations between the test [p. 384]." Recent theorizing within psychology has attempted to integrate explanations of behavior which stress either personality factors or situational conditions (Bowers, 1973; Emmerich, 1973; Mischel, 1973). Clearly, behavior is, as Lewin noted years ago, some function of the person and the situation (Lewin, 1951), and a great deal of research indicates that the person x situation interaction accounts for more of the behavioral variance than either the person or situation alone (Endler, 1973; Endler & Hunt, 1968, 1969; Argyle & Little, 1972). However, stating the precise nature of the person and situation interaction has remained one of psychology's more enduring and elusive problems; and the major difficulty with find- ing a solution to this problem is due more to its enormously complex conceptual nature than to a lack of relevant data. More precisely, the problem is that personality factors and situational conditions represent two distinct conceptual domains. The question, quite simply, is: how does the situation "get into" the person so that it can influence his behavior? Stated another way, the problem is to define the boundaries of the person system and the social system and to describe how these two sys- tems can influence each other. For example, to say that a certain group of children will choose an immediate reward rather than wait for a later reward because they are of a lower socioeconomic class does not explain why the behavior occurs, but relates it to a whole range of other factors, such as father's education and income, a certain urban environment, a particular kind of parent- child interaction system, etc. Conversely, to say that the behavior occurs because the children have been rewarded in the past for accepting an immediate gratification or because the children have low impulse control does not explain why these behaviors occur so frequently in groups of peeple classified as low socio- economic status. Social scientists do not yet possess a single set of conceptual tools that will allow them to build a theoretical structure encompassing both personal and situational factors. Social theories as diverse as Marxian economics (Feuer, 1959), symbolic interactionism (Manis & Meltzer, 1967; Rose, 1962), and cognitive social psychology (Lewin, 1951) have attempted to explain the impact of the situ- ation on the individual in terms of the "consciousness“ that arises from the situation or the meaning or per- ception of the situation. These approaches assume that the social world has no real meaning apart from the various meanings attributed to it by individuals (Gar- finkel, 1967). These approaches view the human actor as an active agent who attempts to control and make sense out of situations and who directs his own behavior as a result of this construction. These approaches have in common a concern for the "meaning" of the situation from the participant's point of View. By focusing on the per- ception of the situation these social scientists are able, to some extent, to integrate aspects of person and situation into a single conceptual domain. This approach to social interaction, more a point of view than a theory, is the approach of the present author. The present study is an investigation of the development of the meanings of certain social situations. The concept of a situation is used in this study in the limited sense of a two-person interpersonal situ- ation. Specifically, four interpersonal dyads will be considered: (1) father-self, (2) mother-self, (3) same sex peer-self, and (4) opposite sex peer-self. This study will examine the meaning that these four situations have for males and females between the ages of 9 and 18 years. Meaning refers to the behavioral expectations that one has for oneself and for the other member of the dyad. For example, how does a child expect to behave when he is with his father and how does this child expect his father to behave towards him. A measure of these expectations is considered to be a measure of the mean- ing of the father-child situation for that child. It has been found that the meaning of a situation is very often shared by groups of individuals who have similar life circumstances. For example, Rabban (1950) found that lower-class children defined masculinity and femininity much more stereotypically than did middle- class children. Lower-class children defined appropriate masculine behavior as assertive and domineering, while females were viewed as nurturant and supportive. Middle- class children defined masculinity and femininity much less extremely. Emmerich (1961) found that the perception of the family situation changed with age. All children saw the father as more powerful than the mother, although this power differential was perceived to be less extreme in middle childhood than in early or late childhood. These studies and others indicate that groups of indi- viduals have common meanings for certain social situations. When the expectations for a certain interpersonal situation are shared by members of a group, we will refer to those expectations as a ”social norm" (Secord & Back- man, 1974). Thus, in this very restricted sense of the word, a social norm is simply a statistical average of social expectations for a group. Emmerich (1959) found that children expect adults to be powerful and they, themselves, to be obedient. The norm for the average child, therefore, is to have adults tell him what to do and to do what adults tell him to do. The concept of norm as used by most sociologists is more extensive than a simple averaging of behavior. A social norm is a rule of conduct (Homans, 1950). As a prescription for behavior two features of social norms are especially important. First, norms allow individuals to anticipate their own behavior and the behavior of others. Second, norms are obligatory in nature. The implications for the present study of these two aspects of norms will be explored in the sections to follow. Role Theory With these beginning observations complete, let us turn directly to the task of presenting the analysis of personality development as role-learning. The simplist form of a social system is a two- person system, or dyad. When two peOple interact, each seeks to obtain certain rewards from the other and to obtain certain goals through the other. The behavior of each person, therefore, can be viewed as an instru- mental attempt to elicit from the other behavior which the first person will find rewarding. This reward does not necessarily have to be a tangible object. The reward may simply be a successful prediction of what the other person will do. In Parson's language, a dyad consists of egg and alter (Parsons, 1951). Ego's problem is to predict how alter will respond, in order to elicit the desired responses from him. Alter's problem is the same. If the relationship is a long-term one or if it is a very formal one, both ego and alter should possess the ability to predict accurately the other's behavior. Alter's expectations for ego's behavior should correspond to ego's expectations for his own behavior, i.e., these expectations should be mutual. Alter's expectations for his own behavior and ego's expectations for his own behavior should be complementary. For example, both adults and children expect children to be obedient. Additionally, adults expect to tell children what to do and children eXpect to be obedient. Since the expec- tations in this situation are both mutual and complemen- tary, we may expect the interaction to be successful and rewarding for both members of the dyad. Since an individual's own behavior in a social situation is contingent upon the behavior of the other, a successful social interaction depends upon the ability of each member of the dyad to anticipate the behavior of the other. To the extent that this anticipation is wrong, i.e., one person does not act in accordance with the accepted social norm, the interaction will not be rewarding. The other person is not only expected to behave in a certain way; it is believed that he must behave in a certain way. This follows from the fact that in order to maximize one's own rewards one must be able to predict consistently the behavior of the other person. Social norms are not just guidelines to behavior; they are rules of social conduct. Those individuals who violate those rules will probably suffer some form of social punishment. It is clear that a single interpersonal norm does not equally apply to all individuals. Children do not expect their parents to behave as they themselves do, nor do children expect to behave similarly in all interpersonal situations. Sociologists say that the applicability of a norm in an interpersonal situation depends upon the status of the individual. Status is defined as the position or identity of an individual within a social structure (Davis, 1949). This identity may be the result of ascribed characteristics such as age and sex, and characteristics which are achieved such as education and wealth. Some aspects of status change from situation to situation, while some remain constant. Although sex remains the same, the status of daughter is different from the status of girl-friend and the norms that apply to a girl acting as a daughter are not necessarily the same norms that apply to the girl act- ing as a girl friend (Goffman, 1959). The preceding discussion points to two important properties of status and norm. First, status and the norms that apply to that status change from situation to situation. Individuals do not occupy a single status but many statuses. Consequently the norms and expec- tations that apply to a person's behavior change from situation to situation. Therefore, it should not be unusual to expect that behavior will also differ between situations. Second, status and norm are closely associated. The two concepts are necessary components of any complete description of behavior. It is this last point that has led sociologists to develOp the important notion of social role. The concept of role occupies a central position within sociology, although it lacks a precise and uni- versally agreed-upon definition. Reviews of role definitions have indicated a multiplicity of meanings (Neiman & Hughes, 1951; Rommetviet, 1954; Thomas, 1969). The concept of role has been used to denote evaluations, descriptions, and prescriptions; and it has referred to both overt and covert processes (Biddle & Thomas, 1966). As it is used in the present study, a role is a generally agreed-upon belief about how a person who occupies a par- ticular status should act. For example, the role of the father is that normative behavior which the culture as a whole expects a man who occupies the status of head of household to perform. Thus the concept of role requires both status and norm. For example, a man who cares about children is not necessarily acting in the role of a father, although he is fulfilling one of the norms of fatherhood, since he could occupy some other status such as teacher. Conversely, a man who is the head of a household but who does not care about his children is also not acting in the role of father since he is violat- ing one of that role's most crucial norms. The present study is not concerned with role behavior but with role expectations. Role expectations 10 are defined as those expectations that an individual has for himself and others in different interpersonal situ- ations. Role expectations are concerned with an indi- vidual's conception of a role. A measure of an indi- vidual's expectations for a particular role is a measure of how that person views the relevance of a particular norm for a certain status. "How often should you tell the truth to your mother; how often should your mother tell the truth to you?" These two questions investigate ego's conception of the role of son and mother for the norm "to tell the truth." As such, these questions are concerned with the meaning of the son-mother situation for the son. Only recently has systematic attention been given by psychologists to the acquisition of roles by children (Brim, 1960; Brim & Wheeler, 1966). Most studies of socialization have not had this particular phrasing of the problem in mind. The Development of Role-Taking Skills A small body of research has dealt with the acquisition of role-taking skills in young children (Chandler, 1973; Feffer, 1970; Feffer & Gourevitch, 1960; Selman, 1971). This work was influenced by Piaget's investigations of the young child's acquisition of the ability to represent the perceptual experience of another individual when that experience differs from 11 his own (Piaget & Inhelder, 1956). The most interesting finding was that young children confuse their own per- spective with the perspective of another. Additionally, Piaget's data on moral development (1932) emphasize the young child's insensitivity to the inner motives and intentions of others, that is, his "egocentrism." Flavell and his associates have carried out the most extensive investigation of the acquisition of role— taking skills in children (Flavell, 1966a, 1966b; Flavell et a1., 1968). As Flavell defines it, role-taking may be thought of as something akin to "person perception" or "social cognition." Basically, it is the ability to take the perspective, or motives, or feelings, or point of view of another. Flavell distinguishes between per- ceptual role-taking, the ability to see something from another perspective, and conceptual role-taking, the ability to understand something from another point of view. Flavell has demonstrated that both abilities are lacking in most three-year-olds and that the development of role-taking skills follows a sequence from simple to complex and from subjective to objective. Although Flavell and his co-workers are concerned with roles, that concern is quite different from the sociological interest in the term. Flavell is inter— ested in the child's ability to take the role of another. Sociologists are interested in the behavior 12 that is expected of a person when he occupies one status as opposed to another. Consider the father-son relation- ship as an example. Flavell would be interested in the child's ability to interpret his own behavior from his father's point of view. And in this respect, Flavell is interested in a child's ability to take the role of his father. Sociologists, on the other hand, are con- cerned with how sons are eXpected to behave in relation to their father. And in this respect, they are inter- ested in the role of the son. Thus, although a body of research has dealt with the child's acquisition of role- taking skills in social interactions, these studies have not investigated the specific role requirements necessary for effective social interaction with peers, parents, or both. Nor have these studies examined how an individual comes to acquire a status system by which he is able to distinguish and differentiate his roles and the roles of others. Maccoby (1961) has outlined the parameters of these issues and has indicated what some of the critical issues are, and Sewell (1963) has reviewed some of the studies that have this orientation. Finally, Brim (1966) has provided a framework for the analysis of socialization in which the acquisition of roles occupies a central position. 13 Socialization as Role Learning The basic premise of Brim's argument is that most of what is learned from socialization in childhood is a series of complex interpersonal relationships. A child interacts with a variety of individuals who have great influence over his behavior because of their frequency of contact, status, and control over rewards and punish- ments. As a result of these interactions, the child learns a series of "self-other systems" in which he becomes oriented toward the expectations of significant others about his own behavior. As Brim states: The individual learns the behavior appropriate to his position in a group through interaction with others who hold normative beliefs about what his role should be, and who reward or punish him for correct or incorrect actions. A number of social scientists have emphasized the importance of normative expectations as a determinant of behavior (Asch, 1952; Mead, 1934; Sarbin & Allen, 1968). Social situations contain within them certain rules of behavior, called norms, which serve to mediate and guide the behavior of the participants. It is believed that knowledge of and obedience to these norms help define the constancies that one observes in these situations. Since norms become internalized during socialization as a result of self—other interactions, it follows that these norms should be differentiated in ways that take into account the relative status of the person. 14 In what appears to be the only study that directly investigated the development and differentiation of social norms, Emmerich, Goldman, and Shore (1971) studied the responses of children and adolescents to normative state— ments that differed in content, sex and generation of source, and sex and generation of the recipient of the norm. Emmerich, et a1., studied the responses of children and adolescents between the ages of 8 and 17 to four norms: give help, argue, seek help, and agree. The children were asked to respond to the norms when both the source and object varied (five sources--self, mother, father, boy friend, and girl friend). The study is rich in interesting results, and some of the more general findings follow. The ranking of the norms in order of importance, for all ages, was: agree with others, help others, seek help from others, and argue with others. These rankings were present in the eight-year-olds and did not change across age. These findings indicate that social norms are reasonably well differentiated by middle-childhood and remain stable through adolescence. A second finding compared the standards that parents and peers had for the self when parents and peers were the objects. The findings were that in middle childhood both parents and peers were seen as having higher standards of conduct for the self when the object 15 of the self's behavior was the parents than when it was peers. During adolescence, this finding remained true for parents; however, it reversed itself for peers. Peers were seen as holding higher standards of conduct for the self for peers than for parents. Emmerich et al. suggest that: . . . the child conceives of a single social world in which obligations are consistently differentiated on the basis of generation, whereas the adolescent conceives of two parallel social worlds, consisting of his relationships with parents on the one hand and with peers on the other, with high (and similar) standards applicable within each context, but with lower standards applicable whenever they intersect. The Emmerich et a1. study is rich in empirical findings which relate to normative differentiation, self-concept develOpment, sex-role development, and parent-peer conflicts. The findings also relate to broader and more general theoretical issues such as the relationship between personality and social structure and theories of socialization. For example, the findings generally support a social learning explanation of socialization rather than a cognitive-develOpmental one (Kohlberg, 1969), since the norms were differentiated relatively early and did not change dramatically during middle childhood and adolescence. Finally, Emmerich's technique represents a methodological advance for examining the "social space" of an individual. Studying a single norm under a variety of systematic conditions (i.e., vary source and vary object) allows for a 16 comprehensive examination of the relation of the self to other significant members in the social environment. Hypotheses The overall aim of the present study is to examine role expectations, using the dyad as the basic unit of investigation. It is assumed that an individual's expectations for himself and for another are a function of the role in which each individual is participating. Two of the most salient dimensions for the division of roles in our society are age and sex. Therefore, it is hypothesized that behavioral expectations will systemati- cally vary as a function of these two variables. A third dimension, self-other, will be considered shortly. The first aim of the present study was to investi- gate the develOpment and differentiation of expectations for the self's behavior when the recipient of the behavior varied among four significant others--mother, father, boy friend, and girl friend. Thus, four dyads were generated in which one member of the dyad was always the self. The specific behaviors under investi- gation were: (a) share; (b) tell the truth; (c) do what is said; (d) tell another what to do; (e) ask for help. The ages of the respondents were 10, 12, 14, 16, and 18 years. The first specific hypothesis of this study is that a strict adherence to social norms will decrease 1'7 between middle childhood and adolescence. It is pre- dicted that children will adhere more strongly to inter- personal norms than will adolescents. Piaget (1932) has pointed out that the social world of the child is a world of absolutes and concrete reality. Social questions are approached from an either-or point of view and there is little appreciation of extenuating circum- stances. Therefore, we should expect a decreasing mean expectation level between childhood and adolescence. The second hypothesis of this study is that the relative ranking of these five norms will remain constant across all the five ages under study. It is hypothesized that norms regulating interpersonal conduct in the five areas sampled in this study are learned early and that their relative importance does not change across these ages. For example, if children have the highest expec- tations for themselves for telling the truth and the next highest expectations for sharing, then it is pre- dicted that this relationship will remain constant across all five ages. For these five norms, socialization is seen to be continuous rather than discontinuous (Benedict, 1938). The third hypothesis concerns the age of the recipient of the norm. It is hypothesized that children will have higher expectations for their own behavior when the object of that behavior is parents rather than l8 peers. Adolescents, however, will have equally high expectations for both parents and peers. It is predicted that the importance of the peer group during adolescence does not diminish the importance of the family. Rather, this division during adolescence is seen as a differen- tiation of the social world along an age dimension. It is postulated that children conceive of one social world, with high standards for parents and lower standards for peers. Adolescents, however, conceive of two social worlds, with equally high standards for both parents and peers. This formulation is in agreement with a number of investigators of adolescent development (e.g., Douvan & Adelson, 1966; Campbell, 1969). The fourth hypothesis is that although both males and females have higher expectations for parents than for peers in childhood, female expectations will be higher for parents at all ages than will male expectations. Additionally, it is predicted that females will evidence higher expectations for parents than for peers for a longer period of time than will males. This hypothesis is based on recent work in the area of parent-peer conflicts. Briefly, it has been found that in early adolescence the peer group is more important for boys than for girls and that girls possess a greater degree of social maturity as evidenced by their orientation towards adult standards (Hartup, 1970). 19 With regard to sex differences, two additional hypotheses are proposed. Discussions of sex differences in socialization suggest that certain behavioral norms apply more to one sex than to the other (Kagan, 1964; Maccoby, 1966; Mussen, 1969). It is expected, therefore, that these five norms will be differentiated on the basis of sex of the respondent. Specifically, it is predicted (Hypothesis 5) that males will endorse the norm "tell another person what to do" more strongly than will females, while females will score higher than males for: sharing, doing what is said, and asking for help. This sex division seems to have face validity as far as our culture is concerned. The sixth hypothesis concerns what Emmerich et a1. (1971) call the "characteristic sex-role orientation." Emmerich found that females at all ages and adolescent males had higher standards of interpersonal conduct for themselves when the object of their behavior was a female than when it was a male. What this means is that most individuals adhere more strongly to norms of accepted social conduct when they are interacting with a female than with a male. The old adage that "females bring out the best in us" seems to be at least partially confirmed by empirical investigation. This same relation- ship was predicted in the present study. 20 The second general aim of this study was to compare the expectations that an individual has for himself when he is the source of the norm with his expectations for others when he is the recipient of the norm. Is there a systematic relationship between the applicability of a norm when the self is either the actor or the recipient? Is the relationship a simple cost- reward one, in which the individual expects to receive what he gives, or, are these expectations mediated along age and sex dimensions? A consideration of the distinction between self and other as a distinction by which individuals dif- ferentially form expectations and evaluate behavior has been ignored in developmental research. Recent reviews of peer relations (Hartup, 1970) and moral development (Hoffman, 1970) do not indicate studies which examine different expectations for behavior or different evalu- ations of behavior based on whether or not the behavior is attributed to oneself or to another. Research in developmental psychology on this self-other distinction has been nonexistent, although the theoretical perspec- tives of Mead (1934) and Piaget (1932) present challeng- ing ideas for empirical investigations. As was previously discussed in this paper, work done by Piaget in the area of moral development has demonstrated that young children conceive of social 21 relationships as hierarchically organized and not based on principles of reciprocity. The child's limited cog- nitive ability prevents him from viewing interactions between adults and himself as relationships based upon agreed-upon social contracts; instead, the child sees these interactions as based upon principles of authority and subordination. Additionally, the child's egocentrism, his inability to take the role of another, influences all his relationships, including those with peers. A second type of influence on interpersonal interaction, besides cognitive ability, is the role that each member of the dyad occupies. One may expect the norm of reciprocity to be in Operation when the dyad is composed of two members who are acting in roles in which both members are equals. However, some roles arenot based on equality, particularly where certain specific norms are concerned. For example, fathers and sons are not expected to be equally obedient to each other. As a general prediction, it is hypothesized that individuals will have different expectations for them- selves and for others. It is predicted that a general reciprocity principle does not characterize normative expectations. Age and sex of self and others should, once again, prove to be salient dimensions for the dif- ferentiation of reciprocity. 22 As Parsons and Bales (1951) point out, within the family, children occupy an inferior power position as compared to their parents. Therefore, it is predicted that within the family, children and adolescents will have higher expectations for their own behavior than for their parent's behavior. For example, children and adolescents should expect to ask for help from their parents more often than they should expect their parents to ask for help from them. Among same age individuals, however, reciprocity should be a strongly adhered-to principle. CHAPTER II METHOD Subjects Subjects in the final sample were 200 children and adolescents selected from four schools in the East Lansing, Michigan school system. Initially 297 students were administered the test questionnaire. Eighty-eight students were eliminated from the study on the basis of their socioeconomic status. Each student was asked to write on his test both his father's and mother's occu- pation and education. These factors were used to select a sample of relatively homogeneous and high socioeconomic standing. All of the fathers had some post graduate education and were employed in areas such as medicine, dentistry, law, and college teaching. Most of the mothers had received a BA, and many were employed full- or part-time. Besides social class, there were two further cri- teria for exclusion from the study. First, four children were excluded either because they did not understand the task or because some test items were skipped. Second, students were given a letter addressed to their parents ’23 24 in which the purpose of the study was explained. Parental permission was requested to allow the student to partici- pate in the study. Two sixth grade and three eighth grade students chose not to take part. A copy of the Iparental permission form is included in Appendix A. Age groups for the final samples are given in Table 1. Twenty males and 20 females from grades 4, 6, 8, 10, and 12 were selected for the study. The age groups were chosen to meet four criteria: (1) inclusion of both pre- and post-pubescent subjects; (2) moderate size N's; (3) about the same average age for both sexes within age groups; and (4) a constant age differential between age groups so that an equal interval age scale could be constructed. Table 1 Subject Characteristics Boys Girls Grade _ _ Age Range X’ N Age Range g' N 4 9.1 - 10.4 9.8 20 9.1 - 10.7 9.8 20 8 13.0 — 15.3 14.0 20 13.4 - 14.4 13.9 20 10 15.5 - 16.4 16.0 20 15.4 - 16.3 15.8 20 12 16.7 - 18.5 17.9 20 16.8 18.3 17.8 20 25 Differentiation of Norms As discussed earlier, the present taxonomy of norm differentiation called for derivation of eight dis- crete normative statements for each of five norms. Each norm was assessed in four dyadic situations (self-mother, self-father, self-boy friend, self—girl friend). Within each dyad, each norm was assessed for the expectations that ego had for himself as both subject and object. For example, the following two questions were asked for the norm "tell the truth" and the dyad, self-father: "How often do you think you should tell the truth to your father? How often do you think your father should tell the truth to you?" Thus, eight questions were required for each norm and the entire test instrument consisted of 40 items. Selection of Norms Since the primary aim of this study was to examine the differentiation of role expectations, it was not possible to construct a comprehensive normative space. Because of the number of questions required for each norm, the present technique precludes the exami- nation of a large number of different norms. The norms chosen for this study were selected on the basis of four criteria. First, norms were selected to sample a wide variety of interpersonal situations. Second, some degree of variability was expected in the endorsement of each 26 norm. Third, each norm was within a normal range of what the community-at-large would regard as acceptable behavior. Fourth, each norm was stated in such a way that its appli- cability was not a function of number of available situ- ations. With these criteria in mind, the following five norms were selected for study: (1) share, (2) tell the truth, (3) do what is said, (4) tell another what to do, (5) ask for help. Test Questionnaire The test questionnaire consisted of 40 statements, each accompanied by five response alternatives--(1) never, (2) sometimes, (3) often, (4) very often, (5) always. The instructions were to circle one answer for each statement. (The problem of an extreme response bias in young children is discussed in Appendix B.) Test items were randomly arranged on each page, eight items per page, and pages were randomly arranged for each test. A copy of the test instrument is included in Appendix C. The questionnaire included the title "What People Do" followed by these instructions: These questions ask about different people in your life, and what they do. Some questions ask about yourself. Others ask about your mother and father. Still others ask about friends of yours. The questions about friends ask about a boy in this school who is your best boy friend, or about a girl in this school who is your best girl friend. Print below the first name of the boy 27 and the girl in this school who are your best friends. Keep in mind this one boy and this girl when answering the questions about friends. First name of your best friend in this school who is a boy: . . First name of yourfibest friend in this school who is a girl: . Consider each question carefully before answering. This is not a test, and there are no right or wrong answers. There is no time limit. Your answers will not be seen by anyone except the interviewer. Please answer all questions, even if you are net sure of your answer. Answer the questions in the order given. Some questions may seem similar to you but each is different and each question appears only once. Do not look back at your earlier answer. Print your first name and date and year of your birth below. Name: Birth: Circle your sex: Male Female Procedure Parental permission forms were distributed to the parents of each subject approximately one week before itesting. Subjects were assessed in classroom groups. All assessments were made during normal school hours. Subjects were scheduled for periods of 50 minutes, although many of the subjects completed the instrument in a shorter period of time. A second test instrument, unrelated to this study, was administered in the remaining time. Subjects were assessed by the principal investi- gator and his assistant. At the beginning of each assessment session, the investigator introduced the 28 study as a research project in human development and made the following points: many students from that school and from other schools in the community were participat- ing; the purpose of the study was to increase our under- standing of young people and it was not an evaluation-- there were no right or wrong answers; responses would be anonymous and individual answers would only be seen by the investigator and would not be available to parents, teachers, or school administrators; participation was voluntary and an individual could choose not to par- ticipate if he so wished. The instrument was then dis- tributed by the investigator who read the instructions aloud as subjects read along silently. Questions were then answered and students began. Each item was read aloud by the investigator for children in grades 4 and 6. Design The design for this study is a 2 x 5 x 2 x 2 x 2 x 5 analysis of variance. The six factors and their levels are: Sex (male, female); Age (9-10, 11-12, 13-14, 15-16, 17-18); Source (self, other); Status (parent, peer); MF, sex of other (male, female); Norm (share, tell the truth, do what is said, tell another what to do, ask for help). All six factors are assumed to be fixed. There are two between factors (Sex and Age) 29 and four within factors (Source, Status, MF, and Norm). The design is similar to a Case I, repeated measures design, described by Winer, 1971. CHAPTER I I I RESULTS A. Overview of Data Analysis The data to be presented in this chapter are organized into two sections. First, role expectations for the self are analyzed by examining only those 20 questions within the test pertaining to self-expectations. These self-expectations are examined in terms of age changes and sex differences. Second, role expectations for self and others are compared in an analysis of the data from the complete questionnaire. These expectations for self and other are also examined in terms of age and sex differences. Within each of the source systems, a lower-order interaction is first presented followed by the relevant higher-order interaction. Three rules concerning the self and self-other data were adopted to facilitate the presentation and interpretation of the results. First, rather conserva- tive guidelines were adopted for setting probably values, since the sample size was large and many statistical tests were carried out. Accordingly, only Erratios that reached a p <:.001 confidence level were accepted 30 31 as significant in the overall analysis of variance. For the post hoc simple effect tests, a critical value of p < .01 was established. For the overall analysis of variance, 15 findings significant between the .001 < p < .05 were disregarded, 3 in the self, and 13 in the self-other analysis. Second, since the self-analysis was primarily concerned with age changes and sex dif- ferences, interactions which did not include an age or sex factor were ignored. Two such interactions were eliminated. Third, the self-other analysis focused on a comparison of role expectations for self and other. Therefore, interactions which did not include a signifi- cant Source effect also were ignored. Nine interactions of this type were disregarded. B. Role Expectations for the Self Table 2 gives the results for the analysis of variance for expectations for the self. As was pre- viously mentioned, these data were examined only for age changes and sex differences. 1. Developmental Changes a. Adherence to social norms.--Tab1e 2 indicates that adherence to social norms was affected by age (F = 5.80, df = 4/190, p ‘<.001). However, Figure l, which depicts mean adherence score as a function of age, reveals that the relationship between age and adherence to 32 Table 2 Analysis of Variance for Self—Expectations Source df Mean Square F Between Subjects A (Sex) 1 2.86 0.87 B (Age) 4 19.03 5.80a AB 4 1.35 0.41 53 w. gps 190 3.28 Within Subjects c (Status) 1 75.90 58.53a AC 1 1.26 0.97 BC 4 13.36 10.30a ABC 4 0.07 0.06 C x 53 w. gps 190 1.30 0 (Norm) 4 437.31 363.35a AD 4 5.60 4.65a BD 16 4.68 3.89a A80 16 0.94 0.78 D x Ss w. gps 760 1.20 E (MF) 1 1.98 2.32 AB 1 18.91 22.17a BE 4 1.61 1.89 ABE 4 1.67 1.96 E x 85 w. gps 190 0.85 CD 4 85.76 133.86a ACD 4 3.00 4.68a BCD 16 3.83 5.98a ABCD 16 0.55 0.85 CD x 85 w. gps 760 0.64 CE 1 0.06 0.07 ACE 1 4.03 4.66 BCE 4 2.97 3.43b ABCE 4 1.62 1.87 CE x Ss w. gps 190 0.87 DE 4 1.36 4.00b ADE 4 4.17 12.28a BDE 16 0.19 0.57 ABDE 16 0.72 2.13 DE x 53 w. gps 760 0.34 CDE 4 1.71 5.38a ACDE 4 0.46 1.44 BCDE 16 0.40 1.25 ABCDE 16 0.52 1.64 CDE x Ss w. gps 760 0.32 ap < .001 bp < .01 33 mucmucoo m>Humauoz .mamm you mGOHumuommxm How mmmcmno mm¢ mod manna mHImH «Alma NHIHH oanm 4 fl 4 fl . II J .cmcflneoo O‘Homnflm 0 Com . mm mm a u 3 X d a D . m on a T: O U S 3 O I m.m a 34 norms was not a simple linear one. A Tukey (a) test (Winer, 1971, p. 201) indicated that the expectations for the 9-10 year-old group were significantly higher than the expectations for the 11-12 and 17-18 year-old groups. All other comparisons were nonsignificant. This finding partially confirms the first hypothesis. It was predicted that adherence to social norms would show a constant decrease between middle childhood and late adolescence. As can be seen in Figure 1, this was not the case, although responses from the 9-10 and 17-18 age groups were different. A clearer understanding of this finding can be obtained when the data for status and norm are considered in relation to age changes. b. Norm ranking.--Tab1e 2 indicates that the Age x Norm interaction was significant (F = 3.89, df = 16/760, E.<:-001)- Figure 2, which presents the mean adherence score for each norm as a function of age, shows that there were many differences in self-expec- tations for each norm at different ages. For the pur- poses of this study, however, the significant finding concerns the relative ranking of the norms across all five ages. It was predicted that the ranking of the norms would remain stable between middle childhood and adolescence. In middle childhood the ranking of the 35 4.5 — Tell the Truth 4.0 c Share 3 o 3.5 '— o m c o '3 3 Ask for o 3.0 P Help 0 m x m a Do What 8 Is Said 2 2.5 - 2.0 ” Tell Another What To Do 1.5:: 0 1 1 1 41 1 9-10 11-12 13-14 15-16 17-18 Age Fig. 2. Differentiation of self-expectations for five normative contents. 36 norms was as follows: (1) tell the truth; (2) share; (3) do what is said; (4) ask for help; (5) tell another what to do. This ordering of the norms appears to have general face validity as far as our society is concerned. Although one reversal occurred in middle adolescence for the norms "do what is said" and "ask for help," the general pattern of the data indicated that by middle child- hood, the relative importance of norms regulating inter- personal conduct is reasonably well established and remains constant throughout adolescence. Although the strength of each norm varies somewhat across ages, the relative ranking of the norms remains generally stable. Thus, this finding supports the second hypothesis. c. Parents and peers.--The third hypothesis was concerned with the relative importance of parents and peers. It was predicted that in middle childhood role expectations for the self would be higher for parents than for peers, while in adolescence expectations for the self would be equal for both parents and peers. Table 2 and Figure 3, which presents the mean adherence score for parents and peers as a function of age, indicates that the Age x Status interaction was significant (F = 10.30, df = 4/190, 2 <:.001), a result which provides preliminary support for the hypothesis. A simple effects analysis revealed that the difference between parent and peer standards is significant at ages 9-10 and 11-12 and 37 .pmcwnsoo mucmucoo 0>Hum8uoz new muconmm MOM mcowumuommxmlmamm mo cowumwpcmummmao .m .mflm mm< manna mHImH wanna NHIHH caum (n _ ~ _ ~; 0 (CON : m.N mumwm \\I// \ // \\\ I, l/ o It 1 o m mucmumm .mummm exoos uornenoadxa ueew 38 nonsignificant at all other ages (see Appendix D for all simple effects tests). Children had higher expectations for themselves when interacting with their parents than with their friends. However, this discrepancy disappeared in early adolescence. The simple effects analysis also indicated that there were significant age changes for both parent and peer objects. A Tukey (a) test revealed the following: When parents were objects, role expectations for the self were significantly higher than all other ages at 9-10 and sig- nificantly less than all other ages at l7-18. When peers were objects, expectations for self were significantly less than all other ages at 11-12. Generally, these findings mean that expectations for self when interacting with parents decreased between middle childhood and adolescence, while expectations for self when interacting with peers remained reasonably stable. Thus, the decrease in the discrepancy between parents standards and peer standards was the result of a lowering of the expectations for the self with parents to the level of expectations for the self with peers. The previous findings focused on age changes in general expectations for the self with parents and peers. It was concluded that parents occupy a higher status than peers in middle childhood and that parents and peers occupy an equal status during adolescence. Additionally, 39 it was found that the expectations for the self with parents decreased somewhat during middle childhood and adolescence, resting essentially on a par in adolescence with expectations for the self with peers. These con- clusions must be regarded as temporary, however, pending an investigation into the effects of specific norms. These normative contents are considered now. It was predicted that the pattern of development for expectations for the self with parents and peers would be different for each norm. As can be seen in Table 2, the Age x Status x Norm interaction was significant (F = 5.98, df = 16/760, 2 < .001), indicating different patterns of normative develOpment. A general overview of the results, illustrated in Figures 4 through 8, reveals that the parent system and the peer system of interaction are significantly differentiated from each other in middle childhood but not in adolescence. Figure 4 illustrates the developmental change that takes place for the norm "share." Nine- and ten-year-old children expected to share more often with their parents than with their friends. At no other age was the dif- ference between parents and peers significant, although the trend of the data indicated increasingly higher expectations during adolescence for the self with peers rather than the self with parents. 40 :.mumcm= Enos onu How muomm paw musmnmm How mcowumuommxmlmamm mo downwaucoumMMHo .v .mwm mmd manna malma vHImH NaIHH calm ‘1 A _ fi 1J—G m.N E \\\\ 91003 uorqeqoedxa ueew 41 The data for "tell the truth," summarized in Figure 5, show a pattern of development which is very similar to "share." Differences between expectations for the self with parents and peers were only significant at ages 9-10, according to an analysis of the simple effects. Once again, the trend of the data indicates higher expectations during adolescence for the self with peers than for the self with parents, but this difference was not significant. Children expect to tell the truth to their parents more often than they expect to tell the truth to their friends. Although the ranking reverses itself during late adolescence, it never quite reached significance. Figure 6 illustrates the developmental pattern for the norm "do what is said." The age trend for this norm is the most unusual of all the norms studied. Expectations for the self with parents are significantly higher than expectations for the self with peers at all ages. Children and adolescents at all ages expect to do what their parents tell them to do more often than they expect to do what their friends tell them to do. How— ever, adolescents eXpect to obey their parents less than younger children do. The data for the norm "tell another what to do" are given in Figure 7. Individuals at all ages have low expectations for this norm. To some extent this norm 42 :.£usuu on» damp: Enos map How mummm pew mucoumm MOM mcowumuowmxmlmamm mo GOHDMHucmHmmmwo .m .mHm mod manta mauma wanna NHnHH oaum fl. . 4) . _(, o uum.~ W - o.m m u 3 X d a O m .. m.m 1. To 0 U S D O m .t.. o.v fimH mu mm L m.v 43 =.©fimm ma was; ow: Enos man How mnwmm pew mucmumm Mow mcoflumuommxmnwamw mo cowumwucmnmmmao .m .mfim 00¢ manna mHImH valma NHIHH OHIm _ _ _ — _ ,c "no.N \\)I:Il \ li/l 110/ \\ III. ‘\|I\I [’4’ \\ .l. MoN muomm l/l \ W I(\ e u 3 X a I o.m o q s 3. T. O u m. ..m.m o 1 a mucmumm 1 o.v 44 =.oo ou umsz umsuosw Ham»: Enos msu MOM mummm psm mesmumm How msoaumuommxmlwamm mo soaumwusmeMMHo .h .mHm mod manna mHImH valma NHIHH calm 4 _ ~ 0 H o ImJ musmumm \ I . \\\ /// om \ \II “““ I’ll..." \\\ // mummm Il.\ // / / /. I m.~ I o.m L m.m 81003 uorqeqoedxa ueew 45 represents the complement of the previous norm--obedience and giving orders. As such, this norm shows a develop- mental pattern that is similar, although in the Opposite direction, to the previous norm. Expectations for the self with peers are higher than expectations with parents at every age, and this relationship did not reverse itself. However, this difference between parent and peer expectations was significant only at ages 9-10. Children expected to tell their friends what to do more often than they expected to tell their parents what to do. Adolescents had somewhat similar eXpectations for both parents and peers, although expectations for peers were slightly higher. Finally, Figure 8 illustrates the data for the norm "ask for help." Children at ages 9-10 and 11-12 expected to ask their parents for help significantly more often than they expected to ask their friends for help. At no time during adolescence is there a sig- nificant difference between expectations for parents and peers, although parents continue to occupy a slightly higher position than peers. d. Summary of developmental changes for expec- tations for the self.--The data for age changes for expectations for the self can be summarized as follows: Norms regulating interpersonal conduct are internalized early, by age nine, and the relative importance of these 46 =.maos How xmmg Euos on» How mummm osm musmumm How msOAumuommxmnmaom mo soquHusmsmMMHo .m .mam mad manna mauma sauna mHuHH cans _ _ _ _ d o I I o.~ W J m.~ m U 3 X d 8 0 3 l o.m W T... O U S O musmumm m 1 m.m a . 0.4 47 norms does not change with age. Children make a clear distinction between parent standards and peer standards, with parents occupying a higher status than peers. Adolescents do not make this distinction and both parents and peers occupy an equal status. There is some evidence that this equalization of standards is the result of a lowering of the very high standards when dealing with parents which exist during childhood. Important differences in individual norms were noted. At age 9-10, parents occupied a higher status than peers for all five interpersonal norms. At age 11-12, this same relationship was true only for the norms "do what is said" and "ask for help." Between the ages of 13 and 18, adolescents expected to do what they were told to a significantly greater extent when interacting with parents than with peers, while the expectations for all other norms were equal with parents and peers. 2. Sex Differences As can be seen in Table 2, there were no signifi- cant Sex x Age interactions. This lack of deve10pmental sex differences disconfirms the fourth hypothesis, which predicted that all children would show higher expectations for parents than for peers, adolesCents would show equal expectations for peers and parents, and this changeover would be earlier for males than for females. Thus, a Sex x Age x Status interaction was predicted but not found. 48 Although the data contain many sex differences, developmental sex differences were not present. In the. area of role expectations, boys and girls apparently develop along similar paths. Table 2 does indicate, however, that there are important sex differences in role expectations, irrespective of age changes. These dif- ferences are considered now. a. Norm applicability.--The fifth hypothesis pre- dicted sex differences in normative expectations. Specifi- cally, it was predicted that males would have higher expectations for "tell another what to do," females would have high expectations for "share," "do what is said," and "ask for help," and males and females would have equal expectations for "tell the truth." Table 2 shows that sex of self influences norma- tive expectations (F = 4.65, df = 4/760, 3 < .001). How- ever, a simple effects analysis of the data in Table 3 indicated that no comparison was significant beyond 2 < .05. Although the overall interaction was highly significant, the individual comparisons were not. Since this interaction was predicted, however, it was thought worthwhile to examine the results of the simple effects analysis. All normative expectations were consistent with the hypothesis except "do what is said," which was . 49 not sex specific. It should be emphasized again, how- ever, that these findings are significant at .01 < p < .05. Table 3 Mean Self—Expectation Scores for Males and Females for Each Norm Sex of Norm Self Do What Tell What Ask for Share Tell Truth Is Said To Do Help Males 3.36a 4.00a 3.08a 2.10a 2.85a Females 3.59b 4.03a 3.07a 1.92b 3.05b For each norm, M-F comparisons having different subscripts are significantly different from each other (p $1.05) according to an analysis of variance for simple effects (Winer, 1971). These results seem to have general face validity as far as sex-role standards in our society are concerned. Further elaboration of these findings will be reserved until the important interactions between sex and norm, and sex of other and status of other are considered. b. Male and female others.--The sixth hypothesis concerns the "characteristic sex-role orientation," that individuals have higher role expectations when interacting with a female than when interacting with a male. Table 2 reveals that the sex of self x sex of other interaction was highly significant (F = 22.17, df = 1/190, 2 < .001). 50 The simple effects analysis indicated that the predicted effect was significant only for females, however. Females had higher expectations for their own behavior when interacting with other females than when interact- ing with males. Males did not show significantly dif- ferent eXpectations for other males or females, although the pattern of the data was in the same direction as the female data; males showed higher expectations for other males than for females (see Table 4). Thus, the hypothesis of a characteristic sex-role orientation was only partially confirmed. A more apprOpriate description of the data, however, is to say that indi- viduals had higher expectations for themselves when interacting with a member of their own sex than when interacting with a member of the opposite sex. The data for Sex x Norm x MF help to clarify this relation- ship. Examination of the relationship between sex of self and sex of other for each particular norm reveals that the previous conclusion is generally correct. Table 2 indicates that the Sex x Norm x MF interaction was highly significant (F = 12.28, df = 4/760, 3 < .001). The mean expectation scores for males and females for each norm when the sex of the other is either male or female are given in Table 5. As can be seen by examining Table 5, the data support the conclusion that males have 51 Table 4 Mean Self-Expectation Scores for Males and Females for Male and Female Others Sex of Other Sex of Self Male Female Males 3.12a 3.03a Females 3.04a 3.22b Within each row, means with different subscripts are significantly different from each other (p < .01) according to an analysis of variance for simple effects (Winer, 1971). Table 5 Mean Male and Female Self-Expectation Scores for Male and Female Others for Each Norm Norm Sex of Other Share Tell Do What Truth Is Said Tell Ask What for To Do Help M 3.49a 4.04a 3.12a 2.06a 2.92 Male F 3.23b 3.97a 3.04a 2.15a 2.78 Sex of Self M 3.39a 3.97a 3.12a 1.85a 2.88 Female F 3‘79b 4.10a 3.02a 2.00a 3.22 Within each sex, M-F comparisons with different subscripts are significantly different from each other (p < .01) for that norm according to an analysis of variance for simple effects (Winer, 1971). 52 higher expectations for their own behavior when inter- acting with other males than when interacting with females. This difference was significant, however, only for the norm "share." Males expected to share more with other males than with females. The female data are similar to the male findings. Females have higher expectations for their own behavior. when interacting with girls than when interacting with boys. This difference was significant only for the norms "share" and "ask for help." Females expected to share with and ask for help from females more often than from males. In view of the general pattern of the data for the Sex x Norm x MF interaction, the following overall conclusion seems warranted. There is some evidence for the notion that males and females had higher expectations for themselves when interacting with others of the same sex than with others of the opposite sex. However, these differences generally were not significant. This lack of significance (7 out of 10 comparisons were non- significant) indicates that individuals have generally equal expectations for themselves when interacting with either males or females. c. Parents and peers.--Table 2 indicates that the Sex x Status x Norm interaction was significant (F = 4.68, df = 4/760, 2 < .001). Males and females had 53 different patterns of role expectations for parents and peers. Table 6 shows that males: (a) expected to do what their parents told them more often than they expected to do what their friends told them; (b) expected to tell their friends what to do more often than they expected to tell their parents what to do; and (c) expected to ask their parents for help more often than they expected to ask their peers for help. Table 6 also shows that females: (a) expected to share more often with their parents than with their peers; (b) expected to do what their parents told them to do more often than they expected to do what their peers told them to do; and (c) expected to ask their parents for help more often than they expected to ask their peers for help. Although these results were not predicted, their outcome was not unexpected. Boys and girls apparently expect to interact in different ways with their parents and friends. d. Summary of sex differences for expectations for the self.--The data for sex differences in the development of role expectations can be summarized as follows: Females expected to share and to ask for help more often than males, while males expected to tell others what to do more often than females. When sex differences were examined in terms of the sex of the other, a slightly different picture emerges. The 54 mammamsm sm ou mswpuooom Enos umsu How A .Aanma .Hmsfi3v meowmmo mHmEHm How mosMHum> mo Ho. V dc Hmsuo comm Eoum uswuwmmwo haysMOHMHs Imam mum mumfluomnsm usmummmap spas msOmwummEoo smmmlusmumm .xmm comm sfisuflz nag.~ 84o.~ amm.~ smo.¢ has.m 000s M m M G mflwemh ma.m Hm.a ms.m vo.v ms4.m usmumm mamm mo xmm nmm.~ nvm.~ naq.m 8mm.m as~.m 000s M m M M m GHMS mo.m sm.a ms.m mo.4 mv.m usmsms mamm oo 09 camm mH . susua Hams mumsm sou sms sass Hams 08s: on msumum Euoz EH02 somm Mow mummm cs0 musoumm uom mouoom soaumuommeImamm mHmEmm m OHQMB ps0 was: sum: 55 patterns of role expectations for males and females were generally similar. With the exception of the situation "share," in which females expected to share more often than did males and both sexes expected to share more often with a member of their own sex, male and female expectations were very similar. Where they do differ, higher standards of personal conduct were usually expected with members of the same sex, and this was particularly true for females. Males and females showed a somewhat different pattern of role expectations for parents and peers. It is not readily apparent, however, why this difference occurred. 3. Summary of Expectations for the Self In combination with the age findings, the follow- ing generalizations emerge about role expectations for the self. Children had higher standards of interpersonal conduct for themselves when interacting with adults than with other children and standards for themselves that were equal for males and females. Adolescents had standards for themselves when interacting with adults and other adolescents that generally were equal for these two groups, although the specific relationship for these two groups depended upon the particular norm in question. Adolescents also had equal standards for males and females. Thus, for children, age of other 56 was an important determinant of role expectation, while sex of other was generally not. For adolescents neither age nor sex of other generally influenced role expec- tations. In those situations where sex of other was an important determinant of self-expectations--e.g., "share"--differences in self-expectations were to the advantage of the same-sex other. Both children and adolescents had somewhat higher standards for themselves when interacting with someone of the same sex than with someone of the Opposite sex and this was particularly true of females interacting with other females. C. Comparison of Role Expectations for the Self and Other The previous section examined role expectations for the self in terms of age changes and sex differences. This section looks at the relationship between expec- tations for the self and expectations for others. The analysis for this data is based on the responses of subjects to all 40 questions. The complete analysis of variance is given in Table 7. It should be recalled that differences in self-expectations and other expec— tations were predicted. The precise pattern of these differences is explored in the sections that follow. 57 Table 7 Analysis of Variance for Expectations for Self and Other Source df Mean Square F Between Subjects A (Sex) 1 2.08 0.35 B (Age) 4 22.77 3.94a A8 4 0.91 0.16 83 w. gps 190 5.79 Within Subjects 0 (Source) 1 11.33 20.38b A0 1 0.90 1.63 80 4 5.19 9.34b ABC 4 1.12 2.02 C x Ss w. gps 190 0.56 0 (Status) 1 109.28 53.45b A0 1 2.49 1.22 80 4 19.31 9.45b A80 4 0.66 0.32 D x 85 w. gps 190 2.04 8 (Norm) 4 796.31 430.42b A8 4 7.57 4.09a 88 16 3.28 1.77c A88 16 1.78 0.96 E x 53 w. gps 760 1.85 F (MP) 1 9.73 8.53a AF 1 23.65 20.73a BF 4 0.77 0.67 ABF 4 1.64 1.44 F x 85 w. gps 190 1.14 00 1 3.49 7.50a ACD 1 0.00 0.00 800 4 0.75 1.60 ABCD 4 0.99 2.12 CD x 83 w. gps 190 0.47 08 - 4 103.81 149.06b ACE 4 0.99 1.42b 808 16 4.45 6.38 A808 16 0.71 1.02 CE x 55 w. gps 760 0.70 58 Table 7 Continued Source df Mean Square F 08 1 1.28 4.63 A08 1 1.65 6.00C 808 4 1.03 3.73a A808 4 0.80 2.91c CF x 85 w. gps 190 0.28 08 4 28.25 40.82b ADE 4 3.16 4.56b 808 16 1.57 2.27a ABDE 16 0.93 1.35 DE x 85 w. gps 760 0.69 08 1 0.60 0.51 A08 1 3.49 3.00 808 4 5.05 4.358 A808 4 3.40 2.93c DF x 53 w. gps 190 1.16 88 4 2.65 7.64: AEF 4 4.60 13.28 888 16 0.26 0.74 A888 16 0.95 2.75 EF x SS w. gps 760 0.35 008 ' 4 80.71 132.63b A008 4 1.16 1.91 8008 16 4.16 6.84b ABCDE 16 0.58 0.96 CDE x 85 w. gps 760 0.61 008 1 0.19 0.53 A008 1 0.95 2.66 8008 4 0.26 0.74 A8008 4 0.02 0.05 CDF x 85 w. gps 190 0.36 088 4 2.38 6.67b A088 4 0.87 2.45c 8088 16 0.22 0.61 A8088 16 0.23 0.65 CEF x 53 w. gps 760 0.36 59 Table 7 Continued Source df Mean Square F DEF 4 0.39 1.21 ADEF 4 0.98 3.02C BDEF 16 0.39 1.20 ABDEF 16 0.35 1.08 DEF x Ss w. gps 760 0.33 0088 4 1.76 6.10b ACDEF 4 0.02 0.07 BCDEF 16 0.29 0.99 ABCDEF 16 0.40 1.37 CDEF x SS w. gps 760 0.29 ap < .01 bp < .001 0p < .05 60 1. Developmental Changes a. Self-other expectations.--As can be seen in Table 7, there was a significant Age x Source interaction (F = 9.34, df = 4/190, 2 <:.001). The pattern of this interaction is illustrated in Figure 9. An analysis of variance for simple effects indicated that self- expectations are higher than other expectations at age 9-10 and 11-12, and that self and other expectations are equal at all other ages. In interpersonal situations, preadolescents apparently had higher expectations for themselves than for others. Adolescents, however, have similar expectations for themselves and others. Although there is some evidence for a reversal of the relative ranking of self and other expectations in adolescence, with expectations for others being higher than expec- tations for self, this reversal did not reach significance. Further interpretations of this relationship are post- poned until the effects of specific normative content are examined below. b. Parents and peers.--Table 7 indicates that the Age x Source x Norm interaction was significant (F = 6.38, df = 16/760, p} <.001). However, since the Age x Source x Norm x Status interaction also was sig- nificant (F = 6.84, df = 16/760, 2 < .001), only this latter interaction is discussed in detail. Role 61 . UQGHQEOU mflfimUCOU m>wumEuoz .muosuo ps8 mamm Mom msOHumuoomxm sH momsmso mod .m .mHm 00¢ manna mHImH valma NHIHH oatm 1 (m (11 1) a 0 Mn o.m exoos uoraeqoedxg ueew 62 expectations for the self and others were related to the age of the individual, the content of the norm, and the status of the other. Figures 10 through 19 illustrate the self-other distinction for each norm and for parents and peers separately. The data for the norm "share" are given in Figures 10 and 11. An analysis of variance for simple effects indicated that children at ages 9-10 expected to share significantly more often with their parents than they did with their friends. At no other age was the self—other distinction significant, although subjects at all ages had consistently higher expectations for themselves than for their parents. The data for sharing with parents are given in Figure 10. Figure 11 illustrates the data for sharing with peers.. An analysis of variance for simple effects indi- cated that the self-other distinction was significant at ages 9-10 and 11-12. Preadolescents expected to share more often with their friends than they expected their friends to share with them. Once again, self- standards were higher than other standards at all ages, although the difference was not significant. A comparison of the self-other distinction for parents and peers indicates no fundamental difference in how this dimension applied to these two groups. Basically, preadolescents distinguished between 63 =.mnwsm= Enos was now mnmmm osm musmnmm now msonumuommxo mo sOnumnusonmmmwo .oa .mnm mod malha malma vHImH NHIHH caum J, a . . ‘1 o H m.~ w 1 o.m 8 ‘III‘||I,,/ u musmnmw 11. ||||||| .11 11 I I l/ m \IIIII-III“ 1 ll. 10 11 e o 3 8 “How 1 m.m 1 T.- o u S o o a 1 o.v 64 now mamm osm mnmmm now msoHumnoomxm mo soHUMHusmnmmmno =.mnmsm= Enos msu .nn .oas om< ma1ha malma «Alma NHIHH OHIm . 1 1. a 1? o nu o.~ W 1 m.~ m u a x d e o 1 . m o m 3 T: o u S o o 1 In mom a waom L o.e 65 self-expectations and other expectations for the norm "share," while adolescents did not make this distinction; and this difference was true for both parents and peers. Thus, the pattern of development for the self-other distinction for parents was similar to the pattern for peers. Figure 12 illustrates the data for the norm "tell the truth" for the parent-self dyad. The difference between expectations for self and expectations for parents was significant at ages 13-14, 15—16, and 17-18. Preadolescents expected to tell the truth to their parents as often as they expected their parents to tell the truth to them. Adolescents expected their parents to be more truthful to them than they, themselves, expected to be truthful to their parents. Figure 13 is the graph for the norm "tell the truth" for the peer-self dyad. At no age was the self- other distinction significant. Expectations for telling the truth to peers were similar to expectations for peers telling the truth to oneself; and this similarity was present at all ages. For the norm "tell the truth" the following con- clusion appears appropriate. Children did not make a distinction between expectations for self and expec- tations for others, while adolescents made this dis- tinction for parents but not for peers. 66 =.susnu msu Ham»: Enos ms» n0m mawm osm musmnmm now msowumuowmxo mo sonDMfluswnommwo .NH .mnm mum malna mHImH valma , «HIHa onlm fl _ q q — O “mts. L o.m W D. u a x .a e o 1 m.m m o u S o o I e 1 o.v musmnmm L m.v 67 =.susnu msu Haous Enos may n0w mamm ps8 mnmmm new ms0flumuoomxm mo sonumwuswnommno .mH .mwm mHIhH wHImH valma NHIHH oatm ~ _ . ‘1 -1 41('_1 o W 1 o.m a p. u 3 x d e a . 1 1 m m 9 q T.- o u S o o . 1 1 o v a 68 The data for the norm "do what is said" are illustrated in Figures 14 and 15. An analysis of variance for simple effects indicated that subjects at all ages expected to do what their parents told them to do significantly more often than they expected their parents to do what they told them. Figure 14 is a graph of this relationship. The data in Figure 15, however, reveal that there were no significant dif- ferences at any ages for expectations for self and expectations for their friends. All subjects expected to do what their friends told them to do as often as they expected their friends to do what they told them. The data for the norm "tell another what to do" exactly parallel the findings for the previous norm. The data in Figure 16 indicate that there was a signifi- cant difference between self—expectations and other expectations at all ages. Children and adolescents reported that they expected their parents to tell them what to do more often than they expected to tell their parents what to do. When the data for the peer-self dyad were examined, however, no significant difference appeared between self and other expectations at any age. Students did not expect to tell their friends what to do any more often than they expected their friends to tell them what to do. Figure 17 illustrates these data. 69 zomuflflm mfl #MSQ» 0U: ECG $5.1... n0m mamm osm musmnmm now msoHumuo0mx0 mo sonumwus0n0mmwo .vH .0Hm 00¢ malwa malma «Alma NHIHH calm ‘ W _ +1 o n o.~ \\- ‘IIIIII‘ “““““““ J\\\ mus0nmm 1 m.N W e u 3 x d a o 1 o.m m 3 To 0 u S o o L m.m a mamm 1 oov 70 r.owmm on puss 00s Enos 0:0 now ma0m osm mn00m now msonumuommx0 mo soquHus0n0mmno .ma .0Hm 00¢ malha malma vHImH NHIHH OHIm W q a _ a O m.H W e .1 o.~ w 3 x d e o 1 P. 1 m.~ "H o u S o o 1 e 1 o.m a. m.m 71 =.oo on 0003 nmsuosm Ha0us Enos msu now wamm ps0 musmnmm now ms0nuwuo0mx0 mo sOnumwus0n0mmno .ma .080 00¢ maIna mHImH wanna NHIHH OHIm H mé 1 o.N W P mnmm u 3 x d I e // m.~ w x I u / u / . o //, 11. u mus0nmm / \\\ II S /\\\l\\ ’l” «Ow ./ l o m a / / / / / I / /L m.m 72 ..oo on 08:3 nosuoso nnmus Enos 0:» now ma0m psm mn00m now msOnumuo0mx0 mo sonumnus0n0mmno .na .0Hm 00¢ mn1sn on1mn an1mn anunn on1o 011 m m m 01 o n” m.n W 1 o.~ m u 3 X .0 8 0 n 1 m.~ m. TL. 0 u S O O 1 1. Com 8 73 The developmental pattern of the self-other distinction for the norm "ask for help" is very similar for both parents and peers. An analysis of variance for simple effects indicated that for the parent-self dyad there were no significant differences at any ages between self-expectations and parent expectations. All subjects expected to ask their parents for help as often as they expected their parents to ask them for help. Figure 18 illustrates this relationship. The data for the peer-self dyad were very similar to those for the parent-self dyad, with one exception. At age 17-18 eXpectations for the self were significantly less than expectations for another. Late adolescents expected their friends to ask them for help more often than they expected to ask their friends for help. As can be seen in Figure 19, the pattern of the relationship was consistent at all ages, although it was significant only in adolescence. c. Summary of developmental changes for the self-other comparison.--The findings for age changes in the self-other distinction do not lend themselves to any straight-forward generalizations. The distinction is present in one form or another at all ages for all norms and for both parents and peers. A few tentative con- clusions do seem in order, however. First, the 74 s.mH0s now xmms Enos man now mamm osm mus0nmm now msowuwuo0mx0 mo sOflumnus0n0mmwo .ma .0nm 00¢ mHIBH mHImH vHIMH NHIHH OHIm " + 11 Jr .1; o N o.~ w 1 as m. u 3 x d 11),, .1 . s0n \11 3 1 mo. on 1 x - o.m m. To 0 U S 3 m “Ham 1 mom a e o... 75 =.ma0s now xmms Enos 0:0 no“ ma0m osm mn00m now msoHuwuo0mx0 mo sowumnus0n0mmwo .ma .0Hm 00¢ malna malma wanna NHIHH caum v x v 0 o o ”u o.~ W e e u . m l m N d e o q 8 1- To . o . u 1 o m S o o 1 e 4 m.m a 3. 76 self-other distinction occurred 34% of the time, that is in 17 out of 50 possibilities (5 ages x 5 norms x 2 statuses). It appears, therefore, that although the distinction was a part of interpersonal expectations, it generally was not a salient one. The validity of this generalization, of course, depends upon the par- ticular norm. Second, use of the distinction does not seem to be related to the age of the subject. Pre- adolescents made the distinction for three norms, while adolescents made the distinction for four norms. Third, the distinction more often was made in the parent-self dyad (14 out of 25) than in the peer-self dyad (3 out of 25). Thus, although the self-other distinction was not made in the majority of cases, if it was made at all, it was more likely to be made in the parent-self dyad than in the peer-self dyad. That is to say, age of other was an important determinant of the self-other distinction. 2. Sex Differences Surprisingly, there were no sex of subject dif- ferences for any self-other comparison. Apparently the expectations that males had for self and others were not different from the expectations that females had. However, sex of other (MF) did prove to be a significant factor for all subjects differentiating eXpectations for self and other. This factor is examined below. 77 a. The importance of status, norm, and sex of gthg£.--As Table 7 indicates, the Source x Norm x MF interaction was significant (F = 6.67, df = 4/760, E.<:'001)- However, since the Source x Status x Norm x MF interaction also was significant (F = 6.10, df = 4/760, 2 < .001), only this latter interaction is dis— cussed in detail. The data for self-father expectations are given in Table 8. For all five norms there was a significant difference between expectations for self and expectations for father. Subjects had higher expectations for them- selves than for their fathers in the following areas: share, do what is said, and ask for help. Subjects also expected their fathers: (a) to tell the truth to them more often than they expected to tell the truth to their fathers and (b) to tell them what to do more often than they expected to tell their fathers what to do. The data for self—mother expectations are given in Table 9. The self-mother comparison for the norm "ask for help" was not significant, but the self-mother com- parisons for all other norms were. Subjects expected to share and to do what was said more often than they expected their mothers to perform these behaviors. Subjects also expected their mothers: (a) to tell the truth to them more often than they expected to tell 78 Table 8 Mean Expectation Scores for Self and Father for Each Norm Norm Status Do What Tell What Ask for Share Tell Truth Is Said To Do Help Self 3.36a 4.04a 3.86a 1.76a 3.08a For each norm, Self-Father comparisons having different subscripts are significantly different from each other (p < .01) according to an analysis of variance for simple effects (Winer, 1971). Table 9 Mean Expectation Scores for Self and Mother for Each Norm Norm Status Do What Tell What Ask for Share Tell Truth Is Said To Do Help Self 3.56a 4.07a 3.65a 1.92a 3.15a For each norm, Self-Mother comparisons having different subscripts are significantly different from each other (p < .01) according to an analysis of variance for simple effects (Winer, 1971). 79 the truth to their mothers and (b) to be told what to do more often than they expected to tell their mothers what to do. Table 10 shows the mean self-other scores for the self—girl friend dyad. There were no significant differences for the norms "tell the truth" and "tell another what to do." However, subjects expected to share significantly more often with their girl friends than they expected their girl friends to share with them. They also expected to do what their girl friends told them to do more often than they expected their girl friends to do what they told them. Finally, subjects expected their girl friends to ask them for help more often than they expected to ask their girl friends for help. The data for the self-boy friend dyad are given in Table 11. The analysis of variance for simple effects indicated that only one comparison was significant. Subjects expected to share with their boy friend more often than they expected their boy friend to share with them. b. Summary of sex differences for the self-other comparison.--In summary, it can be seen that the importance of the self-other distinction, as indicated by the number of norms in which it was significant, is as follows: (1) self-father, significant for 5 norms; (2) self-mother, 80 Table 10 Mean Expectation Scores for Self and Boy Friend for Each Norm Norm Status Share Tell Do What Tell What Ask for Truth Is Said To Do Help Self 3.52a 3.97a 2.38a 2.16a 2.72a Boy friend 3.17b 4.07a 2.24a 2.16a 2.87a For each norm, Self-Boy friend comparisons having different subscripts are significantly different from each other (p < .01) according to an analysis of variance for simple effects (Winer, 1971). Table 11 Mean Expectation Scores for Self and Girl Friend for Each Norm Norm Status Share Tell Do What Tell What Ask for Truth Is Said To Do Help Self 3.46a 4.01a 2.41a 2.22a 2.84a Girl friend 3.26b 4.10a 2.24 2.08 3.17b b a For each norm, Self-Girl friend comparisons having different subscripts are significantly different from each other (p < .01) according to an analysis of variance for simple effects (Winer, 1971). 81 significant for four norms; (3) self-girl friend, sig- nificant for three norms; (4) self-boy friend, significant for one norm. Additionally, subjects in all four dyads expected to share with others more often than they exPected others to share with themselves; and in all dyads except self-boy friend, subjects expected to do what others told them more often than they expected others to do what they told them. Generally, expectations for self and others were in the same direction for all dyad comparisons for any individual norm. The striking difference in these data concerns the comparison of the self-boy friend dyad with all other dyads. Individuals apparently had very similar expectations for self and boy friend but made a strong distinction between self and other when the other was a parent or girl friend. D. Summary and General Conclusions The implications of the data from this study can be summarized as follows: Norms governing interpersonal behavior are reasonably well established in middle- childhood and generally do not fluctuate in relative importance throughout adolescence. The ranking of norms in adolescence is as follows: (1) tell the truth, (2) share, (3) ask for help, (4) do what is said, (5) tell another what to do. 82 Further, children appear to make a strong dis- tinction between parents and peers. Children have higher standards of interpersonal conduct for themselves when they are interacting with their parents than when they are interacting with their peers. Although sex of other is not as salient a dimension as status of other, it does seem to influence interpersonal expectations for oneself in certain specific situations: Boys expect to share more often with other boys than with girls; and, in general, boys have higher expectations for their own behavior with boys than with girls. Girls expect to share with and ask for help from other girls more than boys and also have generally higher expectations for same-sex others than for opposite-sex others. Children also distinguish between eXpectations for self and expec- tations for another for the norm "share," "do what is said," and "ask for help.” The distinction is pri- marily a parent-self division and is most clear in the norms "do what is said" and "tell another what to do." Adolescents show a somewhat similar structuring of their social world, although there are important differences. Adolescents generally do not differentiate between parent standards and peer standards. Expec- tations for the self are similar for those two groups. This does not seem true in the case of the norm "do what is said." Adolescents expect to do what their 83 parents tell them significantly more often than they expect to do what their friends tell them. Sex dif— ferences, where they appear, interact with the sex of the other and are similar to the findings for children. Adolescents have somewhat higher expectations for their own behavior when they expect to interact with someone of the same sex. The distinction between self- expectations and other expectations is most evident in the parent-self dyad. Adolescents expect their parents to tell them the truth more often than they expect to tell the truth to their parents. Adolescents also expect to be told what to do and to do what they are told more often than they expect their parents to perform these behaviors. Finally, adolescents apply the self-other distinction most often according to the following ranking: self-father, self-mother, self- girl friend; self-boy friend. CHAPTER IV DISCUSSION The primary purpose of this study was: (a) to prOpose a model of socialization that is not based on personality development and/or an assumption of cross- situational behavioral consistency and (b) to examine empirically the usefulness of that model. The "trait" approach to human behavior has been shown to be inadequate since it does not take into account the important contri- bution of the social environment. The model that is presented is based on role theory (Sarbin & Allen, 1968) and reference figure analysis (Brim, 1965). Briefly, the model assumes that while behavior is not independent of the social situation, its expression is not unique to each and every social circumstance. The social world is composed of an extremely large number of individuals and settings and it is believed that this world is organized by individuals into categories or dimensions of social space. Social scientists have devoted little time to the systematic investigation of these dimensions. Admittedly, some work has been done 84 85 on this issue, for example, studies which examined possible behavioral differences in an experiment as a function of the sex, age, and race of the experimenter; but the problem has never been presented in the way it is presented in this study, that is, in terms of role expectations. Although a large number of social dimensions may be important, e.g., religion, race, nationality, etc., three were chosen for this study because they seemed to be fundamental and rooted in physical reality. Thus, role expectations for five behaviors were examined as a function of: (1) the sex of other, (2) the status of other, and (3) whether the self was the subject or object of the behavior. The strategy of classifying the social space by sex, status, and source proved to be highly successful for distinguishing between alternative role expectations. Many findings held across a number of normative contents, indicating that these dimensions may be basic social classification schemes. Further investigation with other normative contents could establish the veracity of that possibility. Additional investigation into other classification schemes would also prove useful. As an end product, this type of research would provide a taxonomy of relevant social dimensions and a measure of the robustness of each dimension. 86 The findings support the view that internalized normative contents are aspects of the personality system that arise from and are structurally parallel to the social system, although personality is, itself, indepen- dent of the social context in which it originally arises. The idea that personality and social structure are independent but parallel in organization is not new. As many have observed and as Wrong (1961) has stated: The normative structure of society is more than an environmental obstacle which the actor must take into account in pursuit of his goals in the same way as he takes into account physical laws; it becomes internal, psychological, and self-imposed, as well. Parsons developed this View that social norms are constitutive rather than merely regula- tive of human nature before he was influenced by psychoanalytic theory, but Freud's theory of the super-ego has become the source and model for the conception. . . . It is also valuable to place the above findings into a developmental context. Normative content appears to be internalized early and does not change as a function of increased cognitive and interpersonal skills. This pattern of development is consistent with a continuity model of socialization in which it is postulated that normative content is acquired early through the well-known processes of learning and remains relatively stable throughout development. Despite this surface continuity, however, a certain amount of under- lying structural change may be taking place. The parent-peer dimension appears more important for 87 children than for adolescents. For example, both children and adolescents had equally high expectations for the norm "tell the truth." However, children had a higher standard for the norm for parents than for peers, while adolescents had equal parent and peer standards. It is possible, therefore, that this dif- ferential use of the parent-peer dimension indicates an underlying structural reorganization of the adolescent's social world, at least as far as the parent-peer division is concerned. These findings lend some support to a discontinuous or cognitive interpre- tation of socialization in which it is postulated that dramatic changes occur in social orientation, particu- larly at the onset of adolescence, as a result of cog- nitive, structural changes (Erikson, 1950; Kohlberg, 1969). As was hypothesized, a general developmental trend occurs in the organization of the child's social world. The child's social world is organized hierar- chially, with parents occupying a higher status than peers. However, as the child develops, as his cognitive and interpersonal skills increase, his world becomes organized around egalatarian principles in which both parents and peers occupy equal status. This concep- tualization is in agreement with work done in the area of moral judgments. Children initially view morality 88 as based on an authoritarian structure in which ethical principles are arrived at by resorting to the pre- scriptions of authority, e.g., parents. As children develop, however, a morality appears which is based on egalatarian principles and on an agreed-upon social contract (Piaget, 1932; Kohlberg, 1971). The present findings extend these results in moral development and indicate a general orientation away from a hierarchial social organization and towards a social world in which parents and peers occupy equal status. Furthermore, this change is the result of a lowering of parent expectations. This finding is consistent with research done by Bowerman and Kinch (1959) who found that between the fourth and tenth grades a general decrease in "family orientation" occurred. Expectations for the self's behavior appear higher for parents than for peers in the middle child- hood and equal for both parents and peers in adolescence. This change is due not to a change in expectations for peers, which remain generally constant across ages, but to a decline in the self's expectations for parents. For others, higher expectations occur for parents than for peers in middle childhood and equal expectations for both parents and peers in adolescence. Once again, this change is due to a lowering of expectations for parents. This finding is consistent with a view of the 89 development of social relations in which the child's world is seen as organized vertically with parents occupying a higher status than peers, while the adolescent's world is organized horizontally, with parents and peers occupying equal status, each within their own interpersonal realm. Placed in this context, it is not hard to see why many parents may view their adolescent sons and daughters as unruly and rebellious. Having, at one time, occupied a position of clear high status, parents, during their child's adolescence, must now accustom themselves to occupying a status equal to their adolescent's friends. These data are also in agreement with much of the contemporary thinking about the adolescent period and the relative importance of parents and peers to the adolescent (Conger, 1973; Douvan & Adelson, 1966; Westley & Elkin, 1956). Briefly, these authors suggest that adolescence is probably not the stormy period of cross-pressures that tradition suggests. For the adolescent, the importance of peers does not in all areas surpass the importance of parents. Rather, both these groups are viewed as equally important but in different social areas. This conclusion is also justified by the present study. This general finding of a shift in expectations from parents to peers only for certain specific norms is 90 consistent with work done by Brittain (1963, 1966). Using his "Cross-Pressures Test," in which parent and peer pressures are associated with different norms, he finds that when status norms and identity issues are salient in the items, a shift toward peer-endorsed norms takes place. However, when future aSpirations or achieve- ment in school is salient, shifts toward adult endorsed alternatives occurs. The present study found that males expected to share more often with other males than with females, while females expected to share and ask for help more often with other females than with males. The trend of the data indicates that both sexes have higher standards of conduct for same sex others rather than for Opposite sex others. There are no age changes. Emmerich et al. (1971) concluded from their data that there exists a "characteristic sex-role orientation" by which they mean higher standards Of interpersonal conduct for the self when interacting with females. This relationship was true for females at all ages and for adolescent males. (Before adolescence, however, males had higher standards for other males.) The question of sex differences in role expectations is difficult to interpret but intriguing to speculate about. Is the role of inter- acting with someone Of the same sex different from the 91 role of interacting with someone of the opposite sex? And if the answer to that question is yes, how are these roles different? It first should be pointed out that on the basis of the data gathered in this study, it appears that dif- ferent expectations for the self's behavior for male and female others is norm specific. Although a consis- tent trend is present favoring the same sex other, only in the case of "share" and "ask for help" is the dif- ference significant. Thus, the notion of a general, higher prosocial orientation towards females than towards males may be somewhat overstated. Although the idea of a "characteristic sex-role orientation" has a certain degree of intuitive plausibility to it--individuals do seem to be more "gentle," in a social sense, with females than with males--the idea presently lacks strong empirical support. Intuition aside, one might present a strong case for the Opposite, i.e., higher standards of conduct with a same sex other. Byrne and his associates have demon- strated that an individual is attracted to another person to the extent that the other person is seen as similar to the self (Byrne & Griffitt, 1966; Byrne & Clore, 1967). As Kohlberg (1966) has pointed out, this principle of attractiveness has generally been ignored in studies of sex-typing, although it is a small step from perceived 92 similarity to identification with and desire for norma- tive guidance from same sex individuals. It would not be unreasonable, therefore, to expect individuals to have higher interpersonal standards for same-sex peers than for Opposite-sex peers. This relationship should be particularly strong for pre-pubertal children, since the motive to establish an appropriate sex-role orientation is presumably stronger during this age period than at any later period. How puberty may influence sex-role orientation would be an interesting question to investigate. Although a male may perceive another male as more like himself than a certain female, the sexual aspect of the male-female relationship may be such that attractiveness will be greater in the latter relationship. One sus- pects that the very definition of "attractiveness" is quite different when referring to a same-sex or Opposite- sex other. In any case, the question of different standards for same and Opposite sex others remains Open. The notion of reciprocity has been explored elsewhere (Gouldner, 1960). As the term is used in this study, reciprocity simply means equal expectations for self and others. Given this more narrow definition, the findings of this study lead to a number of con- clusions. The ability to make a self-other distinction is present in middle childhood and the distinction, 93 itself, remains salient throughout adolescence. It appears, therefore, that the cognitive ability to make a self-other distinction is a primitive one requiring neither concrete nor formal Operational thought. Most theorists agree that the capacity to distinguish self from other is not present at birth but it probably is learned during the first few years of life. As the distinction is used here, it is clearly present by age nine, and in some instances is made most strongly at the earlier ages. NO simple develOpmental trend appears between middle childhood and adolescence. It is not true, for example, that children distinguish between self and other, while adolescents do not. Rather it appears that changes in self—other expectations occur for specific norms as a function of the changing nature of the developing individual's role in relation to others. There is some evidence that the social world of the child is fundamentally different from the adolescent's social world, and this will be explored shortly. Relationships between parents and self are less reciprocal than relationships between peers and self. Expectations for self and father are nonreciprocal for all five norms while expectations for self and mother are nonreciprocal for four norms. For peers, three norms are nonreciprocal for girl friend and only one norm for 94 boy friend. The parent-self distinction is a salient one at all ages, while the peer-self division is not as strong. In conjunction with the age changes in the parent-peer system, it seems probable that a reorgani- zation of the social world takes place in early adoles- cence. Children distinguish between the parent system and the peer system, and between the parent system and the self system. Expectations for peers and self, how- ever, are usually reciprocal. For children, age seems to be a critical dimension. A clear distinction is made between Old and young, whether young is peer or self. Adolescents, however, view the parent system and peer system as similar, and the peer system and self system as similar. The parent—self distinction remains, and expectations in this area are nonreciprocal. The following diagram illustrates these relationships: Child Adolescent Parent :2:; //Peer Parent ———— Peer / Self / Self / The social world of the adolescent appears to be a good deal more complex than the child's world. Parents and peers apparently occupy an equal status, although in different areas. Expectations for peers and self are reciprocal, while expectations for parents and 95 self are not. The evidence in regard to develOpmental changes tentatively supports a discontinuous or stage model of socialization. In summary, the data from this study support the following general conclusions: Status of other is a more critical social dimension for children than for adolescents. Sex Of self interacts with sex Of other and individuals generally have higher expectations for a same-sex other than an Opposite-sex other. This dimension is not as pervasive or as powerful as the age dimension is for children. Finally, the self-other distinction is a critical one for all individuals, especially when the distinction involves a parent-self dyad. APPENDICES APPENDIX A PARENT PERMISSION LETTER IUEPERUIIXIZl MICHIGAN STATE UNIVERSITY - East Lansing Department of Psychology Raymond Montemayor Department of Psychology Michigan State University East Lansing, Michigan 48823 Dear Parent: The class that your son or daughter is in at school has been chosen to be part of a large scale investigation of behavioral attitudes in children and adolescents. The study is an attempt to assess changes that may occur between childhood and adolescence in what students believe to be correct ways of behaving. The purpose of the study is to increase our general understanding of young people and it is in no way an evaluation. Almost 300 East Lansing students between the ages of 9 and 17 years will be asked to participate in the study. The study, itself, will involve answering a 40-item questionnaire and will take less than one hour to complete. Students will answer the questionnaire anonymously and their answers will be examined by age and not individually. The study has the approval of the East Lansing Board of Education (Dr. Robert Docking) and the support of the administration and teachers of your child's school. Participation in the study is, however, voluntary. Your son or daughter has the option of taking part or not. In addition, should you decide not to allow your son or daughter to take part in the study, then you should send a note to your child's teacher within the next two days. If we have not heard from you within the next two days, then we will assume that we have your permission to allow your child to take part in the study. If you have any questions, feel free to call me at 353-9166. Thank you for your cooperation. We!“ Raymond Montemayor Sincerely, 96 APPENDIX B EXTREME RESPONSE BIAS APPENDIX B EXTREME RESPONSE BIAS The tendency to use the extreme response cate- gories in a Lickert-type scale has been demonstrated to be related to a variety of subject characteristics. Osgood, Suci, and Tannenbaum (1957) using the semantic differential found that college students used the extremes less than a group Of American Legionnaires. Berg and Collier (1953) found a sex difference for extreme ratings on the Perceptual Reaction Test. Males used the extreme categories less Often than females. Zax, Gardiner, and Lowy (1964) found that ”maturity" was related to extreme response bias. Using a Rorschach measure of maturity-immaturity, they found that immature subjects used the extreme categories more Often than mature subjects. Finally, Light, Zax, and Gardiner (1965) investi- gated extreme response bias as a function of sex, IQ, and age. They did not find a sex difference. However, IQ was significant, high IQ children used fewer extreme response categories than low IQ children. Additionally, 97 98 and most important for the present study, there was a general decrease in extreme responses between grades 4 and 12. The finding of an age difference in extreme response tendency may be explained in the following two ways. First, one might argue that the extreme responses of young children are the result of an artifact in the test situation, itself, and do not accurately reflect the child's true endorsement Of the item. This type of explanation stresses the importance of the manner Of responding, i.e., Lickert scale, and views the child's extreme response bias as a function of this type of testing. The child is viewed as cognitively capable of making the subtle distinctions in answers that a Lickert scale requires, but for whatever reason, does not. How- ever, the findings for IQ indicate that cognitive factors play an important part in extreme response tendencies. A second explanation, therefore, is that the tendency to respond in the extreme is not a function of the test situation, but Of the child's limited cognitive ability. As Piaget has demonstrated, the young child lives in a world Of absolute judgments, while the world of the adolescent is more differentiated and subtle (Piaget, 1932). Therefore, the child's gross judgments of "always" or "never" are a true reflection of his perception of his world. 99 The position of the present study is that the extreme response tendencies Of younger children are a function of the child's developmental level and are not artifacts Of the test situation. The data indicating that maturity, age, and IQ all tend to decrease extreme responding seem to best fit the developmental level explanation. At this point, however, the test artifact position cannot entirely be eliminated. APPENDIX C TEST QUESTIONNAIRE APPENDIX C What Peogleggg These questions ask about different people in your life, and what they do. Some questions ask about yourself. Others ask about your mother and father. Still others ask about friends of yours. The questions about friends ask about your best friend who is a boy, or your best friend who is a girl. Write below the first name of the boy and the girl who are your best friends. Deep in mind this one boy and this one girl when answering the questions about friends. First name of your best friend who is a boy: First name of your best friend who is a girl: Think about each question carefully before answering. This is not a test, and there are no right or wrong answers. There is no time limit. Your answers will not be seen by anyone except the interviewer. Please answer all questions, even if you are not sure of your answer. Answer the questions in the order given. Some questions may seem similar to you but each is different and each question appears only once. Write the date and year of your birth below: Birthdate: Circle your sex: Male Female 100 101 Family Information: How many brothers do you have? What are their ages? How many sisters do you have? What are their ages? Draw a line from 0 through the number of grades of school that your father has completed. 0 1 2 _3_“4 _5_.6“_7W8 9 10 11 12 13 14 15 16 MA PhD Where does your father work? What does he do? (be specific) '“o-m-’--—-~_-.—-.-u .. -.. Draw a line from 0 through the number of grades of school that your mother has completed. 0 1 2 3 4 S 6 7 8 9 10 11 12 13 14 15 16 MA PhD Does your mother work? What does she do? (be specific) Instructions: The questions on the next 5 pages ask about how ygg_think peOple should act. Circle the answer that you think is best for each question. Please answer all questions even if you are not sure of the best answer. There are no right and wrong answers and there is no time limit. Remember to keep in mind the name of the boy and the girl that you wrote down as your best friends when you answer the questions about friends. 102 How often do you think you should tell the truth to your father? Never : Sometimes : Often ° 1 . Very often : 2 3 Always A 5 How often do you think your friend (a boy) should ask you for help? Never : Sometimes I 2 Often : Very_often : Alwgys 3 4 5 How often do you think you should share with your friend (a girl)? Never 3 Sometimes 1 l Often ' 2 . Very often : Always 3 4 5 How often do you think your mother should ask you for help? Never Sometimes 1 Often : Very often : Always 2 3 4 5 Never : Sometimes : Often : Very often : Alway§_ 2 3 4 5 How often do you think you should tell the truth to your friend (a girl)? 1 How often do you think you should share with your friend (a boy)? Never Sometimes : Often : Very often : 1 2 Always 3 4 5 How often do you think your father Should tell you what to do? Never Sometimes : Often : Very often : Alwayg l 2 3 4 5 How often do you think you should share with your mother? Never : Sometimes : Often Very often : Alwayg 1 2 3 4 5 How How How How How How How How 103 often do you think you should do what your mother tells you to do? Never : Sometimes : Often : Very often : Alwgyg l 2 3 4 5 often do you think your friend (a boy) should tell you what to do? Never : Sometimes : Often : Very often : Alwayg 1 2 3 4 5 often do you think your father should ask you for help? Never : Sometimes : Often : Very often : Alwaygy. 1 2 3 4 5 often do you think you Should do what your friend (a girl) tells you to do? Never : Sometimes : Often : Veryyoften : (Alyays l 2 3 4 5 often do you think your mother should share with you? Never : Sometimes : Often : Very often : Always 1 2 3 4 5 often do you think you should share with your father? Never : Sometimes : Often : Very often : Alwsys l 2 3 4 5 often do you think your friend (a boy) should tell the truth to you? Never : Sometimes : Often : Very often : Always l 2 3 4 5 often do you think you should tell your friend (a girl) what to do? Never : Sometimes : Often : Very often : Always l 2 3 4 5 How How How How How How How How 104 often do you think you should ask your friend (a boy) for help? Never : Sometimes : Often : Very often : Always 1 2 3 4 5 often do you think your friend (a girl) should do what you tell her to do? Never : Sometimes : Often : Very often : Always l 2 3 4 5 often do you think your father should share with you? Never : Sometimes : Often : Very often : Always l 2 3 4 5 often do you think you should tell the truth to your mother? Never 2 Sometimegfi : Often = Very‘gften : Agways 1 2 3 4 5 often do you think your friend (a boy) should do what you tell him to do? Never : Sometimes : Often : Very often : Always l 2 3 4 5 often do you think you should tell your father what to do? Never : Sometimes : Often : Verygoften : Alwaya 1 2 3 4 5 often do you think your friend (a girl) should ask you for help? Never : Sometimes : Often : Very often : Always l 2 3 4 5 often do you think you should tell your mother what to do? Never : Sometimes : Often : Very often : Alwayg 1 2 3 4 5 How How How How How How How How 105 often do you think your friend (a girl) should tell the truth to you? Never : Sometimes : Often : Very often : Alwayg 1 2 3 4 5 often do you think your father should tell the truth to you? Never : Sometimes : Often : Very often : Always 1 2 3 4 5 Often do you think you should ask your mother for help? Never : Sometimes : Often : Veryyoften : Always l 2 3 4 5 often do you think your friend (a boy) should share with you? Never : Sometimes : Often : Very often : Alwayg 1 2 3 4 5 often do you think you should do what your father tells you to do? Never : Sometimes : Often :fifiVery often : Always l 2 3 4 5 often do you think your mother Should tell the truth to you? Never : Sometimes : Often : Very often : Alwaya l 2 3 4 5 often do you think your friend (a girl) should tell you what to do? Never : Sometimes : Often : Very often : Alwayg l 2 3 4 5 often do you think you should do what your friend (a boy) tells you to do? Never : Sometimes : Often : Vegy often : Alwaya l 2 3 4 5 How How How How How How How How 106 often do you think you should ask your father for help? Never : Sometimes : Often : Very often : Always l 2 3 4 5 often do you think your mother should do what you tell her to do? Never : Sometimes : Often : Very often : Algays 1 2 3 4 5 often do you think your friend (a girl) should share with you? Never : Sometimes : Often : Very often : Always l 2 3 4 5 often do you think you should tell the truth to your friend (a boy)? Never Sometimes Often Very often Always l 2 3 4 5 often do you think your father should do what you tell him to do? Never : Sometimes : Often : Very often : Always 1 2 3 4 5 often do you think you should ask your friend (a girl) for help? Never : Sometimes : Often : Very often : Alway§_ l 2 3 4 5 often do you think your mother should tell you what to do? Never : Sometimes : Often : Very often : Always 1 2 3 4 5 often do you think you should tell your friend (a boy) what to do? Never : Sometimes : Often : Very often : Always 1 2 3 4 5 APPENDIX D ANALYSES OF VARIANCE FOR SIMPLE EFFECTS APPENDIX D Table D-l Analysis of Variance for Simple Effects (Age x Status) Source df MS F 0 for bl 1 69.62 53.55a 0 for b2 1 44.65 34.35a D for b3 1 6.66 5.12 0 for b4 1 7.61 5.85 0 for b5 1 0.78 0.60 Error 190 1.30 ap < .01 Table D-2 Analysis of Variance for Simple Effects (Age x Status) Source df MS F A for 01 4 24.27 18.67a A for 02 4 8.11 6.24a Error 4000 1.30 ap < .01 107 108 Table D-3 Analysis of Variance for Simple Effects (Age x Status x Norm [Share]) Source df MS F 0 for b1 1 9.02 6.94a D for b2 1 0.15 0.12 D for b3 1 1.05 0.81 Error 190 1.30 ap < .01 Table D-4 Analysis Of Variance for Simple Effects (Age x Status x Norm [Tell Truth]) Source df MS F 0 for 81 1 11.02 8.49a 0 for b2 1 2.03 1.56 0 for b3 1 0.40 0.31 0 for b4 1 0.75 0.58 0 for b5 1 6.00 4.62 Error 190 1.30 ap ‘<.01 109 Table D-S Analysis of Variance for Simple Effects (Age x Status x Norm [Do What Is Said]) Source df MS F 0 for 81 1 156.02 120.02a 0 for 82 1 124.25 95.58a 0 for b3 1 55.22 42.48a D for b4 1 77.00 59.23a 0 for b5 1 10.50 8.08a Error 190 1.30 ap < .01 Table D-6 Analysis of Variance for Simple Effects (Age x Status x Norm [Tell What To Do]) Source df MS F 0 for b1 1 25.60 19.69a D for b2 1 3.02 2.32 D for b3 1 3.02 2.32 D for b4 1 2.02 1.55 D for b5 1 1.30 a p < .01 110 Table D-7 Analysis of Variance for Simple Effects (Age x Status x Norm [Ask for Help]) Source df MS F 0 for b1 1 24.02 18.48a 0 for b2 1 13.80 10.62a 0 for b3 1 0.22 0.17 0 for b4 1 1.05 0.81 0 for b5 1 0.15 0.12 Error 190 1.30 ap.'<.01 Table 0-8 Analysis of Variance for Simple Effects (S ex x Norm) Source df MS F A for s1 1 10.35 6.39a‘ A for e3 1 0.03 0.02 A for e4 1 6.48 4.00a A for e5 1 8.20 5.06a Error 950 1.62 ap < .05 111 Table D-9 Analysis of Variance for Simple Effects (Sex x MF) Source df MS F A for f1 1 4.33 2.09 A for £2 1 16.56 8.00a Error 380 2.07 ap < .01 Table D-lO Analysis of Variance for Simple Effects (Sex [Males] x Norm x MF) Source df MS F 8 for e1 1 7.03 15.98a F for e2 1 0.43 0.98 F for e3 1 0.64 1.45 F for e4 1 0.73 1.66 F for e5 1 1.96 4.45 Error 950 0.44 a p < .01 112 Table D-ll Analysis of Variance for Simple Effects (Sex [Females] x Norm x MF) Source df MS F 8 for el 1 16.00 36.36a F for e2 1 1.83 4.16 F for e3 1 1.11 2.52 F for e4 1 2.11 4.80 8 for e5 1 11.23 25.52a Error 950 0.44 ap < .01 Table D-12 Analysis Of Variance for Simple Effects (Sex [Males] x Status x Norm) Source df MS F D for el 1 3.07 3.99 D for e2 1 1.33 1.73 0 for e3 1 179.56 233.19a 0 for e4 1_ 22.57 29.31a 0 for e5 1 16.00 20.78a Error 950 0.77 a 113 Table D-l3 Analysis of Variance for Simple Effects (Sex [Females] x Status x Norm) Source df MS F 0 for el 1 5.76 7.48a D for e2 1 0.03 0.04 0 for e3 1 191.83 249.13a 0 for e4 1 5.07 6.58 0 for e5 1 7.03 9.13a Error 950 0.77 ap < .01 Table D-l4 Analysis of Variance for Simple Effects (Age x Source) Source df MS F 8 for al 1 21.16 37.79a 8 for a2 1 4.10 7.31a B for a3 1 2.80 5.00 B for a4 1 0.95 1.70 B for as l 3.06 5.46 Error 190 0.56 a p (<.01 114 Table D-15 Analysis of Variance for Simple Effects (Age x Source x Status [Parents] x Norm [Share]) Source df MS F 0 for b1 1 30.62 54.68a 0 for 82 1 1.80 3.21 0 for b3 1 0.62 1.11 0 for b4 1 2.25 4.02 0 for b5 1 0.02 0.04 Error 190 0.56 ap,<:.01 Table 0-16 Analysis Of Variance for Simple Effects (Age x Source x Status [Peers] x Norm [Share]) Source df MS F 0 for b1 1 8.55 15.27a C for b2 1 5.25 9.38a C for b3 1 1.05 1.88 C for b4 1 0.50 0.89 C for b5 1 3.30 5.89 Error 190 0.56 ap <:.01 115 Table D-l7 Analysis Of Variance for Simple Effects (Age x Source x Status [Parents] x Norm [Tell Truth]) Source df MS F 0 for bl 1 0.62 1.11 0 for b2 1 0.00 0.00 0 for b3 1 6.40 11.43a 0 for b4 1 7.65 13.66a 0 for b5 1 16.90 30.18a Error 190 0.56 ap < .01 Table D-18 Analysis of Variance for Simple Effects (Age x Source x Status [Peers] x Norm [Tell Truth]) Source df MS F C for b1 1 0.15 0.27 C for b2 1 0.03 0.05 C for b3 1 0.30 0.54 C for b4 1 0.62 1.11 C for b5 1 1.40 2.50 Error 190 0.56 116 Table D-l9 Analysis of Variance for Simple Effects (Age x Source x Status [Parents] x Norm [Do What Is Said]) Source df MS F 0 for 51 1 213.90 381.97a 0 for b2 1 107.25 191.52a 0 for b3 1 104.00 185.71a 0 for b5 1 13.80 24.64a Error 190 0.56 ap.<.01 Table D-20 Analysis of Variance for Simple Effects (Age x Source x Status [Peers] x Norm [DO What Is Said]) Source df MS F C for b1 1 1.22 2.16 C for b3 1 3.60 6.43 C for b4 1 0.10 0.18 C for b5 1 0.15 0.27 Error 190 0.56 117 Table D-21 Analysis of Variance for Simple Effects (Age x Source x Status [Parents] x Norm [Tell What To DO]) Source df MS F 0 for bl 1 131.40 234.64a 0 for b2 1 64.02 116.11a 0 for b3 1 25.60 45.71a 0 for b4 1 39.00 69.64a 0 for 85 1 4.90 8.75a Error 190 0.56 ap < .01 Table D-22 Analysis of Variance for Simple Effects (Age x Source x Status [Peers] x Norm [Tell What To Do]) Source df MS F C for bl l 3.60 6.43 C for b2 1 0.22 0.39 C for b3 1 0.15 0.27 C for b4 1 0.00 0.00 C for b5 1 0.23 0.41 Error 190 0.56 118 Table D-23 Analysis of Variance for Simple Effects (Age x Source x Status [Peers] x Norm [Ask for Help]) Source df MS F C for b1 1 1.05 1.88 C for b2 1 2.25 4.02 C for b3 1 1.40 2.50 C for b4 1 0.22 0.39 0 for b5 1 11.55 20.63a Error 190 0.56 ap < .01 Table D-24 Analysis of Variance for Simple Effects (Source x Norm [For Fatherl) Source df MS F 0 forel 1 4.85 14.26a 0 for e 1 7.57 22.25a 0 for e3 1 299.29 880.26a 0 for e4 1 126.57 372.25a 0 for e5 1 4.41 12.97a Error 950 0.34 a 119 Table D-25 Analysis of Variance for Simple Effects (Source x Norm [For Mother]) Source df MS F 0 for el 1 13.69 40.26a C for e2 1 7.03 20.66a C for e3 1 175.57 516.37a 0 for e4 1 93.13 273.90a C for e5 1 0.31 0.90 Error 950 0.34 ap < .01 Table D-26 Analysis Of Variance for Simple Effects (Source x Norm [For Boy Friend]) Source df MS F 0 for e1 1 12.61 37.07a C for e3 1 1.96 5.76 C for e4 1 0.00 0.00 C for e5 1 2.25 6.62 Error 950 0.34 a p < .01 | .l’. I'll! .II II] IIIIJII '1 Illll'llll 120 Table D-27 Analysis of Variance for Simple Effects (Source x Norm [For Girl Friend]) Source df MS F c for el 1 4.00 11.76a C for e2 1 0.91 2.66 c for e3 1 2.89 8.50a C for e4 1 2.11 6.19 C for e5 1 10.89 32.03a Error 950 0.34 a p < .01 REFERENCES REFERENCES Argyle, M., & Little, B. Do personality traits apply to social behavior? Journal for the Theory of Social Behavior, 1972, 3, 1-35. Asch, S. Social psychology. New York: Prentice-Hall, 1952. Benedict, R. Continuities and discontinuities in cultural conditioning. Psychiatry, 1938, 1, 161-167. Berg, 1., & Collier, J. Personality and group differences in extreme response sets. Educational and Psy- chological Measurement, 1953,13, 164-169. L/ . Biddle, B., & Thomas, E. Role theory: Concepts and research. New York: Wiley, 1966} Bowerman, C., & Kinch, J. Changes in family and peer orientation of children between the fourth and tenth grades. Social Forces, 1959, 31, 206-211. Bowers, K. Situationism in psychology: An analysis and a critique. Psychological Review, 1973, fig, 307-336. Brim, 0. Personality development as role-learning. In I. Iscoe and H. Stevenson (Eds.), Personality development in children. Austin: University of TEkas Press, 1960. Brim, O. Adolescent personality as self-other systems. Journal of Marriage and the Famil , 1965, £1, ./ Brim, 0., & Wheeler, S. Socialization after childhood: Two essays. New York: Wiley, 1966. Brittain, C. Adolescent choices and parent-peer cross- pressures. American Sociological Review, 1963, gg, 385-391. 121 122 Brittain, C. Age and sex of siblings and conformity towards parents versus peers in adolescence. Child Development, 1966, 31, 709-714. Byrne, D., & Clore, G. Effectance arousal and attraction. Journal of Personality and Social Psychology Monograph? 1967, 6(4). Byrne, D., & Griffitt, A. Developmental investigation of the law of attraction. Journal of Personality and Social Psychology, 1966, E, 699;702. Campbell, E. Adolescent socialization. In D. Goslin (Ed.), Handbook of sogialization theory and research. ’Chicago: Rand McNally, 1969. Chandler, M. Egocentrism and antisocial behavior: The assessment and training of social perspective- taking skills. Developmental Psychology, 1973, 9, 326-332. Conger, J. Adolescence and_youth: Psycholggical develop- ment in a changing world. New YorE: Harper and Row, 1973. Davis, K. Human society. New York: Macmillan, 1949. Douvan, E., & Adelson, J. The adolescent experience. New York: Wiley, 1966. Emmerich, W. Young children's discrimination of parent and child roles. Child Develgpment, 1959, 30, 403-419. Emmerich, W. Family role concepts of children ages six to ten. Child Development, 1961, 33, 609-624. bEmmerich, W. Socialization and sex-role development. In P. Baltes and K. Schaie (Eds.), Life-span developmentalypsychology: Personality and socialization. New York: Academic Press, 1973. Emmerich, W., Goldman, K., & Shore, R. Differentiation and development of social norms. Journal of Personality and Social Psychology, 1971, lg, 323-353. . Endler, N. The person versus the situation--A pseudo issue? A response to Alker. Journal of Per- 123 Endler, N., & Hunt, J. S-R inventories of hostility and comparison of the proportion of variance from persons, reSponses, and situations for hostility and anxiousness. Journal of Personality and Social Psychology, 1968, 2, 309-315. Endler, N., & Hunt, J. Generalizability of contributions from sources of variance in the S-R inventories of anxiousness. Journal of Personality, 1969, 37, 1-24. Erikson, E._ Childhood and society. New York: Norton, 1950. Feffer, M. DevelOpmental analysis of interpersonal behavior. Psychological Review, 1970, 11, 193- 214. Feffer, M., & Gourevitch, V. Coqnitive aspects of role- taking in children. Journal of Personality, Feuer, L. (Ed.) Marx and Engels: Basic writings on politics and philosophy. New Yofk: Doubleday, 1959} Flavell, J. Role-taking and communication skills in children. Young Children, 1966, 21, 164-177. (a) Flavell, J. The development of two related forms of cognition: Role-taking and verbal communication. In A. Kidd and J. Rivoire (Eds.), Perceptual development in children. New York: International UnIVersities Press, 1966. (b) Flavell, J., Botkin, P., Fry, C., Wright, J., & Jarvis, P. The development of role-taking and communication skills in Ehildren. New YorE: Wiley, 1968. Garfinkel, H. Studies in ethnomethodology, Englewood Cliffs, N.J.: Prentice-Hall, l9 . Goffman, E. The resentation of self in everyday life. New YorE: Dofibleday, 1959. Gouldner, A. The norm of reciprocity: A preliminary statement. American Sociological Review, 1960, 22, 161-178. Hartshorne, H., & May, M. Studies in the nature of character. Vol. I. Studies in deceit. New York: Macmillan, 1928. 1 v 124 Hartup, W. Peer interaction and social organization. In P. Mussen (Ed.), Carmichael's manual of child psychology. New York: Wiley, 1970. Hoffman, M. Moral development. In P. Mussen (Ed.), Carmichael's manual of child psychology. New York: Wiley,ii970. Homans, G. The human_group. New York: Harcourt, Brace, 1950. Kagan, J. The acquisition and significance of sex- typing and sex-role identity. In M. Hoffman and L. Hoffman (Eds.), Review of child development research. Vol. 1. New York: Russell Sage Foundation, 1964. Kohlberg, L. A cognitive-developmental analysis of children's sex-role concepts and attitudes. In E. Maccoby (Ed.), The development of sex differences. Stanford: Stanford University Press, 1966. Kohlberg, L. Stage and sequence: The cognitive- developmental approach to socialization. In D. Goslin (Ed.), Handbook of sggialization theory and research. Chicago: Rand’McNally, 1969. Kohlberg, L. From is to ought: How to commit the naturalistic fallacy and get away with it in the study of moral development. In T. Mischel (Ed.), Cognitive development and epistemology. New York: Academic Press, 1971. Lewin, K. Field theory in sogial science: Selected theoretical papers. D. Cartwright TEd.), New York: Harper, 1951. Light, C., Zax, M., & Gardiner, D. Relationship of age, sex, and intelligence level to extreme response style. Journal of Personality and Social Psy- chology, 1965, 2, 907fi909. Manis, M., & Meltzer, L. (Eds.) Symbolic interaction. New York: Allyn and Bacon, 1967. Maccoby, E. The choice of variables in the study of socialization. Sociometgy, 1961, 21(4). ‘/ Maccoby, E. (Ed.) The development of sex differences. Stanford: Stanford University Press, 1966. 125 Mead, G. Mind, selgyand society. Chicago: University of Chicago Press, 1934. Mischel, W. Personality and assessment. New York: Wiley, 19692 Mischel, W. Toward a cognitive social learning reconcep- tualization of personality. Psychological Review, 1973, fig, 252-283. Mussen, P. Early sex-role development. In D. Goslin (Ed.), Handpook of sogialization theory and research. ChiCago: RandiMcNally, 1969. L/ ’/ Neiman, L., & Hughes, J. The problem of the concept of role--A re-survey of the literature. Social Forces, 1951, 32, 141-149. Osgood, C., Suci, G., & Tannenbaum, P. The measurement gfmeanin . Urbana: University of Illinois Press, 19 7. Parsons, T. The social system. Glencoe, I11.: The Free Press, 1951. Parsons, T., & Bales, R. Family sgcialization and inter- action process. Glencoe, 111.: THe Free Press, 1955. Piaget, J. The moral 'udgment of the child. London: Kegan Paul, 19 2. Piaget, J., & Inhelder, B. The child's conception of Space. London: Routledge and’Kegan Paul, 1956. Rabban, M. Sex-role identification in young children in two diverse social groups. Genetic Psychology Monographs, 1950, 32, 81-158. Rommetveit, R. Social norms and roles. Minneapolis: University ofiMinnesota Press, 1954. Rose, A. A systematic summary of symbolic interaction theory. In A. Rose (Ed.), Human behavior and social processes. Boston: Houghton Mifflin, 1962. Sarbin, T., & Allen, V. Role theory. In G. Lindzey and E. Aronson (Eds.), The handbook of socia1_psy- cholo . (2nd ed.) Reading, Mass.: Addison- Wesley, 1968. 126 Secord, P., & Backman, C. Social psychology. New York: McGraw-Hill, 1974. Selman, R. Taking another's perspective: Role-taking development in early childhood. Child Develgp- Sewell, W. Some recent developments in socialization theory and research. The Annualsof the American Agademy of Political and Social Science, 1963, 349, 163-181. L/' '__— . Thomas, E. Role theory, personality and the individual. In E. Borgatta and W. Lambert (Eds.), Handbook of personality theory_and research. Chicago: Rand McNally, 1969. Westley, W., & Elkin, F. The protective environment and adolescent socialization. Social Forces, 1956, 35, 243-249. Winer, B. Statistical principles in experimental desigp. New York: McGraw-Hill, 1971. Wrong, D. The oversocialized conception of man in modern sociology. American Sociological Review, 1961, 26, 183-193. Zax, M., Gardiner, D., & Lowy, D. Extreme response ten- dency as a function of emotional adjustment. Journal of Abnormal and Social Psychology, 1964, 62, 654-657. HICHIGAN STATE UNIV. LIBRARIES llWWI”WWINHWHIW‘\IHIWIHNWIWHI 31293106748969