SEX ROLE PREFERENCE 0F PRE-SCHGOL EGYS‘Q [N fiELANON TO PRESENCE, SEX, AND POSITBON 0F SIBUNGS Thesis {Pote- ftrm Degree of M. A. MICHEGAN STATE UNIVERSE? Phyfiis F. Newman 1968 __-#M.-m.# —-'* " mun LIDRARY ”this" 59!» University MIMIHHllilHWllUlfllliHINlllltlfli\IHHILHW I.~ 3 1293 10441 [AM-{14m M M (We)??? : 3;? géfi ABSTRACT SEX ROLE PREFERENCE OF PRE-SCHOOL BOYS IN RELATION TO PRESENCE, SEX, AND POSITION OF SIBLINGS by Phyllis F. Newman This study deals with the relationship between sex role preference of pre-school boys and the presence, sex, and position of siblings. To study this problem, sixty—five pre- school boys were administered the It Scale for Children (ITSC). Ten boys had older brothers only, ten boys had younger brothers only, ten boys had younger sisters only, and ten boys had older sisters only. Twenty-five boys had no siblings. Since the sample was non—random, non-parametric statis- tics were used on the data, Specifically the Kruskal Wallis One Way Analysis of Variance, and the Median test. The Hypotheses were: 1. Boys with older brothers will have the highest mascu- line score on sex role preference as measured by the ITSC. 2. Boys with younger brothers will have the next highest masculine score on sex role preference as measured by the ITSC. 5. Boys with younger sisters will have the next lowest Phyllis F. Newman masculine score on sex role preference as measured by the ITSC . 4. Boys with older sisters will have the lowest mascu- line score on sex role preference as measured by the ITSC. 5. Boys without siblings will have a mean score sig- nificantly different from boys with siblings in sex role preference as measured by the ITSC. None of the hypotheses were supported. SEX ROLE PREFERENCE OF PRE-SCHOOL BOYS IN RELATION TO PRESENCE, SEX, AND POSITION OF SIBLINGS BY Phyllis F. Newman A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Home Management and Child Development 1968 ACKNOWLEDGMENTS Acknowledgment is made of the help received from the members of my Master's Committee, Miss Phyllis Lueck, and Mrs. Thelma Hansen, and my committee chairman, Dr. Martha Dale. Dr. Sarah Hervey was also of great help in statisti— cal analysis of the data. Of course, the study would have been quite impossible without the cooperation of the boys tested, the permission granted by their parents, and the assistance of the nursery school teachers. The nursery schools which participated in the study were: The Storybook Nursery, California; The Highland Nursery, California; The Redwood Christian Nursery, California; The Little Ones Nursery, California; The Baptist Pre-School, California; The Franklin Nursery, California; The Lakeside Nursery, California; The Fleur de Lis Nursery, California; The Christopher Robin Pre-School, California; The University Methodist Co-Operative Nursery, Michigan; The Peoples Church Co-Operative Nursery, Michigan; The Spartan Co-Operative Nursery, (Michigan State University), and the Laboratory Pre-School, (Michigan State University). ii TABLE OF CHAPTER ACKNOWLEDGMENTS. . . . LIST OF TABLES . . . . I. INTRODUCTION . . . . . Assumptions Objectives Basic Terms Hypotheses II. REVIEW OF LITERATURE . Introduction CONTENTS Page Various Theories of Identification Current Research Findings in Sex Role Identification III. PROCEDURE. . . . . . . . . . . . . . . . . . . 42 Description of the Instrument Reliability Validity Criticisms of the The Study IV. DATA ANALYSIS. . . . . It Scale for Children V. RESULTS, IMPLICATIONS, RECOMMENDATIONS, SUWY O O C O O 0 REFERENCES . . . . . . APPENDIX . . . . . . . . . . . . . . . . . . . 50 . . . . . . . . . . . . 56 . . . . . . . . . . . . 65 . . . . . . . . . . . . 71 iii 5. 4. 5. LIST OF TABLES Classification of Testing Conditions. . . . . . Means of Samples 1-4. . . . . . . . . . . . . . Ranks of Samples 1-4. . . . . . . . . . . . . . Range of Samples 1—4. . . . . . . . . . . . . . Median Score for Combined Group of Boys Having Siblings and Boys without Siblings. . . . . . . Range of Scores for Boys Having Siblings and Boys without Siblings . . . . . . . . . . . . . Percentages of Boys Object Choices on the It Scale for Children. A Comparison of Choices Made by Boys in Newman's Study (1968) and Brown's Study (1956). . . . . . . . . . . . . . iv Page 47 52 55 55 54 54 62 CHAPTER I INTRODUCTION CHAPTER I INTRODUCTION It has been said of the times in which we live that the only thing certain is change itself. Certainly this has been true in the area of male and female roles. In the last generation alone, significant changes have taken place in the traditional concepts of what is masculine and what is feminine. It has been the concern of researchers and others whether such changes have been abrupt enough to be considered a cultural revolution or gradual enough to be simply degrees of cultural variation. Whichever is the case, there are a number of implications for individual, group, and institu- tional behavior. Brown has raised the following questions regarding the implications of cultural change: (1) What are some of the changes in sex roles that have taken place? (2) In which role (masculine or feminine) have the changes been more pronounced? (5) How have these changes effected the individual's life adjustment, and the relationship of the sexes with each other? (4) What is the effect on boys and girls at the present time, and what will the future effect be?1 1Daniel G. Brown, "Sex—role develOpment in a changing culture," Psychological Bulletin, LX, (July, 1958), p. 252. It is this last question which is particularly pertinent to child study, for the child‘s acquisition of normal sex role behavior is fundamental to total personality deve10pment and adjustment. Two reasons indicate a need for better under- standing of the process whereby a girl comes to adopt a feminine role and a boy a masculine one. The first is theo- retical, i.e., there is much speculation concerning the nature and dynamics of sex role adjustment, based largely on the study of adults. However, relatively little study has dealt specifically with sex role development in children. The second is a practical consideration and is based on the increasing recognition by workers in clinical psychology and psychiatry that difficulties of distortions in sex role adjustments appear to be functionally related to the occurrence of personality maladjustment and certain forms of emotional disorders. This suggests a direct link between childhood learning and development in sex role behavior and adult personality disturbances.l There is consensus that the individual learns to identi— fy with a given sex role, to prefer one role or the other, and to adopt aspects of one role or the other. Recent stud— ies2 have shown that chromosomal sex and gonadal sex can be overridden by learning experiences. The learned aSpect of sex or gender role acquisition is crucial to the hypotheses 3 of this paper. Furthermore, Brown states that sex role 1Daniel G. Brown, "Masculinity—femininity deve10pment in children," Journal of Consulting Psychology, XXI, (June, 1957), p. 197. 21bid., p. 158. 3Ibid., p. 155. identification does not emerge as an automatic unfolding, but rather results from familial and other influences as the individual develOpS. This brings us to the statement of the concern of this paper which is: are the presence, sex and position of sib- lings related to a boy's sex role development? In discussing the influence of siblings in the child's sex role acquisition, Brown in 1956 reported no difference for girls, but a seven point mean difference significant at the .05 level between boys who had sister(s) only and those who had brother(s) only. The former group was less masculine in their reSponses than the latter. There was also a differ- ence of about the same magnitude between boys having sister(s) only, and those having both brothers and sisters. This dif- ference approached significance (.10 level). Brown used the It Scale for Children (ITSC). These results suggest the possibility of some degree of feminization in the case of boys who have sisters only as siblings. Whether the child's siblings were older or younger than S was not considered. It is possible that the influence of an older sister on a boy's developing sex role pattern would be different from the in— fluence of a younger sister. "In any event, position of a child in the family with reSpect to the presence and absence of siblings of the same or opposite sex undoubtedly has a formative influence on his or her sex-role development and should receive further study."1 In a study conducted by Helen Koch2 reported in the same year it was found that boys with sisters were less masculine than boys with brothers. Furthermore, it was found that dif- ferences associated with ordinal position are frequently contingent on sex of child and sex of sibling.3 For a further discussion of Koch's findings see Chapter II, the section on sibling influences. Hartup and Zook in 1960 reported a study using the ITSC and found that their data did not indicate a relationship between birth order and the acquisition of sex-role prefer- ences. However, they stated that they did not know if Older siblings of a particular sex (e.g., older brothers only, or older sisters only) effect the deve10pment of such preferences, as their present investigation did not offer a means of check- ing findings of this kind.4 Sutton-Smith and Rosenberg in 1965 hypothesized that siblings reinforce their own sex traits in the opposite 1Daniel G. Brown, "Sex-role preference in young children, Psychological Monographs, LXX, No. 14 (1956), p. 17. 2Helen L. Koch, "Attitudes of young children toward their peers as related to certain characteristics of their siblings, Psychological Monographs, LXX, No. 19 (1956), pp. 1-41. 3Ibid., p. 40. 4rWillard W. Hartup, and Elsie A. Zook, "Sex-role prefer- ence in three and four year old children," Journal of Consult- ing Psychology. XXIV, (October, 1960), p. 425. sibling irreSpective of that sibling's sex; and that older siblings, because of their greater power, have a greater reinforcement value for younger siblings than vice versa. Using the MMPI on college males and females they found that in general boys with sisters yield a more deviant profile than boys with brothers. Boys without siblings appeared in a high- ly favorable light in terms of the total MMPI configuration. The sex of the subjects sibling, they concluded, has a defi— nite effect on the subjects sex preference.1 Richard Q. Bell, reporting in the 1965 Annual Review of Psychology, noted that further studies using the ITSC should control for sex of siblings.2 The major assumption of this paper is: That the It Scale for Children (ITSC) is a reliable instrument. (Test—retest, interval approximately one month: r .71 for boys, .84 for girls.)3 This is an individually administered structured projective. For a further explana- tion of the ITSC see Chapter III. The objectives of this paper are: 1. To determine if there is a difference in the sex role preference of boys without siblings as compared with boys 1Brian Sutton—Smith, and B. G. Rosenberg, "Age changes in the effects of ordinal position on sex-role identification," Journal of Genetic Psychology, CVII, (1965), p. 69. 2Richard O. Bell, "Developmental psychology," Annual Review of Psychology, ed. Paul R. Farnsworth, Palo Alto, California, Annual Reviews, Inc., (1965), XVI, p. 24. 3Brown, Psychological Monographs, LXX, No. 14 (1956), p. 6. who have siblings. 2. To determine if male siblings and older siblings have a greater effect on boys' sex role preference than female sib- lings and younger siblings. The basic terms are: 1. Male maleness and female femaleness should be used to refer only to the biological aspects of sexuality. Masculine, masculinity and feminine, femininity refer to psychological social characteristics of behavior.1 2. Sex-role is that modal system of reponses which consti- tute the culturally expected behavior of a member of a particular sex.2 3. Identification refers to the general process by which a person learns the role of another by interacting with him.3 4. Sex—role identification refers to the actual adoption of behavior characteristics of one sex or the other, not simply to the desire to adopt such behavior. The term is reversed for reference to the introjection and incorporation of the role of a given sex and to the basic underlying reactions 1Daniel G. Brown, and D. B. Lynn, "Human sexual deve10p- ment: an outline of components and concepts," Journal of Marriage and the Family, XXVIII, No. 2 (May, 1966), pp. 156- 157. 2M. Rabban, "Sex-role identification in young children in two diverse social groups," Genetic Psychology Monographs, LXII, (1950), p. 97. 3Ibid., p. 98. Also in J. W. M. Whiting, "Resource mediation and learning by identification," Personality Develop— ment in Children, ed. I. Iscoe, and H. W. Stevenson, (Austin: University of Texas Press, 1960), p. 115. characteristic of that role.1 Other authors2 are not so precise in their definition stating that sex role identity is the degree to which an individual regards himself as masculine or feminine. 5. Sex-role preference refers to the desire to adOpt the behavior associated with one sex or the perception of such behavior as preferable or more desirable.3 6. Sex-role adoption refers to overt behavior characteristics of a given sex. This is contrasted with sex role identifi— cation which refers to a more basic internalized process in which behavioral characteristics of one sex role or the other are incorporated.4 The prime Operational definition of this paper is: 1. Sex-role preference may be operationally defined in terms of preferential responses of children to sex-typed objects and activities.5 The hypotheses are: 1. Boys with older brothers will have the highest masculine score on sex role preference as measured by the ITSC. 1Brown and Lynn, p. 155. 2Jerome Kagan, "Acquisition and significance of sex typ- ing and sex-role identity," Review of Child Development Research I, ed. Lois W. Hoffman and Martin L. Hoffman, (New York: Russell Sage Foundation, 1965). P. 159. 3Brown and Lynn, p. 157. 4Ibid. 5Brown, Psych. Monographs, LXX, No. 14 (1956), p. 4. 2. Boys with younger brothers will have the next highest masculine score on sex role preference as measured by the ITSC. 5. Boys with younger sisters will have the next lowest masculine score on sex role preference as measured by the ITSC . 4. Boys with older sisters will have the lowest masculine score on sex role preference as measured by the ITSC. 5. Boys without siblings will have a mean score significantly different from boys with siblings in sex role preference as measured by the ITSC. CHAPTER II REVIEW OF LITERATURE CHAPTER II REVIEW OF LITERATURE Introduction Formerly it was thought that the psychological differ- ences of masculine and feminine were constitutional. However, the present consensus is that the determinants are environmental conditioning and social learning experiences of the individual.1 Brown and Lynn have stated that This rather recent investigations suggest that the psycho- sexual status of the individual is undifferentiated at birth. The individual begins life psychosexually plastic, capable of develOping along a variety of lines depending upon the definition of sex roles in his particular culture as well as his unique learning ex- periences in the first few years of life especially. This psychosexual plasticity has been convincingly demonstrated by research showing that hermaphroditic children, i.e., those with a mixture of inconsistency of male and female components, usually grow up as masculine or feminine depending on the sex assigned them and the sex role in which they are reared. Research also suggests, however, that at least as far as sex role identity is concerned, this plasticity does not persist beyond early childhood; once a masculine or feminine sex role is established, it may be extremely difficult for this Basic pattern to be changed or reversed in later life. idea that sex-role is a learned phenomenon is shared by 1Brown and Lynn, p. 157. 2Ibid. 11 12 other researchers too.l Margaret Mead's cross cultural studies add further evidence to this concept. In her book Male and Female, she discusses the many varieties of attributes and qualities dichotomized as male or female in various cultures. Though this division is often contradictory and by and large arbi- trary, there has been no known culture that has said, articulately, that there is no difference between men and women.2 The concept that sex—role acquisition is a learned phenomenon is central to this paper, which is an attempt to test some of the variables in this learning process.3 It is generally agreed that this learning process begins through the child's imitation of a model. As Lazowick4 states, 1Ruth E. Hartley, "A developmental View of female sex- role identification," Role Theory: Concepts and Research, ed. Bruce Biddle, and Edwin Thomas, (New York: John Wiley, 1966), pp. 555-56. 2Margaret Mead, Male and Female, (New York: William Morrow and Company, 1949), pp. 5-12. 3Other writers corroborate the concept of culturally conditioned sex role. These include: C. Landreth, "Four year old's notions about sex appropriateness of parental care and companionship activities," Merrill-Palmer Quarterly, IX, (1965), p. 175; J. Bieliauskas, "Recent advances in the psy- chology of masculinity and femininity," Journal of Psychology, LX. (1965), p. 256. 4L. M. Lazowick, "On the nature of identification," Journal of Abnormal and Social Psychology, LI (1959), p. 176. Also Jerome Kagan and Henker, "Developmental psychology," Annual Review of Psychology, XVII, (1966), p. 27. 15 imitative behavior is the deve10pmental root of identifica- tion. Maccoby elucidates this point explaining that all role taking is imitation, but not all imitation is role tak- ing. ~Role behavior is the expected behavior characteristic of a position. A child's imitation of someone with whom he interacts may properly be called "taking the role of another" only if the action imitated is inappropriate for a child and is appropriate instead for the occupants of some other posi- 1 A child acquires a repertoire of actions tion or status. by covertly practicing (imitating) the actions characteristic of the adults with whom he interacts most frequently and who control the resources that he needs. This covert role play- ing is also a means of learning reactions toward the self.2 Lazowick discusses three ways in which the term identi- fication is used in literature treating this subject. The first of these he labels pseudo-identity, and states that this is the situation where the subject behaves as if he and the model were one and the same person. The second he labels imitation, stating that most definitions of identification fall here. "Identification takes place when one person copies another person, in this sense it is practically synonymous with imitation." A distinction has been drawn on the grounds that identification refers to the action of the entire 1Eleanor Maccoby, "Role taking in childhood and its consequences for social learning," Child Develppment, XXX, (1959), pp. 241-242. 21bid., p. 251. 14 personality, while imitation is more restrictive in terms, referring to isolated skills or acts. Furthermore, identi- fication usually presupposes an alteration of the ego after a pattern set by the model. The third aspect Lazowick labels as personality change and characterized this as the Freudian point of view. According to this usage, identification occurs at the point where the superego takes the place of the parental function (introjection). Identification is said to have occurred, it is the accomplished fact, not the pro- cess (introjection or learning) which is referred to as identification. Identification is the result of "taking into the self."1 A further review of identification theories will be presented in the next section titled Various Theories of Identification. The reason for such concern over the construct of identi- fication is that studies appear to indicate that psychological masculinity and femininity are related to the identification 2 process. This involves the relationship between child and parents but especially with the same sex parent. lLazowick, pp. 175-185. 2A. F. angrilli, "The psychosexual identification of pre- school boys," Journal of Genetic Psychology, XCVII, (1960), pp. 529-540. See also J. Kagan, Review of Child Development Research, ed. Hoffman and Hoffman, pp. 145-146; and P. Mussen and L. Distler, "Child rearing antecedents of masculine identi- fication in kindergarten boys," Child Development, XXI, (1960), p. 90. 15 Sears elaborates on the child's need of models in the learning process of develOping the appropriate sex identifi- cation. He states that it is these models upon which be- havior the child may pattern his own. For a boy child the father is usually the chief model. Furthermore, there must be someone available who has a sound knowledge (not necessar- ily verbalized) of what constitutes right or wrong sex-typed behavior to provide a continuous rewarding and punishing of such actions.1 These models are not exclusively parents, although it is thought that the parents are the primary models. Other persons who may function to greater or lesser degrees as models are teachers, peers, and siblings.2 Hartley reports Piaget (1962) as saying that imitative behavior is both meaningful and perceived as related to the child's interests.3 Kagan further outlines the determinants for sex role identity. The first is a perception of simi- larity to the same sex parent. The second is the degree to which the child adopts the games, and masters the skills that are traditionally encouraged for his sex. The child who perceives major elements of similarity to the parent of 1Robert R. Sears, and M. H. Pentler, and Pauline 8. Sears, "Effect of father separation on pre-school children's doll play aggression," Child Development, XVII, (1946), 219- 245. 2Pat Minuchin, "Sex-role concepts and sex typing in childhood as a function of school and home environments," Child Development, XXXVI, (1965), pp. 1055-1048. 3Hartley, p. 559. 16 the same sex will initially regard himself as masculine (or feminine), for the parents are the original prototypes of masculinity and femininity for the young child.1 The establishment of an optimally strong identification requires that three conditions be met: (a) the model must be perceived as nurturant to the child; (b) the model must be perceived as being in command of desired goals, eSpecially power, love from others, and task competence in areas the child regards as important; and (c) the child must perceive-- before the identification belief begins its growth--some objective bases of similarity in external attributes or psycho— logical properties between himself and the model.2 Kagan goes on to say that there are at least three kinds of experiences that determine the degree to which an indi- vidual regards himself as masculine or feminine: (a) differ- ential identification with mother, father, parental surro- gates, older siblings and special peers; (b) acquisition of the attributes or skills that define masculine and feminine behavior; and (c) a perception that other peOple regard the individual as possessing appropriate sex-typed characteris— tics.3 When does identification begin, and at what age can one say that the process is complete? Brown and Lynn state that lKagan, Review of Child Develppment Research I, p. 145. 2Ibid., p. 147. 31bid., p. 146. 17 differentiation of sex roles is a gradual process beginning as early as the first and second year of life and becoming definitely established by or during the fifth year.1 On the basis of hermaphroditic cases of sex reassignment, Money concludes that the critical period for "gender imprinting" is between eighteen months and three years of age, beginning with the onset of mastery of language. The die is considered well cast by the age of six with major realignment of gender role and sexual identity rare after that.2 Vener attests to the early age at which the child antici- 3 Levin and Sears state that by age five pates adult roles. most children may be assumed to have developed their strong— est identification with the same sex parent.4 Kagan states that the child as young as four has dichotomized the world into male and female people and is concerned with boy-girl 5 differences. Hartley states that awareness of one's own sex identity would appear to be crucial for the conscious rejection 1Brown and Lynn, p. 158. 2Ibid., citing J. Money, "Sex hormones and other variables in human eroticism,“ in Man and Civilization: The Potential of Women, ed. S. M. Farber and R. H. L. Wilson, (New York: McGraw—Hill, 1965), p. 56. 23A. M. Vener, and C. A. Snyder, "The preschool child's awareness and anticipation of adult sex roles," Sociometpy, XXIX, No. 2 (1966), pp. 159—168. (Abstract). 4H. Levin, and Robert R. Sears, "Identification with parents as a determinant of doll pay aggression," Child De- velopment, XXVII, (1957), p. 158. 5Kagan, Review of Child Development Research I, ed. Hoffman and Hoffman, p. 162. 18 of non-appropriate play objects. Hartley notes that Rabban found consistently increasing sex oriented limitations of toy choices with increase in awareness of own sex, from age five to six years.1 Various Theories of Identification McCandless describes three general classes of identifi- cation theorists. The first are Freudian. Very briefly, this theory states that the boy first identifies with mother; then between the age of five and six identifies with the aggressor or father as a result of fearing him. The next are the Social Learning theorists. These maintain that the boy identifies with the father because of love and reSpect for the parent. They see identification facilitated by a warm nurturant father, and a mother who loves the husband and wishes her son to be like him. The third are the Power theo— rists. This theory is a logical extension of the Freudian and Learning theories, and combines both, saying that the boy identifies with the father because he is both a rewarder and an effective punisher; i.e., because he is powerful.2 Freudian Freud considered the process of identification in three stages. The first occurs when the child is very young and 1Hartley, p. 557, citing Rabban (1950). 2Boyd R. McCandless, Children and Adolescents, (New York: Holt, Rinehart and Winston, 1965), pp. 559-541. 19 incorporates some of his mother's actions because he is un- able to distinguish between himself and the object of his identification, the mother. The second stage is the develop- ment of the specific object choice due to some action of the mother that disappoints the child, and the child subsequently identifies with the father. Freud posited two different mechanisms of identification in this third stage depending on the child's sex. Boys change their primary identification from mother to father as a function of fear of the aggressor, for it is the father who controls and frustrates the child. By this means (identification) the boy seeks to defend himself against the punishing father by identifying with him, thus resolving the Oedipus complex.1 Freud states that the antecedent of the Oedipus complex includes an identification of an affectionate sort with the boy's father, an identification which is still free from any sense of rivalry in regard to his mother.2 He (Freud), also states that identification with the father helps pave the way for the Oedipus complex. He explains the process thus: at the same time as this identification with his father, or a little later, the boy has begun to develop a true object-cathexis towards his mother according to the lAldous and Kell, "A partial test of some theories of identification," Journal of Marrigge and Familpriving, XXIII, No. 1, (1961), p. 15. 2Sigmund Freud, "Some psychological consequences of the anatomical distinction between the sexes," Collected Papers, (London), Hogarth, 1959), V, p. 188. 20 anaclitic type. He then exhibits, therefore, two psychologically distinct ties: a straightforward sex— ual object-cathexis towards his mother and a typical identification towards his father. The two subsist side by side for a time without any mutual influence or interference. In consequence of the irresistible ad- vance towards a unification of mental life they come together at last, and the normal Oedipus complex origi- nates from their confluence. The little boy notices that his father stands in his way with his mother. His identification with his father then takes on a hostile coloring and becomes identical with the wish to replace his father in regard to his mother as well.1 As we have noted, the mother is the first love object. The young boy at about the age of three or four recognizes his father as a dangerous, powerful rival for his mother's affection and attention. The danger arises from the threat of castration which his father and rival may carry out upon him if his incestuous desire for his mother is discovered. At first this threat is not taken too seriously, but when the small boy discovers that little girls have no penis he jumps to the conclusion that they are guilty persons who have been castrated, and therefore the threat of castration is not an empty one. In view of his danger it seems wiser to repress his desire for his mother and to identify with his father to the best of his ability, deferring to maturity those qualities which he is now too immature to assimilate. The more success- ful his repression of his incestuous desire for his mother and his identification with his father, the less he has to fear. The less successful he is, the more he has to fear. 1Sigmund Freud, Group Psychology and the Analysis of the Ego, (London: Hogarth, 1949), pp. 60-61. 21 Thus, fear of the father becomes a driving force and evantual- ly it is incorporated where it becomes the child's superego, or punishing force when the ego ideal is violated.1 Thus the Oedipus complex is resolved by what Anna Freud refers to H2 as "identification with the aggressor and repression of the desire for the mother. In short, "The Oedipus complex suc- cumbs to the threat of castration."3 Social Learning Theory In contrast to the Freudian and Power theories of identi- fication the Social Learning theory has fewer adherents. As McCandless stated, this theory maintains that the boy identifies with the father not, as Freud postulates, due to fear, but rather because of love and respect for the parent. In support of this theory there have been various research studies which have shown a correlation between the child's identification with the parent (via such methods as toy pref- erence tests, or using the ITSC) and the parent's score on a scale of some type measuring warmth, nurturance, and highly rewarding parent characteristics.4 18. M. Stoke, "An inquiry into the concept of identifi— cation," Journal of Genetic Psychology, LXXVI (1950), p. 165. 2Anna Freud, The Ego and the Mechanisms of Defense, (New York: International University Press, 1946), Chapter IX. 3Sigmund Freud, "The passing of the Oedipal complex," Collected Papers, Volume II (London: Hogarth, 1924), p. 275. 4McCandless, p. 541. 22 Power‘Theory The Power theory of identification is an attempt to blend both the Freudian and Social Learning theories. This theory states that when the father is both a rewarder and a punisher identification is likely to be greatest. Levin and Sears support this view when they state that there is a pre- sumption that the child identifies in some degree with all persons who both reward him and place demands on him. They note that this includes all members of the family but pre- eminently the parents.1 This theory also states that the boy whose father rejects him would probably find it difficult to identify with his male parent.2 After stating that both males and females first identify with their mother, as she is their main care taker, Sears, Maccoby and Levin state that boys transfer to a masculine identification due to the follow- ing motivations: (1) direct rewards, (2) identification with the aggressor, (5) the power advantage of the father role, and (4) the greater number of opportunities for practicing the masculine role which are available to male children.3 Other Theories There are other variations of identification theories. Perhaps that of Talcott Parsons is one of the more widely 1Levin and Sears, Child Development, XXVII, p. 158. ZSears, Maccoby and Levin, Patterns of Child Rearing, (Evanston, Illinois: Row Peterson, 1957), p. 586. 31bid., p. 575. 25 mentioned of these. Parsons stated that identification was the development of a "we feeling." He believed that in the Oedipal phase of development a child underwent not one but three new identifications (as opposed to the first identifi- cation with the mother in which he and the mother become one). These are: (1) the internalization of the familial "we category, (2) the internalization of the sibling cate— gory, and (5) the internalization of the same sex category (it is here that the boy identifies with the father and the girl with the mother).1 Parsons viewed the whole process of identification as a dynamic one with variability at each stage of development and influenced by the whole interaction system of the family. Sex role identification was con- sidered to be internalization not of a total personality or of personality traits, but of a reciprocal role relationship that was functional at a particular period in the child's deve10pment.2 Current Research Findings In Sex Role Identification In evaluating the Freudian identification theory Stoke states that there is a serious scientific difficulty in the 1 Talcott Parsons and R. F. Bales, Family Socialization and Interaction, (Glencoe, Illinois: Free Press, 1955), p. 95. 2M. M. Johnson, "Sex role learning in the nuclear family," Child Development, XXXIV, (1965), p. 519—55. 24 problem of proof of the Oedipus complex as a significant factor in the development of an identification.1 Broadbeck points out since there is, also, strong enough identification with the opposite-sex parent even within the youngest age groupings, which would not be predicted from an all-or- none identification process often implied in the Oedipal interpretation, it appears that there are still other factors than sexual frustration leading to parental identification in childhood.2 The data (based on a study of forty adolescents) from his study suggests that there are "multiple and independent de- terminants of identification and that, by and large, the value systems of adolescents are on the whole not patterned in the degree of manner implied by the exclusive and simple use of an Oedipal theory."3 Mussen and Distler, working with thirty-eight white middle-class kindergarten boys in a doll play and story com- pletion situation, attempted to correlate these findings with ratings of parent nurturance, punishment, and power (nurturance plus punishment). They concluded that role theory with its emphasis on both reward and punishment best inte- grated the data they found.4 1Stoke, Journal of Genetic Psychology, LXXVI (1950), pp. 166-167. 2A. J. Broadbeck, "Learning theory and identification: IV Oedipal motivation as a determinant of conscience de- ve10pment," Journal of Genetic Psychology, LXXXIV (1954), p. 225. 3Ibid. 4Paul Mussen and Luther Distler, "Masculinity identifi- cation and father-son relationships," Journal of Abnormal and Social Psychology, LIX, (November, 1959), p. 556. 25 Some other general findings in the area of sex role identification are the following. Rabban, in his often quoted study of 1950, found that: (1) Boys are more clearly aware of sex appropriate behavior than are girls in both middle class and working class groups. (2) Boys and girls of work- ing class groups are earlier and more clearly aware of the sex role pattern than are both boys and girls of the middle class group. This class difference is especially great be- tween girls. (5) Three year old boys and girls of both groups show incomplete recognition of sex differences and as a group are unaware of any appropriateness of sex-typed toy objects. (4) The fourth and fifth years are periods of growth and clarification of sex role for working class boys, while the sixth year is particularly significant for middle class boys. (5) Working class girls accept the sex appro- priate pattern by six years of age, but middle class girls do not fully acquiesce to the definition of appropriate sex patterning even by the eighth year, when all other groups have accepted the social expectations.l Walker, in a study of children's game choices noted that more girls crossed over into masculine scores than vice versa, and these scores were more masculine as compared with the few boys who crossed over into feminine scores, and then with lM. Rabban, "Sex role identification in young children in two diverse social groups," Genetic Psychology Monographs, XLII, (1950), pp- 140-141. 26 only marginally feminine scores. Furthermore, he reported that this type of finding is common to other masculine feminine studies.1 Hartley, in interviewing eight to eleven year old males, noted the following sources of conflict in establishing male roles. (1) lack of adequate male models, (2) extensive supervision by women, (5) multiple conflicting role demands, (4) and lack of a clear, positive definition of the male sex role in socialization. Hartley goes on to elaborate on the problems in establishing masculine identification, stating that more stringent demands are made on boys and at an earlier age when they are least able to understand either the reasons for or the nature of the demands. Those demands are fre- quently enforced harshly and defined as something the child should not do because he will be regarded as a "sissy."2 Boys are more aware of female roles than are girls of male roles. It is suggested that a negative directive plays a greater part in boy's sex role identification, forcing an awareness of Opposite sex role activities for the purpose of avoiding them.3 It has been concluded by others that 1R. N. Walker, "Measuring masculinity and femininity by children's game choices," Child Develgpment, XXXV, No. 4 (1964). PP. 161~171. 2Ruth E. Hartley, "Sex role pressures and the socializa- tion of the male child," Psychological Reports, V, No. 5 (1959), p. 458. 3Ruth E. Hartley and F. Hardesty, "Children's perceptions of sex role activity in childhood," Journal of Genetic Psy— chology, cv (1964), pp. 45—55. 27 masculinity for boys is defined as avoidance of femininity, but just what constitutes masculinity in positive actions, thought patterns, and so forth, seems to be a mystery.1 Some have suggested that the problem of having the masculine role defined in negative (i.e., don't be a sissy) terms rather than in positive (boys do such and such) terms stems from fathers being away from home during most of the young child's wakeful moments. The process of identification is thus available only minimally to boys since their natural identification objects, their fathers, are simply not around much of the time to serve as models. Hartley, in her study of eight to eleven year old males, concluded that boys are thus forced to turn to the peer group and somewhat older boys as guides for Specifics of their behavior as males. But she goes on to note that this amounts to a pooling of impressions and anxieties they derived from early training, since peers have no better source of information. This results, accord- ing to some researchers, in greater rigidity of male role demands.2 One other interesting finding was reported by Kagan and Moss. According to their report, Escalona and Herder (1959) at the Menninger Foundation, attempted to predict the behavior 1Willard W. Hartup, S. Moore, and G. Sager, "Avoidance of inappropriate sex typing in young children," Journal of Consulting Psychology, XXVII (1965), p. 467. 2Hartley, Psychological Rgports, V, No. 5 (1959), pp. 448—449. 28 of preschool children from observations of them during the first year of life. Sex role interests in the five year olds could be estimated from the infancy data. Sex role identification and the pattern of sexual behavior in adult- hood were each related to reasonably analogous behavioral dispositions during the early school years. The period from six to ten years was particularly predictive, and to a lesser degree the period of three to six years.1 As has been mentioned earlier, parent attributes are thought to be of prime importance in the development of the child's sex role identification.2 Of course the child's perception of male and female roles is a determinant of the type of identification he makes. Hartley discovered that from the child's point of view there are no changes from traditional sex role conceptions. He sees only the picture as it appears in his time, and this picture shows remarkably little change from traditional values. The basic home mak- ing duties are the woman's and the money getting ones the man's.8 The variable of parental warmth and affection has gen- erated several studies, the majority of which seem to support ~ 1Kagan and Moss, Birth to Maturity, (New York: Wiley, 1962), pp. 9 and 266. 2Frederick Elkin, The Child and Society:. The Process of Socialization, (New York: Random House, 1960), p. 54. sHartley, "Children's concepts of male and female roles, Merrill-Palmer Quarterly, VI, (1960), p. 91 29 a correlation between the parent's warmth and affection for the child and the degree of the child's identification, especially in the father son dyad. One study showed that fathers high in warmth and affection had sons who scored high in masculinity.1 Another closely related study showed that boys who perceived their fathers as nurturant and power- 2 ful scored more highly masculine. Sears, Maccoby and Levin reported that the mother's warmth and affectionate demonstra- tiveness were slightly related to the amount of dependency, and they expected that these same aspects of maternal be- havior would be related to the child's tendency to practice 3 parental roles. Mussen and Distler reported that the vari- able of the father-son relationship was more directly associ- ated with sex typing than that of the mother-son relationship.4 Another finding somewhat related to this is that the father's discipline and authority is necessary to the boy's sex role identification.5 lP. Mussen and L. Distler, "Child rearing antecedents of masculine identification in kindergarten boys," Child DevelOp- ment, XXXIV, (1965), p. 596. 2Paul Mussen and Eldred Rutherford, "Parent-child rela- tions and parental personality in relation to young children's sex-role preferences," Child DevelOpment, XXXIV, (1965), p. 596. 3Sears, Maccoby and Levin, Patterns of Child Rearing, p. 572. 4Mussen and Distler, Child Development, XXXI (1960, p. 98. 5L. H. Mitchell, "Dominance and femininity as factors in sex role adjustment of parents and children," Dissertation Abstracts, No. 12, Pt. 1 (1966), p. 7440. “ 50 Other variables under study have been father-absent and mother-dominant families. Sears reported that if the father is not present the boy must model his behavior after his mother. This does not mean exclusive feminine sex typ- ing because there are extra-family influences, but at earlier ages sex differences would be less clearly established in children whose fathers were absent than in those whose fathers were in the home.1 Hetherington reported that matern- al dominance disrupts masculine sex role preference in boys.2 The differential imitation of parent and sibling was studied in a simulated situation in which it was shown that the child imitates the adult model who possesses rewarding power (parent) rather than the adult who is awcompetitor (sibling) for rewards. The main implication was that children probably have a greater tendency to imitate parents than siblings. Other aspects of the study were also valuable; one of these showed that when children are exposed to multiple models they may select one or more of them as a primary be- havior source, but rarely reproduce all the elements of a single model's repertoire or confine their imitation to that model. Furthermore, even within the same family, even same 1R. R. Sears, M. H. Pentler, and P. S. Sears, "Effect of father separation on preschool children's doll play aggres- sion," Child DevelOpment, XVII (1946), pp. 219-245. 2E. Marvis Hetherington, "A developmental study of the effects of sex on the dominant parent on sex role preference, identification and imitation in children," Journal of Personality and Social Psychology, II, (1965), pp. 188-194. 51 sex siblings, exhibit quite different response patterns, ow- ing to their having selected for imitation, different elements of their parents' response repertoires.1 Several studies relate both the parent and sibling vari- ables to sex role identification of the child by studying how the presence of siblings and birth order affect the parent- child dyad. Heilbrun and Fromme report that only-child males are more highly identified with their mothers than are males 2 In another study by Heilbrun he states that with siblings. since mothers tend to be more nurturant with first born children than those following (Sears £5.3l. 1957), the well- documented positive relationship between nurturance and identification would lead one to predict a greater maternal identification for first born boys and girls. His study showed that males without siblings were more highly identified with their mothers than all children with siblings combined.3 Fauls and Smith seem to agree in their finding that the child without siblings (whose relations within the family are limited to adults) receives more parental supervision and lA. Bandura, D. Ross, and S. A. Ross, "A Comparative test of the status, envy, social power, and secondary reinforcement theories of identificatory learning," Journal of Abnormal and Social Psychology, LXVII, (1965), pp. 527-554. 2A. B. Heilbrun and D. K. Fromme, "Parental identifica— tion of late adolescents and level of adjustment: the import- ance of parent-model attributes, ordinal position and sex of the child," Journslyof Genetic Psychology, CVII (1965), PP. 49-59. 3A. B. Heilbrun, "The measurement of identification," Child Development, XXXVI, (1965). PP. 124-127. 52 guidance and is parentally directed to adult activities and interests at an earlier age than the child with older sib— lings. Furthermore, the only child's competitive field is limited within the family to adults, and the only child is the sole object of parental aspirations. All these factors are seen as tending to motivate the only child toward early adoption of the adult roles, including the adult sex role, and to result in close agreement between the child's attitude and his parents attitude regarding sexually appropriate be- 1 These researchers found that only children more havior. often choose sexually apprOpriate activities than do children with one or more older like-sex siblings. It may be specu- lated either that older siblings have no influence on the younger child's learning of sexually appropriate behavior, or that the greater permissiveness in the relationship be- tween the parents and the younger child (as compared with the only child) counterbalances the teacher-pupil relationship 2 between older and younger siblings. Sears and others found that there is a maternal tendency to overprotect only children, and that fathers of boys without siblings tend to assume the chief disciplinary control of their sons.3 1L. S. Fauls, and W. D. Smith, "Sex role learning of five year olds," Journal of Genetic Psychology, LXXIX (1956), pp. 106-107. 21bid., p. 115. 3Sears, Maccoby, and Levin, Patterns of Child Rearing, p. 415. 55 Similarities between only and oldest children have been noted by several researchers. Fauls and Smith state that these probable similarities are related to the fact that the oldest child is the only child in the family before the birth of a sibling.1 Koch thinks that the similarity is heightened at wider age spacings (4-6 years) where the boys had been only children for a longer period of time before having siblings. The boys in this particular study who were only or oldest siblings with a wider age spacing were rated 2 Sears and others found rather "sissyish" by their teachers. that only and oldest children had more strongly developed consciences (conscience development being a correlate of identification). They were also more commonly disciplined by the fathers and this may have accounted for some of the rapid conscience development in these boys. Furthermore, when later children were born and the mother required help from her husband, it was the oldest child, rather than the baby, whom she was most likely to turn over to him for brief periods. His relationship with the oldest child, however, was not a particularly nurturant one, and the father tended to be more strict than the mother with this child.3 1Fauls and Smith, Journal of Genetic Psyshology, LXXXIX, p. 107. 2Helen Koch, "The relation of certain formal attributes of siblings to attitudes held toward each other and toward their parents," Society for Research in Child Develgpment Monographs, XXVI, No. 4 (1960), p. 114. 3Sears, Maccoby, and Levin, Patterns of Child Rearing, pp. 417-418. 54 Sears and others, in discussing the child's position in the family as it effects personality deve10pment, note that sex and especially ordinal position place the child in 1 However, in 1950 Sears cautioned a particular social role. that ordinal position is an ecological variable--not a psycho- logical one-~and for that reason could be only a starting point for an inquiry into personality discussion. Ultimately, explanations of an individual's behavior must be driven back to the exact circumstances of his rearing and to the imme- diate stimulational forces acting upon him.2 Some of the findings related to the position of oldest child are as follows: The oldest child is generally allowed to show more aggression toward younger siblings than middle and youngest children. The oldest child's father takes a greater part in his upbringing than with succeeding children. Regardless of family size the mother plays a greater role in caretaking and discipline than does the father. The father has less to do with the youngest child than with other 3 children, regardless of their sex. Fauls and Smith reason that the younger child's learning situation would include not lIbid., p. 419. 2B. G. Rosenberg, and B. Sutton—Smith, "Ordinal posi- tion and sex role identification," Genetic Psyghological Mono- graphs, LXX (1964), p. 299, citing R. R. Sears, "Ordinal posi- tion in the family as a psychological variable," American Sociological Review, XV (1950), pp. 597-401. 3Sears, Maccoby, and Levin, Patterns of Child Rearing, pp. 415—416. 55 only the sample and direction of the parent, but also the example and direction of the older sibling. As a consequence, it would appear that the younger sibling might be expected to learn the appropriate sex role more quickly and accurate- ly than the only child.1 Perhaps one of the more noteworthy aspects of Koch's extensive studies in the area of sibling influences on sex role identification is her careful study of the effect of spacing between siblings and the relative influence of one sibling on another. Her data show that age difference between siblings and the interaction between sibling spacing and ordinal position are significant to sissyness in boys. Koch states that in first-horns there is a decrease in sissyness as the age difference between the child and his sibling in- creases up to four years, but a rise occurs at the four to six year level.2 Thus it can be seen that birth order dif- ferences which obtain at the under two year spacing may not obtain at the wider Spacings. Ordinal position group differ- ences at the close Spacings reflect to a relatively greater extent direct sibling interaction effects. At the wider spac- ings the differences are increasingly expressions of child 1Fauls and Smith, Journal of Genetic Psychology, LXXXIX (1956), p. 107. 2Helen Koch, "Sissyness and tomboyishness in relation to sibling characteristics," Journal of Genetic Psychology, LXXXVIII (1956), p. 252. 56 parent relations conditioned, to be sure, by the fact of the sibling and his relation to the parents.1 At any rate, differences associated with ordinal position are frequently contingent on the sex of the child and the sex of the sib- ling.2 Several studies have found that, in general, boys with girl siblings are less masculine than boys with boy siblings. Rosenberg and Sutton-Smith reaffirm this point. However, they also found that boys with older brothers are more mascu— line than boys with older sisters, or boys with younger brothers.3 Koch's findings on boys with older brothershave shown that the more males above them in the family hierarchy, the 4 Furthermore, an increase greater the male identification. in the spacing between brothers enhances the degree of male identification of the younger brother due possibly to an expanding exposure to boys. The older the older brother is, the more male friends he has. Much scorn of girls and girl's 1Helen Koch, "Some emotional attitudes of the young child in relation to characteristics of siblings," Child Develppment, XXVII (1957), p. 422. 2Helen Koch, "Attitudes of young children toward their peers as related to certain characteristics of their siblings, Psychological Monographs, LXX, No. 19 (1956), p. 40. 3Rosenberg and Sutton-Smith, Genetic Psycholpgical Mono— graphs, LXX (1964), pp. 297-528. 4Koch, Society for Research in Child Development Mono- graphs, XXVI, No. 4 (1960), p. 107. 57 activities is likely to be expressed. Little brother has 1 Koch has summarized some "big ears" for such comments. other possible effects for a boy having an older brother; these are that (1) the younger brother doubtless has less sex-identification conflict than a boy with an older sister, and (2) the child's identification with his sibling is prob— ably greatest at the close Spacing in the sense of feeling like the sibling, but his desire to have the sibling's assets and advantages may increase with Spacing.2 Koch reports that boys with a younger brother spaced 2-4 years apart were more father allied, and rated more masculine than were those with spacing less than two years.3 She also stated that the boy with a younger brother might have some sexual conflict but in the main this would not probably be very great. However, his conflict might be greater than that of a boy with an older brother because the boy with an older brother has a male sibling from the first to reinforce his identification. It is possible that the closer the siblings are in age, the less conflict the boy with a younger brother has in establishing a male sex role identification. Furthermore, the child with a younger sib- ling has a period of time when his sole stimulation and 1Koch, Journal of Genetic Psychology, LXXXVIII (1956), p. 259. 2Koch, Psychological Monographs, LXX, No. 19 (1956), p. 29. 3Koch, Sociepy For Research in Child DevelOpment Mono— graphs, (1960), p. 106. 58 instruction is by adults; the wider the spacing, the longer the period of adult influence. The amount of identification decreases as the spacing between a boy and his younger brother increases. At any rate, the identification is not so great as when the male sibling is older and commands more respect. Another point to remember is that the boy with a younger brother is the prime chooser of playmates at all Spacings, in contrast to the boy with the older brother.1 In comparing the sex of sibling with the degree of mascu- linity, Koch found that boys with sisters were less masculine 2 Sutton-Smith and Rosenberg made than boys with brothers. further suggestions that siblings reinforce their own sex traits in the opposite sibling, irrespective of that sibling's sex, and that older siblings, because of their greater power, have a greater reinforcement value for younger siblings than the reverse.3 Furthermore, Brown stated that his findings suggested the possibility of some degree of feminization in the case of boys who have only sisters as siblings.4 lKoch, Psychological Monographs, (1956), p. 106. 2Koch, Psychological Monogrsphs, LXX, No. 19(1956), p. 41. See also Rosenberg and Sutton-Smith, Genetic Psycholggi- cal Monographs, LXX (1964), pp. 297-528. 3Sutton-Smith and Rosenberg, "Age changes in the effects of ordinal position on sex role identification," Journal of Genetic Psychology, CVII (1965), pp. 66-68. 4Brown, Psychological Monographs, LXX, No. 14 (1956). p. 17. 59 Koch has published the following findings relative to boys with younger sisters, compared with boys with younger brothers. Boys with younger sisters less frequently prefer male playmates, and less frequently name a male as best friend. They more frequently express equal enjoyment in playmates of both sexes.1 Boys with younger sisters are in the driver's seat socially, that is, it is they who bring home friends, hence boys with younger sisters have a more marked male identification and interests than boys with older sisters.2 Nonetheless, boys with younger sisters have much sex role identification conflict. At the wider spacings the boy is more stimulated and instructed by adults. The identi- fication with the sibling and the amount of association with the sibling and her friends doubtless decreases more rapidly with spacing than if the sibling is a younger brother.3 Boys with older sisters doubtless have a strong sex identification conflict since the mother and sibling both are opposite in sex to the child. This conflict may increase with spacing at least up to the two to four year difference. The model the child has is strongly feminine and it is not unlikely that the father favors the sister. The boy with an 1Koch, Child Development, XXVI (1960), p. 114. 2Koch, Journal of Genetic Psychology, LXXXVIII (1956), p. 259. 3Koch, Psyshological Monogrsphs, LXX, No. 19 (1956), p. 55. 40 older sister may be more indulged and his sibling and mother may serve and protect him more. This may tend to unfit him for an effective role with his peers. There is probably a sharp reduction in identification with or conscious desire to be like the sibling with an increase in Spacing.1 Koch has reported other findings related to sex role identifica— tion of boys with older sisters. Compared with boys with older brothers, boys with older sisters more frequently had a best friend who was female, and more often expressed preference for girl playmates and preference for older play- mates than did the boys with younger sisters. These boys were also judged more sissyish than boys with brothers.2 If the older sibling is a girl but near in age, the boy's contact with her and her girl friends will be extensive. The sister, being older, will probably bring more friends into the home than the boy does and the former's friends will be chiefly girls. Thus a boy with an older sister is more apt to have girl playmates. When the sister is more than two years older, however, the two children in the family each tend to have their own group of friends, because the sex distance tends to be great in elementary school years. Eighty per cent of the boys with sisters two to four years older said they preferred male playmates (even though both 11bid., p. 29. 2Koch, Child Development, XXVI (1960), p. 115. 41 groups were judged very masculine).1 Koch also reports that at Spacings up to four years second-born boys with sisters (i.e., boys with older sisters) were more sissyish than were first-born boys with sisters; whereas, when the distance between siblings was more than four years the relationship between second-borns was reversed.2 1Koch, Journal of Genetic Psychology, LXXXVIII (1956), p. 255. 21bid., p. 245. CHAPTER III PROCEDURE CHAPTER I I I PROCEDURE Description of the Instrument The It Scale for Children (ITSC) was developed by Daniel G. Brown, presently a consultant in Mental Health with the public health service in Atlanta, Georgia. The ITSC is fully described in a noncommercial American Psychological Association monograph which serves as the manual.1 The ITSC is composed of 56 picture cards, 5 by 4 inches in size. The main character of the test is "It" (see Appendix, p. 66), a presumably sexless stick figure drawing. Supposedly, the subject identifies with It, and it is It rather than the child who "takes the test." Thus Boyd McCandless has labled "2 The most feminine pos- the test a "structured projective. sible score It can earn is zero, the most masculine 84. This score is based on three subtests: (a) The first subsection: 8 points for choosing all masculine toys 8 of which are masculine--a jack knife, car, train, dump truck, earth digger, 1D. G. Brown, Sex Role Preference in Young Children, ("Psychological Monographs," Vol. LXX, no. 14; The American Psychological Association, Inc., 1956). 2Boyd R. McCandless, "It Scale for Children," Sixth Mental Measurement Yearbook, ed. 0. K. Burros, (Highland Park, New Jersey: The Gryphon Press, 1965), p. 150. 45 44 gun, toy soldiers, tractor; and 8 of which are feminine-- baby buggy, doll, cradle, purse, necklace, tea set, high chair, baby's washing or bathing stand. Zero is scored for 8 feminine toy choices. (b) The second subsection: 64 points for totally masculine choices for 8 pairs of pictures: Indian chief and Indian squaw; boy's clothing (slacks and shirt) and girls clothing (dress); things with which to make an airplane, and simple sewing articles; shaving articles and ladies' cosmetics; tools (screwdriver, pliers, wrench) and washing machine, iron and ironing board; mens' shoes and ladies' shoes; boys playing on swings and girls playing on swings; tools (saw, hammer, T-square, nails), and rolling pin, muffin tin, and sifter. The child receives 8 points for each masculine choice made and zero points for each feminine choice made. The child is asked, for instance, "Which Indian would It rather be?" (c) The third subsection: 12 points (completely masculine) if It's preference is for the picture of a boyish boy over the pictures of a girlish boy (8 points) a boyish girl (4 points) and a girlish girl (0 points). Marked deviations, plus or minus, from 42 are interpreted as indicative of M or F, respectively. This measure is within the attention span of children, taking 8-10 minutes to administer individually to each child. It can be given (and was in this study) in a single testing session. 45 Reliability Brown reports that reliability of the ITSC was determined by the test-retest method (interval approximately one month). The sample included 78 male and 68 female kindergarteners from Denver, aged 5-4 to 6-4 years who came predominantly from the middle class. For boys the coefficient is .71, and for girls .64.1 Validity The ITSC manual reports item validity, but neglects statistical workup on the power and attractiveness (aside from their differential sex pull) of items or subsections, and does not report the adequacy of the assignment of subtest and sub- section weights.2 The Study Description of Subjects The subjects for this study were boys, ages 4.0 to 5.6 years. There were ten boys with older brothers only, ten boys with younger brothers only, ten boys with younger sisters only, ten boys with older sisters only, and twenty-five boys without siblings. The total number of boys participating in 1Brown, Psychological Monpgraphs, (1956), p. 6. 2McCandless, Sixth Mental Measurement Yearbook, p. 150. 46 the study was sixty-five. All boys who participated were from families in which both own father and own mother resided together with the child. Forty-one boys were from the Lansing-East Lansing area of Michigan, three from the Berrien Springs area of Michigan, and twenty-one from San Diego, California. Twenty-one boys were in attendance at all day nursery schools, twenty-eight attended half-day co-operative nursery schools and thirteen boys attended either the Spartan Co-Operative nursery or the Laboratory Pre-School at Michigan State University. Three other boys (who were used in the study) were not in attendance at any nursery school. Establishment of Rapport The majority of boys participating in the study (62 out of 65) were in some type of nursery school setting. The examiner would generally plan to visit the nursery at least one day before beginning the testing sessions. This first day was Spent in gaining necessary information and permission to include each subject in the study. Also on the first day the examiner met each child, sometimes played with him, had juice with him, and explained to him that at a future time the examiner would return with a very Special game to play. If the test room had been determined the examiner and subject took a walk to see the room. 47 TestingyConditions TABLE 1 CLASSIFICATION OF TESTING CONDITIONS no other persons present child Sized table and chair no other persons present no child sized table and chair Name of No. nursery tested California: Baptist Christopher FleurLiS Franklin Highland Lakeside LittleOne Redwood Storybook >eto >< no child Sized table and other persons present chair 48 Presentation of the Test The child, on being brought to the testing Situation, was told that he was going to play a game with the examiner. After putting the child at ease (by saying something like: have you ever played games with cards before? That's what we are going to do. We're going to play with these cards). The plastic box containing the pictures used in the ITSC was then brought out. The test was administered according to the directions given on page five of Brown's monograph.1 Introduction: We are going to play a little game with this child here. See this child? Let's call this child It. O.K.? So this game will be about It. Here, you hold It. Now, we're going to Show this child whose name is It, some cards with pictures on them. (If child asks about the sex of It say: "It's just a child, isn't it? Let's just say It's a child.")2 The last section has the four child-figures. The pictures were placed in rows of four in front of the child (all Sixteen cards were presented in the first section at the same time). In the second section two cards at a time were placed in front of the child. In the last section the four cards were placed on the table in front of the child. In every case the child was asked to indicate "It's" choice by placing "It" on the object, or person it would choose. During the testing there was nothing else on the small table in front of the child except the cards used in the test. After playing the lBrown, Psychological Monographs, (1956), p. 5. 21bid. 49 game the child was told, that's all, thank you for playing with me. The information on whether the child lived with both parents was obtained through the nursery school teacher, except in the cases where boys were tested in their own homes. In the later case this information was known by the examiner due to personal acquaintance with the family. The information on family position and sex of Siblings was ob- tained from the parents when the examiner contacted the parent to obtain permission to use the subject in the experi— ment. The date of birth was also verified at this time. A score sheet is included in the Appendix. CHAPTER IV DATA ANALYSIS CHAPTER IV DATA ANALYSIS Procedure AS is the case in many studies in the social sciences, a random sample for a study of this type would have been most difficult to obtain. Since the random sample is one of the requirements for using parametric tests, nonparametric tests were chosen. To test hypotheses 1-4 (see pages 8-9) a test developed by A. R. Jonckheerel was originally chosen. The Jonckheere test determines whether the differences among the samples signify genuine population differences, and also tests the order of the expected differences. However, after examining the data and discovering that the order of the means was not in the direction hypothesized, the Kruskal- Wallis One Way Analysis of Variance by Ranks2 was used to determine whether the four samples were from different pOpu- lations. 1A. R. Jonckheere, "A distribution-free k-Sample test against ordered alternatives," Biometrika, XLI, (1954), pp. 155-145. 2Sidney Siegel, Nonparametric Statistics For The Behavorial Sciences, (New York: McGraw-Hill, 1956), pp. 185-194. 51 52 For hypothesis 5 (see pages 8-9) the Median Test1 was chosen to determine whether the two independent groups (Group I boys with siblings and Group II boys without sib- lings) had been drawn from the same population (that is, were their scores Significantly different). Findings Hypotheses 1-4 The Kruskal-Wallis test was applied. The value for X2 was 2.04; that is, it did not approach significance. Thus hypotheses 1-4 were not supported; the means were not in the direction hypothesized, and the samples were not from dif- ferent populations. The means for samples 1-4 are given in Table 2. The ranks for each sample are given in Table 5, and the range of scores for each sample are given in Table 4. TABLE 2 MEANS OF SAMPLES 1-4 prs having Older Younger Younger Older brothers brothers sisters sisters o_i1) (2) (5) (4) 59.41 64.00 64.20 65.40 lIbid., p. 111. 55 TABLE 5 RANKS OF SAMPLES 1-4 Boys havipg Older Younger Younger Older brothers brothers sisters Sisters (1) (2) (5) (4) 2.5 1.0 6.0 4.0 7.0 2.5 10.5 5.0 8.0 12.0 10.5 19.5 9.0 17.5 15.5 21.0 15.5 25.0 15.5 25.0 15.5 29.0 17.5 26.5 19.5 51.0 22.0 26.5 24.0 55.0 29.0 29.0 55.5 56.0 51.0 56.0 58.5 58.5 56.0 40.0 TABLE 4 RANGE OF SAMPLES 1-4 Boys having Older Younger Younger Older brothers brothers Sisters Sisters (1) (2) (5) (4) 29-85 28-85 48-82 50-84 54 HypOtheSis 5 The Median test was then applied to the data. (The Median test is the nonparametric analog of the parametric "t" test for differences between means). Thus hypothesis 5 (that boys without Siblings will have scores significantly different from boys with siblings in sex role preference as measured by the ITSC) was not supported. Table 5 Shows the median for the combined scores of boys having Siblings and boys without Siblings. Table 6 Shows the range in scores for boys having siblings and boys without Siblings. TABLE 5 MEDIAN SCORE FOR COMBINED GROUP OF BOYS HAVING SIBLINGS AND BOYS WITHOUT SIBLINGS __—-_—_—_-__—————__ Boys with Boys without siblings siblipgs No. of scores above 65.5 12 14.5 No. of scores below 65.5 15 10.5 TABLE 6 RANGE OF SCORES FOR BOYS HAVING SIBLINGS AND BOYS WITHOUT SIBLINGS Boys with Siblings » Boys without Siblings 28-84 51-85 55 In summary, hypotheses 1-4 tested by the Kruskal-Wallis One Way Analysis of Variance, and Hypothesis 5, tested by the Median Test were not supported by the statistical analy- sis. The possible reasons for and implications of these results will be discussed in Chapter V. CHAPTER V RESULTS, IMPLICATIONS, RECOMMENDATIONS, SUMMARY CHAPTER V RESULTS, IMPLICATIONS, RECOMMENDATIONS, SUMMARY Results The hypotheses for this experiment are: 1. Boys with older brothers will have the highest mascu- line score on sex role preference as measured by the ITSC. 2. Boys with younger brothers will have the next high- est masculine score on sex role preference as measured by the ITSC. 5. Boys with younger sisters will have the next lowest masculine score on sex role preference as measured by the ITSC. 4. Boys with older sisters will have the lowest mascu- line score on sex role preference as measured by the ITSC. 5. Boys without siblings will have a mean score sig- nificantly different from boys with siblings in sex role preference as measured by the ITSC. None of these hypotheses were supported (see Chapter IV). Implications The possible reasons that the hypotheses were not sup- ported can be divided into three main groups. First it is 57 58 possible that the hypotheses are actually incorrect; that is, that the presence, sex, and position of Siblings does not Significantly effect sex role preference in preschool boys. The second is that the sample for this experiment was not representative. The third is that the ITSC is not valid. The first possibility--that the hypotheses are actually incorrect must be viewed in light of other research to the contrary.1 Koch (see Chapter 2, pages 56-41) has produced the most abundant data in this area. She has noted that dif- ferences in “sissiness” and tomboyishness" correlated with birth order differences which obtain at the under two year spacing may not obtain at the wider Spacings.2 ’However, in the present study the number of years of Spacing was not controlled nor was the number of older or younger Siblings controlled. Boys were included as long as all the Siblings were the same sex and either all older, or all younger than the subject. It should also be noted that Koch relied on teacher ratings to establish the degree of masculinity or feminity. This is a very different instrument than a struc- tured projective such as the ITSC. Furthermore, much of what Koch has published in this area was published in the mid 1See pages 54-41 for a review of literature relating to the hypotheses. 2Helen Koch, "Sissyness and tomboyishness in relation to sibling characteristics," Journal of Genetic Psychology, LXXXVIII (1956), p. 252. 59 1950's and one article in 1960. Looking at the current 1968 population including "hippies" and others, one realized that at least for some young adults, the definition of masculine and feminine roles is rapidly changing. To what extent this has penetrated to influence preschool boys' toy choices does not appear answerable at the present time. However, I would tend to think that the hypotheses are true as stated; but due to the complexity of variables inter- vening in the development of sex role preference (such factors as parental warmth, and affection, for instance, have been found to have a high correlation with sex role preferencel), the specific variables of presence, sex and position of Sib- lings are difficult to isolate in a statistical correlating sense. The second possibility—-that the sample for this experi- ment was not representative should also be considered. Look- ing at the wide range of scores (see Table 4 and Table 6, Chapter IV), one sees several very low (or feminine) scores around 28-51, as well as many very high (or masculine) scores in the 80's. The high scores would be expected, but one wonders about the frequency of low scores in all groups (except boys with younger Sisters). It Should be noted, how- ever that Brown reports a Similar frequency of very low lP. Mussen, and L. Distler, "Child rearing antecedents of masculine identification in kindergarten boys," Child Development, XXXI (1960), p. 98. 60 scores with an increasing rate of very high scores.1 It should also be noted that many (16 out 65) of the subjects were from known student families. It is possible that other boys were also from student families. This is mentioned because frequently in student families there is a greater Sharing of work roles around the home. That is, the mascu- line and feminine roles may be less precisely divided. Mother may be gone from home as much as father. Father may participate in such traditionally feminine tasks as food preparation, child care, and so on, as much as mother. The last question concerns the test itself. The valid- ity was established with a rather small sample (for a norma- tive study) with 78 male and 68 female kindergarteners from Denver aged 5.4:6.4 who came predominantly from the middle class. In developing the ITSC Brown states "it was assumed that toy objects commonly associated with boys and those commonly associated with girls constitute one source of dif- "2 ference in sex-role preference. Brown gives an item analy- sis of percentage of boys and percentage of girls making 1D. G. Brown, Psychological Monographs, LXX, no. 14 (1956), p. 8. The subjects in Brown's study were 5.4 to 6.4 years old. However through personal correspondence (see Appendix) Brown suggested that an age range between 4.0 and 5.6 would be preferable because older children tend to "see through" the purpose of the ITSC. However, in this study children under 4.5 made considerably lower scores than those 4.6 and above. Perhaps the less mature subjects had diffi- culty identifying some objects. 2Brown, ibid., p. 10. 61 differential choices for each item. All but 4 of these items discriminated beyond the .05 level, when these tests were run prior to 1956. Whether they discriminate today in the same degree (e.g., do 74% of the boys still prefer building tools, and 69% of the girls still prefer baking articles), is another unanswerable question. In this particular study there were more feminine object choices than Brown reported in 1956. However, more boys said "It" would prefer to be a boy than any of the other child figures. From this, one might suspect that although there may be a less precise definition of what objects constitute masculine interests, boys have just as strong a desire to be masculine (see Table 7). McCandless in his review of the ITSC stated that find- ings from its use, suggest that progression with age is clear for boys, 'full masculinity' apparently having been 'gained' by early school ages. This con- clusion is based on several United States populations. . . . Findings are less clear for United States girls, the theory adduced being that in our culture the feminine role is both less desirable and less clearly modeled, and thus later and more reluctantly assumed. The logical flaw here is that there seems to be a tend- ency for United States girls, even at very early ages, to identify It as a boy and thus, presumably, to re- Spond in terms of cultural expectations rather than projectively.1 The other problem McCandless touched Upon is that of the subjects perception of "It"° The test is based on the lBoyd McCandless, "It scale for children," Sixth Mental Measurement Yearbook, ed. 0. K. Burros, (Highland Park, New Jersey: The Gryphon Press, 1965), p. 151. 62 TABLE 7 PERCENTAGES OF BOYS OBJECT CHOICES ON THE IT SCALE FOR CHILDREN. A COMPARISON OF CHOICES MADE BY BOYS IN NEWMAN'S STUDY (1968) AND BROWN'S STUDY1 (1956) Newman's Studyy Brown's Study, Male Female Male Female choice choice choice choice 9’6 5% Indian Princess Indian Chief 78 22 86 14 Trousers and Shirt Dress 85 17 77 25 Sewing materials Airplane parts 92 8 88 12 Cosmetic articles Shaving articles 72 28 91 9 Mechanical tools Household objects 71 29 82 18 Men's shoes Women's Shoes 71 29 76 24 Girls playing Boys playing 62 58 71 29 Building tools Baking articles 65 55 74 26 Totals: eight paired items 74 26 81 19 Child figure Girl Girlish Boy Boyish Girl Boy Newman's Study 15 5 11 71 Brown's Study 12 9 10 69 1D. G. Brown, Psychological Monogrsphs, LXX, no. 14, (1956). pp. 15_14. 65 assumption that the "testee identifies with "It" and it is "It" rather than the child who 'takes the test'."1 Thus the perception of It is crucial to test results. In the present study several boys made such comments as "I choose the truck" when using "It", thus revealing that they personally identified with "It" and "It's" choices. Others said such things as "Oh, he'd choose the pants, boys don't wear dressesi" (when the choice between a dress, or slacks and a Shirt was made). Recommendations Possibilities for further research include repeating this study using a larger, and ideally a random sample. A sample which was controlled for age spacing of siblings, and number of siblings might also prove profitable. A com- parative study using several instruments on the same subjects, or on matched subjects might also prove worthwhile. If a sufficiently large study were undertaken, it would be well to control for other intervening factors such as parental warmth and affection. Summary The hypotheses that presence, sex and position of sib- lings is related to sex role preference of pre—school boys 11bid., p. 150. 64 was not supported. The possible reasons that the hypotheses were not supported were classified into three groups, first, that the hypotheses were actually incorrect, second, that the sample was not representative, and third that the ITSC is an invalid instrument. The first, that the hypotheses are incorrect, seems unlikely. Rather due to other intervening variables, and a test which may not be sufficiently precise to detect possible slight differences, the hypotheses were not supported. The second, that the sample was not repre- sentative, seems more likely. The many student families and young age of subjects are factors to consider here. The possibility that the ITSC is invalid was discussed. Also it is possible that the test items do not discriminate to the same degree in 1968 as in 1956. The possibilities for further research were reviewed. Finally, although the variables of presence, sex, and position of siblings did not seem to appreciably effect dif- ferences in sex role preference in this experiment, it would seem that identification of significant variables in masculine feminine role preference iS necessary. Furthermore it seems that identification of significant variables will be valuable not only in the basic sense of adding insight into develop- mental psychology, but eventually prove a useful clinical tool as well. REFERENCES REFERENCES Articles and Periodicals Aldous and Kell, "A partial test of some theories of identi- fication," Journal of Marriage and Family Living, XXIII, No. 1 (1961), 15-19. Angrilli, A. F., "The psycho—sexual identification of pre- school boys," Journal of Genetic Psychology, XCVII (1960), 529-540. Bandura, A., Ross, D. and Ross, S. A., "A comparative test of the status envy, and social power, and secondary rein- forcement theories of identificatory learning," Journal of Abnormal and Social Psychology, LXVII, (1965), 527-554. Bell, Richard 0., "Developmental psychology," in Annual Review of Psychology, XVI, (edited by Paul R. Farnsworth), Palo Alto, California: Annual Reviewers, Inc., (1965), 24. Bieliaukas, V. J., "Recent advances in the psychology of masculinity and femininity," Journal of Psychology, LX (1965), 255-265. Broadbeck, A. J., "Learning theory and identification: IV Oedipal motivation as a determinant of conscience de- velopment, Journal of Genetic Psyghology, LXXXIV (1954), 219-227. “ Brown, D. G., "Masculinity-femininity development in children," Journal of Consulting Psychology, XXI (July, 1957), 197-202. . "Sex-role development in a changing culture," Psychological Bulletin, LV (July, 1958), 252-242. . "Sex-role preference in young children," Psychological Monographs, LXX, No. 14 (1956), 1-19. . "Sex—role preference in children: methodological problems," Psycholpgical Reports, XI, (October, 1962), 477-478. 66 67 Brown, D. G., and Lynn, D. B. "Human sexual deve10pment: an outline of components and concepts," Journal of Marriage and The Family, Vol. XXVIII, No. 2 (1966), 155—161. Fauls, L. B., and Smith, W. D., "Sex role learning of five year olds," Journal of Genetic Psychology, LXXXIX (1956), 105-117. Freud, Sigmund, "Some psychological consequences of the anatomical distinction between the sexes," in Collected Papers, Vol. V, London: Hogarth, (1959), 186-197. . "The passing of the Oedipal Complex," in Collected Papers, Vol. II, London: Hogarth Press, (1924), 269—276. Hartley, Ruth B., "A developmental view of female sex-role identification," in Role Theory: Concepts and Research, (edited by Bruce Biddle and Edwin Thomas), New York: Wiley, (1966). . "Children's concepts of male and female roles," Merrill-Palmer Quarterly, VI (1960), 85-91. . "Sex role pressures and the socialization of the male child,“ Psychological Reports, V, No. 5 (1959), 457-468. , and Hardesty, F., "Children's perceptions of sex role activity in childhood," Journal of Genetic Psy- chology, CV (1964), 45-51. Hartup, W. W., Morre, S. and Sager, G., "Avoidance of inap- propriate sex typing in young children,“I Journal of Consulting Psychology, XXVII (1965), 467—475. , and Zook, E. A., "Sex role preference in three and four year old children," Journal of Consulting Psychology, Hetherington, E. M., "A developmental study of the effects of sex of the dominant parent on sex role preference, identification and imitation in children," Journal of Personality and Social Psychology, II (1965), 188-194. Heilbrun, A. B., "The measurement of identification," Child Development, XXVI (1965), 111-127. , and Fromme, D. K., "Parental identification of late adolescents and level of adjustment: the importance of parent-model attributes, ordinal position, and sex of the child," Journal of Genetic Psychology, CVII (1965), 49-51. 68 Jonckheere, A. R., "A distribution-free K—sample test against ordered alternatives,“ Biometrika, XLI, (1954), 155-145. Kagan, Jerome, “Acquisition and significance of sex typing and sex identity," Review of Child Development Research, (edited by L. W. Hoffman and M. L. Hoffman), New York: Russell Sage Foundation, (1964), 157-176. Koch, H. L., "Attitudes of young children toward their peers as related to certain characteristics of their siblings," Psychological Monographs, LXX, No. 19, (1956), 1-41. "Sissyness and tomboyishness in relation to sibling characteristics," Journal of Genetic Psychology, LXXXVIII (1956), 251-245. . ”Some emotional attitudes of the young child in relation to characteristics of siblings," Child Develop- ment, XXVII (1957), 595-426. . "The relation of certain formal attributes of Siblings to attitudes held toward each other and toward their parents," Society For Research in Child DevelOpment Monographs, XXVI, No. 4 (1960), 1-124. Landreth, C., "Four year old's notions about sex appropriate— ness, of paternal care and companionship activities," Merrill-Palmer Quarterly, IX, (1965), 175-182. Lansky, L. M., and McKay, G., "Sex role preferences of kinder- garten boys and girls: some contradictory results," Psychological Reports, XIII, (1965), 415-421. Lazowick, L. M., "On the nature of identification," Journal of Abnormal and Social Psyshology, LI (1955), 175-185. Levin, H., and Sears, R. R., "Identification with parents as a determinant of doll play aggression," Child Development, XXVII, (1957), 155-155. Maccoby, Eleanor, "Role taking in childhood and its conse- quences for social learning " Child Develppment, XXX (1959), 259-252. Minuchin, Pat, "Sex-role concepts and sex typing in childhood as a function of school and home environments," Child Development, XXXVI, No. 4, (1965), 1055-1048. Mitchell, L. H., "Dominance and femininity as factors in sex role adjustment of parents and children," Dissertation Abstracts, No. 12, Pt. 1 (1966), 7440. 69 Mussen, P., and Distler, L., "Child rearing antecedents of masculine identification in kindergarten boys," Child Development, XXXI, (1960), 89—100. , and , "Masculinity identification and father-son relationships." Journal of Abnormal and Social Psychology, LIX (November, 1959), 550-556. , and Rutherford, Eldred, "Parent-child relations and parental personality in relation to young children's sex-role preferences,“ Child Development, XXXIV (1965), 589-607. Rabban, M., "Sex role identification in young children in two diverse social groups," Genetic Psychological Mono- graphs, XLII, (1950), 81-158. Rosenberg, B. G., and Sutton-Smith, B., "A revised conception of masculine feminine differences in play activities," Journal of Genetic Psychology, XCVI, (1960), 165-170. , and , "Ordinal position and sex-role identification," Genetic Psychological Monogrgphs, LXX, (1964), 297-528. Sears, R. R., and Pentler, M. H., and Sears, P. 8., "Effect of father separation on pre-school children's doll play aggression," Child Develppment, XVII, (1946), 219-245. Stoke, 8. M., ”An inquiry into the concept of identification," Journal of Genetic Psychology, LXXVI, (1950), 165-189. Sutton-Smith, B., and Rosenberg, B. G., ”Age changes in the effects of ordinal position on sex role identification," Journal of Genetic Psychology, CVII, (1965), 61-75. Vener, A. M., and Snyder, C. A., "The pre-school child's awareness and anticipation of adult sex roles," Sociometry, XXIX, No. 2, (1966), 159-168. Walker, R. N., "Measuring masculinity and femininity by children's game choices," Child Development, XXXV, No. 5, (1964), 961-971- Books Burros, A. K., The Sixth Mental Measurement Yearbook, Highland Park, New Jersey: The Gryphon Press, 1965. Elkin, Fred, The Child and Society: The Process of Sociali- zation, New York: Random House, 1960. 70 Freud, Anna, The Ego and the Mechanism of Defense, New York: International University Press, 1946. Freud, Sigmund, Group Psychology and the Analysis of the Ego, London: Hogarth Press, 1949. Kagan, J. and Moss, Birth to Maturity, New York: Wiley, 1962. McCandless, Boyd R., Children and Adolescents, (New York: Holt, Rinehart and Winston, 1965), 559-541. Mead, Margaret, Male and Female, New York: William Morrow and Company, 1949. Parsons, T., and Bales, R. F., Fgmily Socialization and lpteraction Process, Glencoe, Illinois: Free Press, 1955, 55-151. Sears, Robert R., and Maccoby, Eleanor, and Levin, Harry, Patterns of Child Rearing, Evanston, Illinois: Row, Peterson and Co., 1957. Siegel, Sidney, Nonparametric Statistics for the Behavioral Sciences, New York: McGraw-Hill Book Co., Inc., 1956. APPENDIX 72 ITSC Information Sheet Testing Date Name Birth date School “ Test Age Lives with both parents Yes No Test Score 0' Brothers Birth date Sisters Birth date 75 II It" PUBLIC HEALTH SERVICE 74 DEPARTMENT OF HEALTH. EDUCATION. AND WETI-‘A’as‘ '." ‘ REGIONAL OFFICE ' Room 404 - 50 Seventh Street, N. E. Atlanta, Georgia 30323 April 3, 1967 Miss Phyllis Fehlmann Newman 602 North Pennsylvania Lansing, Michigan 48912 Dear Miss Newman: I think your prOposed study is an interesting one and quite A worthwhile. Enclosed are several reprints that you might like to look over. A There are two considerations that I would suggest: I) if you can do so, I think a better age range would be between 4.0 and 5.0 or 4.6 and 5.6 years; the IT Scale seems to be more sensitive and useful at these pre-school ages; older children tend to "see through"*the purpose, etc., of the Scale; and 2) while your emphasis on sibling status and birth order is important, the parent status and relation- ship is probably crucial as far as sex role development is concerned; in this connection is there some way you could determine the family structure, i.e., one or both parents. or foster parents? Preference of child for one parent or ~the other? Father-absent families? etc. I would be interested to know about your study after you have carried it out. Please let me know'if I can help in any other way. Sincerely, 7’7 Q ' 'f' y M "' r“‘~—"\4(/tt,t£.p ‘ - 4")“ . Daniel G. Brown, Ph.D. Consultant in Mental Health HICHIGQN STQTE UNIV. LIBRARIE