Ll lllllllfilflllsllllfllljfllcllflflll winging» ' L This is to certify that the thesis entitled ASPECTS OF THE FATHER-CHILD RELATIONSHIP AND FATHER-AVAILABILITY DURING CHILDHOOD AND SEX-ROLE DEVELOPMENT AMONG COLLEGE STUDENTS FROM INTACT FAMILIES presented by Edward Thomas Manzitti has been accepted towards fulfillment of the requirements for Ph . D . degree in Educational Psy . r7<:4,///1Ws \anmmawr M' ’/ ,,,. Date may/J U, /f/(/ 0-7639 OVERDUE FINES ARE 25¢ PER DAY PER ITEM Return to book drop to remove this checkout from your record. _---- 0155 i fzfio‘i ASPECTS OF THE FATHER-CHILD RELATIONSHIP AND FATHER AVAILABILITY DURING CHILDHOOD AND SEX-ROLE DEVELOPMENT AMONG COLLEGE STUDENTS FROM INTACT FAMILIES BY Edward Thomas Manzitti A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Counseling, Personnel Services, and Educational Psychology 1979 ABSTRACT ASPECTS OF THE FATHERFCHILD RELATIONSHIP AND FATHERPAVAILABILITY DURING CHILDHOOD AND SEX-ROLE DEVELOPMENT AMONG COLLEGE STUDENTS FROM INTACT FAMILIES BY Edward Thomas Manzitti Six hypotheses were developed to test the effects of paternal nurturance and father-availability during child- hood upon sex-role development among college students from intact families. First, college males having highly nur- turant fathers during childhood were expected to have higher Masculinity scores than were college males with less nurturant fathers during childhood. Similarly, college females with highly nurturant fathers during childhood were expected to have higher Femdninity scores than were college females with less nurturant fathers during childhood. Third, college males with highly nurturant and highly avail- able fathers during childhood were expected to have higher Masculinity scores than were college males that had fathers with other combinations of nurturance and availability during childhood. Moreover, college females with highly nurturant and highly available fathers during childhood were expected to have higher Femininity scores than were college females that had fathers with other combinations Edward Thomas Manzitti of nurturance and availability during childhood. Fifth, college males having less nurturant and highly available fathers during childhood were expected to have the lowest Masculinity scores among college males. Finally, college females having less nurturant and highly available fathers during childhood were expected to have the lowest Femininity scores among college females. The subjects were sixty-eight males and seventy-two females that ranged from seventeen to twenty-three years of age and all of them were undergraduates at a large mid- western university. In order to control for a variety of variables that might affect the results, all of the par- ticipants were native American, White, first-born or only- children, middle or upper middle class, and resided with both of their natural parents from birth until age sixteen. A retrospective method was used to ascertain the subjects' perceptions of their fathers' behavior and avail- ability toward them as children and the instruments were several questionnaires. Siegelman and Roe's (1978) Parent- Child Relations Questionnaire II was used to assess the father-child relationship during childhood and the Love- Reject Factor score was used to measure the subjects' per- ceptions of paternal nurturance during childhood. Reuter and Biller's (1973) Paternal Availability Scale measured the subjects' perceptions of their fathers' availability during childhood. The Ben Sex-Role Inventory Scales Edward Thomas Manzitti measured sex-role development. A Personal Background Questionnaire was administered to ascertain characteristics of the participants that might affect the results. Several statistical procedures were followed. Two 2 x 2 analyses of variance were made. One analysis was for males Masculinity scores and the other analysis was for females Femininity scores. Post-hoc analyses were performed for some of the hypotheses. Additional analyses included several correlational analyses and analyses of variance for some other dependent measures of sex-role development. The results of the statistical tests supported four of the six hypotheses. The findings indicated that both paternal nurturance and paternal availability during child- hood were important for sex—role development among college students from intact families. Fathers that were both highly nurturant and highly available to their children druing childhood were more effective in facilitating sex- role development among their children as adults, than were fathers that were less nurturant and either highly avail- able or less available to their children during childhood. Fathers that were both less nurturant and highly available to their children during childhood had negative effects upon the sex-role development of their children as adults. ACKNOWLEDGEMENTS While preparing this dissertation, I requrested and I received the advice of a number of indivisuals. At this time, I would like to acknowledge their help. Louise Sause, my Chairperson, adviser, and friend spent time helping me organize this work and she helped in the editing of the numerous drafts of this manuscript. Moreover, she has been a source of encouragement and inspiration to me throughout my doctoral program. The other members of my committee also seserve recog- nition for their efforts. John P. McKinney and LeRoy Olson made constructive comments and both of them provided en- couragement. J. Edwin Keller has been accessible to me throughout my years at Michigan State University and I have appreciated his advice and encouragement in my studies and in preparing this dissertation. A number of other individuals aided me in my work and deserve credit for their assistance. I am indebted to the instructOrs of Michigan State University that permitted me to use their classes to obtain participants for my study. Without their help the data collection would have been difficult. I would also like to thank the numerous under- graduates that gave their time and participated as subjects ii in this study. The consultants at the Computer Labor- atory helped me with the use of the computer facilities and programs and I appreciate their efforts. iii II. III. TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . . . . . . . INTRODUCTION, REVIEW OF LITERATURE, STATEMENT OF THE PROBLEM, AND THE HYPOTHESES . . . . . Introduction . . . . . . . . . . . . . . . . Review of Literature . . . . . . . . . . . . Theories of sex-role development . . . . . Paternal attitudes toward children. . . . Parent-child relationship and the child's sex-role development . . . . . . . . . . . Identification, sex-role development, and adjustment . . . . . . . . . . . . . . . Paternal nurturance during childhood, adjustment, and sex—role development. . . Father-absence and sex-role development . . Father-absence and sex-role development in males . . . . . . . . . . . . . . . Father-absence and sex-role development in females . . . . . . . . . . . . . . . Other factors affecting sex-role development . . . . . . . . . . . . . . . . Summary . . . . . . . Statement of the problem and the hypotheses. Summary . . . . . . . . . . . . . . . . . . . MTHOD O O O O O O O O O O O O O C O O O O 0 Subjects . . . . . . . . . . . . . . . . . Descriptions of the instruments . . . . . . . Personal background questionnaire . . . . . A parent-child relations questionnaire (PCR)................. Paternal availability scale (PAS) . . . Bem sex-role inventory (BSRI) . . . . . Procedure . . . . . . . . . . . . . . . Statistics and research design . . . . . . . Summary . . . . . . . . . . . . . . . . . . . RESULTS 0 O O O O O O O O O O. O O O O O O O 0 Results of the tests of the hypotheses . . . Additional analyses . . . . . . . . . . . . . Pair comparisons for BSRIM scores among males . . . . . . . . . . . . . . . . . . Pair comparisons for BSRIF scores among females . . . . . . . . . . . . . . . . . . iv cannot-4 H 10 13 15 17 18 23 27 29 30 34 35 35 37 37 37 39 40 46 48 56 57 57 '64 64 65 IV. Correlational analyses . Descriptive statistics for variables. Self-ratings of males and females on BSRI Masculine and Feminine items, respectively . . . . Analyses of variance procedures for other Analyses of variance procedures for BSRIF scores among males . Analyses of variance procedures for BSRIM scores among females Additional descriptive statistics Summary . . . . . . . DISCUSSION . . . . . . and post-hoc dependent measures. and post-hoc scores and BSRIA and post-hoc scores and BSRIA Discussion of the results Paternal nurturance during childhood and sex-role development . Paternal availability during childhood and sex-role development . Paternal nurturance and paternal avail- ability during childhood and sex-role development . . . . Intelligence and sex-role development Limitations of the study . Conclusions . . . . . Recommendations for future research APPENDICES . . . . . . Appendix A Letters of permission . Appendix B Instructions for instructors and for subjects . . . . . . Appendix C Consent form and instruments Consent form . . . Personal background questionnaire . Parent-child relations questionnaire . Paternal availability scale . Bem sex-role inventory . BIBLIOGRAPHY . . . . Page 66 69 71 71 74 77 80 84 85 85 85 88 89 93 94 96 97 99 99 101 103 103 104 108 116 117 119 TABLE I. II. III. IV. VI. VII. VIII. IX. XI. XII. XIII. XIV. LIST OF TABLES PAGE Person correlation coefficients and signifi- cance levels for ACT scores and BSRIF, BSRIM, and BSRIA scores for males . . . . . . . . . . 50 Pearson correlation coefficients and signifi- cance levels for ACT scores and BSRIF, BSRIM, and BSRIA scores for females . . . . . . . . . 50 Analysis of variance for males Masculinity (BSRIM) scores . . . . . . . . . . . . . . . . 58 Descriptions of the sub-samples for males' BSRIM scores 0 O O O O O O 0 O O O O O I O O O 59 Analysis of variance for females Femininity (BSRIF) scores 0 O O O O O O O O O O O I O O I 60 Descriptions of the sub-samples for females BSRIF scores . . . . . . . . . . . . . . . . . 60 Results of the Scheffe tests for the complex comparisons that were used in Hypotheses 3 through 6 . . . C C C C C . . O . . O O . O . C 64 Results of the Tukey tests of BSRIM scores among males . . . . . . . . . . . . . . . . . 66 Results of the Tukey tests of the BSRIF scores among females . . . . . . . . . . . . . 67 Pearson correlation coefficients for males . . 68 Pearson correlation coefficients for females. . 69 Descriptive statistics for variables among males I O O O I O O O O O O O O O O O O O O O O 70 Descriptive statistics for variables among females . . . . . . . . . . . . . . . . . . z . 70 Mean self-ratings of males on Masculine items as a function of father-nurturance and father-availability . . . . . . . . . . . . . . 72 vi 75 77 79 79 80 81 81 82 82 83 TABLE PAGE XV. Mean self-ratings of females on Feminine items as a function of father-nurturance and father-availability . . . . . . . . . . XVI. Analysis of variance of BSRIF scores for females . . . . . . . . . . . . . . . . . . XVII. Analysis of variance of BSRIA scores for males 0 O O O O O O O O O O O O O O O . XVIII. Results of the post-hoc procedures for BSRIA scores among males . . . . . . . . . XIX. Analysis of variance for BSRIM scores among females . . . . . . . . . . . . . XX. Analysis of variance for BSRIA scores among females . . . . . . . . . . . . . XXI. Results of the post-hoc tests for BSRIA scores among females . . . . . . . . . . . XXII. Descriptive statistics for BSRIF scores amcng males 0 O O O O O O O O O O O C O O O XXIII. Descirptive statistics for BSRIM scores among males . . . . . . . . . . . . . . XXIV. Descriptive statistics for BSRIA scores among males . . . . . . . . . . . . . . XXV. Descriptive statistics for BSRIF scores among females . . . . . . . . . . . . . XXVI. Descriptive statistics for BSRIM scores among females . . . . . . . . . . . . . XXVII. Descriptive statistics for BSRIA scores among females . . . . . . . . . . . . . vii 83 CHAPTER I INTRODUCTION TO THE STUDY There has been an increasing amount of scholarly inquiry on sex-role development. According to the lit- erature, the masculine sex-role is usually characterized by instrumental or goal-oriented behaviors such as ag- gression, assertiveness, leadership, and self-reliance while the feminine sex-role is usually characterized by expressive behaviors such as being pleasing, solicitous, and understanding (Johnson, 1963). The literature in- dicates that sex-role development begins early in life, usually before age three (Brown, 1958; Kagan, 1958b; Maccoby and Jacklin, I974). A great deal of the literature on sex-role develop- ment has concentrated on the effects of the father. Gen- erally, the literature states that the father plays a very important role in the devel0pment of children's sex- roles. According to a number of theorists, the father facilitates sex-role development by rewarding his male children for behaving in a masculine way and by rewarding 1 his female children for behaving in a feminine way (Cot- trell, 1942; Parsons, 1955, 1958; Johnson, 1963). The father that rewards, encourages, and praises his children is called nurturant and the non—nurturant father is us- ually detached, cold, and discouraging of his children. While the position of the role theorists has been supported by empirical studies, there has also been a sub- stantial amount of research on the effects of father- absence upon sex-role development. Most of the research considers the effects of father-absence upon male sex-role development. There has also been some research on the effects of father-absence upon female sex-role development. These studies indicate that father-absence has an adverse effect upon male and female sex-role development. More- over, the research indicates that father-absence in early childhood has a greater, adverse effect upon sex-role development than does father-absence during later life (Biller, 1970, 1974, 1976; Lynn, 1974). There was a need for research that considered the combined effects of paternal nurturance and paternal avail- ability during childhood upon sex-role development among adolescents and young adults from intact families. There was, in addition, a need to study the stability of the effects of the father upon sex-role development. This study considered the effects of combinations of paternal nurturance and paternal availability during childhood upon sex-role development among college students from in- tact families. REVIEW OF THE LITERATURE Theories of Sex-role Development The research on sex-role development has been stimu- lated by a variety of theories. The psychoanalytic, learning, social learning, and role theories consider sex- role development. Freud has been a major proponent of the psychoana- lytic point of view. According to Freud, the three to five year old boy desires to have an exclusive relation- ship with his mother, and, at the same time, fears his father as an aggressive competitor for maternal affection. The boy also fears castration. In order to attain a mascu- line identification, the boy must identify with his aggres- sive father and repress his desire for his mother. For Freud, the young girl initially rejects her mother and desires her father. Freud suggests that the daughter fears maternal rejection or loss of love from the mother and, thereafter, identifies with the mother. The girl's fear of loss of love is not as strong as the boy's fear of castration, and the girl does not completely identify with the mother, nor does she completely resolve the oedipal complex (Biller, 1974). Several authors have extended Freud's theory of identification. Whiting (1959, 1960), suggested a status envy theory. According to Whiting, a masculine identifi- cation in boys will result if the boy perceives that the father has access to more priviledges, attractive objects, and activities than does the mother. Leonard (1966) in- dicated that the girl must establish "a desexualized ob- ject relationship with the father" in order to accept a feminine role without guilt or anxiety. Leonard also suggested that paternal rejection may retard the develop- ment of a feminine identification. The learning theorists have also presented a theory of the development of identification. Mowrer (1950) pre- sented some Freudian concepts in terms of learning theory. Mowrer suggests that there are two types of identification. Defensive identification is similar to Freud's identifica- tion with the aggressor. Developmental identification is similar to Freud's anaclytic identification and is an attempt to imitate the beloved parent. Mowrer empahsizes the importance of parental nurturance in rewarding the child's sex appropriate behaviors. The child becomes de- pendent upon parents for supplying nurturance and learns to perform the behaviors that the parents reward. Thus, parents can facilitate masculine behavior in boys and feminine behavior in girls by rewarding those behaviors. In addition, Mowrer suggests that early parental nur- turance toward children is important for sex appropriate identification. A similar position was presented by Sears (1957). While the social learning theorists' position is also based upon the premise that sex-roles are learned, these theorists emphasize that learning often takes place vicariously (Bandura & Walters, 1963; Mussen & Distler, 1959). That is, an individual observes the behavior of a model and learns to imitate that model without nec- essarily receiving reinforcement. In the case of sex-role learning, the child usually has male and female models available to him. Usually, while the most available models are the father and the mother, other individuals can serve as models. Thus, the child can imitate either the masculine or the feminine behaviors that he observes. Whether the child demonstrates masculine or feminine be- havior depends upon which behaviors are rewarded and which behaviors are punished. While the boy will develop a mas- culine sex-role only if he is rewarded for demonstrating masculine behavior and punished for demonstrating feminine behavior, a girl will develop a feminine sex role only if she is rewarded for demonstrating feminine behaviors and punished for demonstrating masculine behavior. While the parents are the most frequent administrators of reward and punishment, other individuals may dispense reward and punishment that affect sex role development. Perhaps the most integrative approach is presented by the role theorists (Cottrell, 1942; Parsons, 1958). According to these theorists, the child's behavior is not necessarily identical to the parent with whom he identifies. Rather, it is a result of reciprocal role relationships of the child and the parent. Reciprocal roles differ for male and female children. According to Parsons (1955, 1958) and Johnson (1963), the father provides differential rein- forcements in these relationships as a function of the sex of the child. While the father rewards the male children for demonstrating masculine behavior, he rewards female children for demonstrating feminine behavior. According to these authors, masculine behaviors included aggression, dominance, a preference for quantitative and mechanical tasks, self-reliance, assertiveness, and leadership. Fem- inine behaviors included being solicitous, affectionate, appealing, understanding, pleasing toward others, and sub- missive. The paternally administered differential rein- forcements are responsible for sex-role learning in children. Althouth the role theorists have stressed the role of the father as the primary transmitter of conceptions of masculinity and femininity, they have not neglected the role of the mother. Johnson (1963) suggests that fathers and mothers differ in their relationships with their children and she states that while fathers reward their sons and daughters for masculine and feminine behaviors, respectively, mothers are generally rewarding and en- couraging toward their children. Moreover, Johnson suggests that while mothers can facilitate their children's sex- role develoPment by communicating positive feelings about their husbands to their children, the mothers can inhibit the effects of the fathers by communicating negative feelings about their husbands to their children. The research will be considered as follows. First, there will be a discussion of the research on parental attitudes toward children's behavior. This is followed by a presentation and discussion of the research on the parent-child relationship and the degree of masculinity in males and the degree of femininity in females. Third, there will be a discussion of the literature on the rela- tionship of identification with the father, adjustment of the child, and sex-role development of the child. Finally, there will be a discussion of the research on father absence. Parental Attitudes Toward Children Two major studies considered parental attitudes toward young children's behavior. Goodeneough (1957) studied parents' attitudes toward their male and female children. Interviews were used to assess mothers and fathers attitudes toward their children's behavior. The findings indicated that the fathers were more interested than were the mothers in facilitating appropriate sex- role development in their children. Fathers considered sex typed behaviors as a major difference between the sexes. (Sex typed behaviors are those behaviors that are socially acceptable for one's sex. A sex typed individual is simultaneously motivated to keep his behavior consis- tent with a standard that reflects social desirability and suppress any behavior that might be considered unde- sirable for an individual of his gender.) Fathers ex- pressed restraint in showing affection for their sons while mothers indicated that they showed affection for both their sons and daughters. Fathers, moreover, were more concerned than were the mothers about masculinity in their sons. Similar findings were reported by Sears, Maccoby and Levin (1957). The later study used interviews of mothers of young children and reported that mothers expressed little to moderate concern over the masculinity of their sons. Sears et a}: reported that mothers disliked the idea of treating boys and girls differently. The study also suggested socioeconomic status differences. Working class mothers expressed more severity and less permissiveness in their child—rearing practices. Middle class mothers expressed more warmth than working class ‘mothers toward their young children. In contrast, working class mothers were more critical of their husbands than were the middle class mothers. Three additional studies considered fathers attitudes toward children of various ages and found that fathers' attitudes toward their male and female children differ as a function of the sex of the child. Tasch (1952) inter— viewed urban, middle-class fathers in the New York City area. While the fathers were between thirty and fifty years of age, their children ranged in age from infancy through adolescence. Several findings were reported. First, fathers indicated a preference for participating in the daily personal care of their daughters more than in the daily care of their sons. Second, the fathers indi- cated a preference for participating more often in the ‘motor activities with boys than with girls from early childhood through adolescence. Third, the fathers reported assigning different chores to boys and girls as a function of gender. Girls were asked to iron and care for younger siblings. Boys were asked to remove garbage and help 10 their fathers with his work. Adolescent boys were often encouraged to find employment. Related findings were reported by Aberle and Naegele (1952). The latter authors interviewed middle class fathers at home and reported that fathers were more concerned about career planning for boys than they were for girls. In addition, while these fathers expressed concern for obedience, responsi- bility, school achievement, athletic ability, and an over- all masculine role in their sons, they wanted their daughters to be nice, sweet, pretty, affectionate, and well liked. The findings are consistent with an earlier study by Gardner (1947). Gardner considered male and female pre-adolescents and adolescents and reported that although boys were given physical chores by their fathers, the girls were assigned to sedintary tasks. In summary, the findings of Goodenough (1957), Sears et a1" (1957), Tasch (1952), Aberle & Naegele (1952), and Gardner (1947) are consistent and support the role theorists' position. Parent-Child Relationship and the Child's Sex-Role Development A number of studies have considered the child's rela- tionship with his parents and the child's sex-role develop- ment. Generally, the research indicates that the father is important for the development of masculinity in males and 11 femininity in females. The research on young children indicates that fathers are important for the development of sex-roles in males and females. P.S. Sears (1953). used maternal interviews to gain information about the father-child relationship and a doll play projective test to measure the child's sex-role orientation. Her findings indicated that children with appropriate sex-role orientations also had warm re- lationships with their fathers. Similar findings were re- ported by Mussen & Distler (1959) and Mussen & Rutherford (1963). The former study used the It Scale to measure the boys sex-role orientation and a doll play projective test to measure the boy's perception of paternal nurturance, punishment and power. Paternal power was measured by the number of situations in the doll play where a father was mentioned. Paternal punishment was determined by the num- ber of situations where the subject described the father as punitive. Paternal nurturance was defined as the number of situations where the father was described as encouraging and warm. The finding indicated that highly masculine boys, as opposed to less masculine boys, perceived more paternal nurturance, more paternal power, and more paternal punishment. Mussen & Rutherford (1963) reported similar findings with males and females. The latter study used a questionnaire to measure paternal encouragement of sex 12 appropriate play. Examples of sex appropriate play in- clude rough and tumble play for boys and doll play for girls. The results indicated that highly masculine boys and highly feminine girls reCeived more paternal encour— agement than did less masculine boys and less feminine girls, respectively. Moreover, the research on older children indicates that the father continues to be important for sex-role development. Bronson (1959) considered a sample of nine through thirteen year old boys. While the father-son relationship was measured by a combination of interviews with both parents and children, a projective test was used to measure the boy's masculinity. The findings suggested that boys with.warm, non—stressful relationships with their fathers had higher masculinity scores than boys with stressful relationships with their fathers. Mussen (1961) studied adolescent male sex-role orien— tation and reported that a positive father-son relationship was important for the development of masculinity in sons. He used seventeen and eighteen year old boys and found that highly masculine boys had more positive relationships with their fathers than did the less masculine boys. On the other hand, boys that scored low on masculinity had more negative relationships with their fathers than did boys that scored 13 high on masculinity. While the research supports the notion that the father- child relationship is important for the development of sex- roles, there is some evidence that the effects of the mother- child relationship differ with the sex of the child. The research on young children by Mussen & Distler (1959) and Mussen & Rutherford (1963) indicated that while ma- ternal warmth, power, and punishment had no effect upon the development of a masculine sex-role in males, the mother-child relationship seems to have an effect upon the development of a feminine sex-role in females. Mussen & Rutherford (1963) reported that maternal warmth and en- couragement differentiated high femininity girls from low femininity girls, with the high femdninity girls receiving more maternal warmth and encouragement than the low fem- ininity girls. Among adolescent boys the research indicates that the mother-son relationship has little effect upon the development of a masculine sex-role (Mussen, 1961). Un- fortunately, there is no evidence for the effects of the mother-child relations upon feminine sex-role development among adolescent females. Identification, Sex-Role Development, and Adjustment The research suggests that there is a relationship between identification with the father, appropriate sex-role 14 development, and adjustment. Most of the evidence comes from studies with adolescents and young adults. Sopchak (1952) studied male and female college students. He used the MMPI as a measure of adjustment, sex-role, and iden- tification. While positive adjustment was defined as the lack of abnormal personality traits, negative adjustment was represented by the presence of those traits. Students were required to take the MMPI three times; in the usual manner, as father would answer, and as mother would answer. The MMPI that was taken in the usual manner was used to measure adjustment and sex-role. The students identifica- tion with each parent was measured by the degree of sim- ilarity between responses to the MEET taken in the usual manner and the responses to the MEET taken as the father or the mother would respond. The findings suggested that masculine men identified with their fathers more than did the less masculine men. The men that demonstrated high identification with their fathers were also better adjusted and more masculine than were the men that showed less iden- tification with their fathers. Masculine women identified less with their fathers than did the less masculine women. Women who lacked identification with their fathers also tended to be less well adjusted than were women that were highly identified with their fathers. Similarly, Mussen (1961) reported that highly masculine boys scored higher on adjustment than did the less masculine boys. Studies 15 by Gray (1959) and Helper (1955) indicated support for the relationship among children and adolescents. More- over, Osgood, Suci, & Tannenbaum (1957), reviewed the literature and gave additional support for the relationship. Some evidence suggests that while paternal sex-role orientation is not related to boys identification with the father, the mother's sex-role orientation seems to be re- lated to the son's identification with his father. Payne and Mussen (1956), administered a variety of personality tests to adolescent males, their fathers and their mothers. They found that there was no relationship between the boy's identification with his father and the father's masculinity score, although there was a negative relationship between maternal masculinity scores and the son's identification with his father. It seems that masculine mothers interfere with the relationships of their husbands and sons. Paternal Nurturance During Childhood, Adjustment, and Sex-Role Development The evidence indicates that paternal nurturance, in the form of encouragement and warmth during childhood, is important for the development of positive adjustment in late adolescence and adulthood. One study has examined college females perceptions of paternal nurturance in child- hood and their personal adjustment. Fish & Biller (1973) considered father—absent college females. While the 16 authors used Schaeffer's Parent-Perception questionnaire as a measure of perceived paternal nurturance during childhood, they administered the Adjective Checklist as a measure of personal adjustment. The findings indicated that those females that perceived a high degree of pater- nal nurturance during childhood had higher adjustment scores than did the females perceived a lower degree of paternal nurturance during childhood. A related study was conducted by Reuter & Biller (1973). ,These authors demonstrated that both paternal nurturance and paternal availability during childhood were important for positive adjustment in college males. These authors measured college males perceptions of paternal availability and paternal nurturance and college males' overall adjustment. The findings suggested that while paternal availability alone did not affect overall adjust- ment, nurturance and nurturance x availability affected adjustment. High paternal nurturance combined with at least moderate paternal availability and high paternal availability combined with at least moderate paternal nur- turance were related to high scores on personal adjustment. On the other hand, high paternal nurturance combined with low paternal availability and high paternal availability combined with low paternal nurturance were associated with low scores on personal adjustment. 17 These studies indicate that paternal nurturance during childhood is important for positive adjustment in both male and female college students. While Renter & Biller (1973) showed that paternal availability combined with paternal nurturance was necessary for positive adjustment among college males, there is a need for research that examines the effects of paternal availability combined with paternal nurturance upon female adjustment. Moreover, taken with the findings on the relationship of sex-role development and positive adjustment, the evidence suggests that pater- nal availability combined with paternal nurturance during childhood is important for the development of masculinity in males and femininity in females. Father-Absence and Sex-Role Development The findings of some of the research on father-child relationships suggest that paternal availability may affect sex-role development. While there has been little research that considers the effects of paternal availability upon sex-role development among children from intact families, a substantial number of studies have compared children from father—present and father-absent families. These compari- sons have considered the sex-role development of males and females. The review that follows considers the research on father-absence and sex-role development. First, there will be a discussion of the research on males. This will 18 be followed by a discussion of the studies on females. Father-Absence and Sex-Role Development in Males Generally, the research on males indicates that father-absence has a negative effect upon the development of a masculine sex-role. Some of the early work considered young, male children. These studies compared father-present and father-absent males on sex-role development. While father-present males had fathers available in the home, the father-absent males had no father available. Sears, Pintler, and Sears (1946), and Bach (1946) used doll play tests to measure sex-role development in young boys, and in both studies aggression was the measure of masculinity. While a high level of aggression indicated high masculinity, a low level of aggression indicated low masculinity. In the former study, sex-role was measured by the number of aggressive acts that were demonstrated by the subject. The findings of Sears eg_al., (1946) indicated that father present boys were more aggressive than were the father- absent boys. Similar findings were reported by Bach (1946). In the latter study, subjects described a fantasy about the dolls. These stories were rated on aggression. Bach (1946). reported that the stories of the father-present boys included an aggressive father more often than did the stories of the father—absent boys. Moreover, the stories of the father-separated boys included descriptions 19 of an affectionate father more often than did the stories of the father-present boys. The evidence indicates that the duration of father- absence is an important factor. Biller (1969b) examined kindergarten boys. He used Brown's It Scale as a measure of sex-role orientation. The findings indicated that father-absence of more than two years had a greater nega- tive effect on sex-role development than did father— absence of less than two years. Those boys who were father-absent for more than two years had less masculine It scores than did the boys whose fathers were absent for less than two years. Moreover, as in the previous studies, boys with fathers available continuously had higher mascu- linity scores than did any of the father-absent boys. The findings from the research on adolescents indi- cates that early father-absence has a greater negative effect upon sex—role development, than does late father- absence. Biller & Bahm (1971) compared early father- absent boys, late father-absent boys, and father-present boys. While early father—absence was defined as father- absence before age five, late father—absence was defined as separation from the father after age five. While the early father-absent boys had lower masculinity scores than did the late father-absent boys, there was no difference between father—present and late father—absent boys. 20 The conclusions of several literature reviews consis- tently support the fact that early father-absence has a greater negative impact upon male sex-role development, than does late father-absence (Herzog & Sudia, 1968; Biller, 1970, 1974, 1976; Lynn, 1974). These reviews in- dicated that early father-absence, before age five, had an especially damaging effect upon male sex-role development. The evidence suggests that reason for father-absence does not affect male sex—role development. McCord.g§_§1. (1962) studied ten through fifteen year old males and used observations and interviews to measure aggression among the boys. They considered the effects of father- absence due to divorce, death, or legal separation and found that the reason for paternal separation did not con- tribute to the differences in male sex—role development. Furthermore, these authors also reported that the father- absent boys demonstrated less aggressive behavior than did the father—present boys. In addition to onset and duration of father—absence, it appears that race is an important factor in male sex— role develOpment. Biller (1968) compared black and white boys that ranged from five through six years of age. These boys were given a series of tests that were used to deter— ndne their abilities to discriminate male typed items from female typed items. While male typed items were items that were judged to be masculine by a group of raters, 21 female typed items were items judged to be feminine by a group of raters. Several findings were reported. First, white boys received higher masculinity scores than did black boys. Moreover, father-absent boys received lower masculinity scores than did the father—present boys. While father-absent whites received higher masculinity scores than did father-absent blacks, there were no differences between father-absent whites and father- present blacks. Similar findings were reported by two studies on adolescents. Barclay & Cusumano (1967) used the Rod and Frame Test to measure field dependence. Field depen— dence is the degree to which an individual's judgment of an object are affected by the surrounding of the object. A field dependent individual's judgments of an object are affected to a greater degree than are the judgments of a field independent individual. Moreover, field dependence has been associated with females more often than with males. Field independence, on the other hand, has been associated with males more often than with females. Thus, a low field dependence was indicative of high masculinity while a higher field dependence score suggested that the individual was low on masculine orientation. The findings indicated that blacks were more field dependent than were whites. Furthermore, father-absent males were more field dependent than were father-present males. Comparable findings were 22 reported by Hartnagel (1970). In the latter study, the boys were asked to rate themselves on self-potency. Self- potency included power in social relations, self-reliance, and assertiveness. The findings indicated that the whites had higher self-potency ratings than did the blacks. Moreover, the father present group had higher self-potency scores than did the father-absent group. The theoretical position of the role theorists in— dicates that maternal encouragement may effect sex-role development (Parsons, 1958; Johnson, 1963). This position is supported by empirical evidence. Biller & Bahm (1971) indicated that adolescent boys' perceptions of maternal encouragement of masculine behaviors was positively re- lated to masculinity scores among early father-absent males. Masculine behavior was defined as aggression, assertive- ness and self-reliance. Moreover, McCord e;_al. (1962) indicated that boys with rejecting mothers were more de- pendent than were boys with encouraging mothers. Thus, it seems that mothers can facilitate masculine behavior in their sons by encouraging such behavior. Some evidence indicates that father-absent boys re— ceive less maternal encouragement of masculine behavior than do father—present boys. Biller (1969b) used a ques- tionnaire to assess maternal encouragement cfmasculine behavior among kindergarten boys. Masculine behavior in- cluded aggression and self—assertiveness. He found that 23 mothers of father-present boys encouraged masculine be- havior more often than did the mothers of father-absent boys. Apparently, part of the difference between the father—present and the father absent boys sex-role develop- ment is accounted for by the differences between father- present and father-absent boys"mothers behaviors. It appears that the availability of a surrogate father counteracts the effects of father-absence. The father surrogate is an older male that is available to the father-absent child. Santrock (1970) considered a sample of male, black pre-school children. While inde— pendence among the boys was indicative of a masculine sex-role, dependence suggested a lack of masculinity. This study showed that father-absent boys with an avail- able surrogate father were more independent than were the father—absent boys without a surrogate father. Father Absence and Sex-Role Development in Females While there is less evidence on the effects of father absence upon female sex—role development, the research in- dicates that the absence of the father has a negative effect upon the girl's identification with her father. Bach (1946) considered father-absent and father-present girls and used the subjects'doll play stories to measure the girl's identification with her father. The findings indicated that the stories of the father-absent girls 24 contained a father less often than did the stories of the father-present girls. Similarly, Lynn & Sawrey (1959) reported less identification with the father among the father-absent girls as opposed to the father-present girls. The latter study also used a doll play task and indicated that the father-absent girls selected the father doll less often than did the father-present girls. The research also indicates that father-absent girls are more dependent upon their mothers than are father- present birls. Lynn & Sawrey (1959),.stated that the father-absent girls selected a mother doll more often than did the father-present girls. Hetherington (1972) cOnsidered adolescent girls and reported that father-absent firls showed a greater dependence upon their mothers than did the father-present girls. Apparently, father-absent women have difficulty in their relationships with men. Jacobson & Ryder (1969) compared father—present and father-absent women on marital relationships. While the father-absent women were separated from their father during childhood, the father-present women reported an available father during childhood. The findings indicated that the father—absent women had less favorable relationships with their husbands than did the father-present women. The relationships of the father- 25 absent women were characterized by discord and difficulties in sexual relationships with their husbands. The evidence indicates that both the time of father- absence and the reason may be important for female sex- role development. Hetherington (1972) considered adoles- cents using personality tests and observational ratings to measure aspects of sex-role development. She found that more daughters of divorcees, versus other father- absent girls, sought attention from male peers and contact with male peers. Second, daughters of widows avoided males and preferred females. Third, early father separated girls showed less preference for feminine activities, such as housework and child care, than did the late father sep- arated girls. While early separation was defined as father-absence before age five, late separation was defined as father-absence after age five. Finally, the early father separated females spoke less to a male interviewer than did the late father separated females. Some evidence indicates that the presence of older ‘male siblings has a negative effect upon sex-role develop- ment in females. It appears that father-absent girls with older male siblings only are more aggressive than are father-absent girls with older female siblings only. Santrock (1970) considered father-absent girls with older male siblings only and father-absent girls with older female 26 siblings only. He found that father—absent girls with older male siblings only were more aggressive than the father—absent girls with older female siblings only. Possibly, the greater aggression that was demonstrated by the girls with the male siblings is a defensive reaction to their older brothers. Taken together, the findings of the research on the effects of father-absence upon adolescent and adult females is consistent with the research on males. The research on females suggests that father-absence, especially during early life, has a negative effect upon the develop- ment of heterosexual relationships during adolescence and adulthood. The finding of differential effects for the reason for father-absence also provides support for the notion that paternal availability is important for female sex-role development. While daughters of divorcees sought male attention and contact, the daughters of widows avoided males and preferred females. Conceivably, the daughters of divorcees had more contact with their natural fathers during early life than the daughters of widows. It is possible that the greater paternal contact that was ex- perienced by the daughters of divorcees facilitated the greater heterosexually oriented behavior of the divorcees' daughters. On the other hand, the limited paternal con- tact that was experienced by the daughters of the widows may have had a negative effect upon the development of 27 heterosexual interests among the daughters of widows. Finally, the effects of surrogate fathers and siblings are important. While surrogate fathers can counteract the effects of father-absence in males, older male siblings may have an negative effect upon female sex-role develop- ment. Other Factors Affecting Sex-Role Development Some authors have indicated that socioeconomic status may affect sex-role development. Biller (1974, 1976), along with Altus (1958), has stated that there are differences in sex-role development between middle and low socioeconomic status males and females. According to some authors, lower class fathers often assert less in- fluence over the assignment of chores to his male and female children than does the middle class fathers. Other authors have suggested that children from lower class families often perceive their fathers as less dominant than their mothers in the family decision making process (Bowerman & Elder, 1964; Distler, 1964). Moreover, S.E.S. differences have been reported with respect to paternal punitiveness and affection. McKinley (1964) indicated that lower class fathers were more punitive and less affectionate than were middle class fathers. Apparently, the literature indicates that lower class children may experience more difficulty than middle class children in 28 their relationships with their fathers. Some authors have suggested that race, father surro- gates, and IQ may affect the development of sex-roles in males and females. Several studies have suggested that blacks often score lower on measures of masculinity than do whites (Barclay & Cusumano, 1967; Hartnagel, 1970; Biller, 1974, 1976). Furthermore, the presence of father surrogates in the form of older male siblings or older males residing in the home with the children may affect the development of sex-roles. Santrock (1970)_ suggested that father-absent boys with an available substitute father were less dependent than father-absent boys without a substitute. Dependency has often been associated with feminine sex-role (Maccoby & Jacklin, 1974). A number of authors have suggested that intelligence may affect sex- role (Kohlberg, 1966; Biller, 1970, 1974, 1976; Jones, 1975). Kohlberg (1966), among others, argued that individuals with lower IQs have more difficulty learning socially defined concepts of sex-role. Moreover, Kohlberg presented data that indicated that differences between father-present and father—absent boys were reduced when the subjects were ‘matched on IQ. In addition, some studies have suggested that among blacks, father-absent boys scored lower on IQ and achievement tests than did father-present boys (Deutsch, 1960; Deutsch & Brown, 1964). 29 Summary After considering the research that has been reported, several observations can be made. 1. Paternal warmth and encouragement during childhood is important for development of appropriate sex- roles in males and females. Father-absence during childhood is detrimental to the sex-role development of males and females. Father-absence during early childhood has a greater negative effect than does later father-absence upon sex-role development in males and females. A highly available and highly nurturant father is most effective for positive adjustment among males and females. Furthermore, it seems that this kind of a father is most effective for appropriate sex- role development in males and females. Similarly, an unavailable and non-nurturant father is least effective for positive adjustment in males and females. Moreover, it seems that this kind of a father is least effective in appropriate sex-role development in males and females. While the mother may have either a positive or a negative effect upon the development of masculinity in males, there is a need for more research on the effects of the mother upon feminine sex-role development. 30 7. Maternal encouragement of masculine behavior may counteract the effects of father-absence. 8. There is some evidence that there are racial and possibly S.E.S. differences in sex-role development. 9. The effects of the father may be equally important for male and female sex-role development. 10. The presence of a surrogate father may compensate for the negative effects of father absence upon sex- role development in males. Statement of the Problem and the Hypotheses After considering the evidence, a problem for study was formulated and a discussion of this problem follows. The research on the parent-child relationship and sex-role development indicates that paternal nurturance facilitates sex-role development in childhood and adoles- cence, but there was a need for research that considers the lasting effects of this nurturance into adulthood. The positive effects of paternal nurturance have been demon- strated among young children (Mussen & Distler, 1959; Mussen & Rutherford, 1963). There is also support for the relation— ship from studies with pre-adolescent and adolescent males (Payne & Mussen, 1956; Mussen, 1961). While the research indicates that father-absence during childhood has a negative effect upon sex-role de- velopment, there was a need for research that examined 31 the effects of paternal availability during childhood upon sex-role development among adolescents and adults from intact families. The negative effects of early father-absence upon sex-role development have been shown among samples of males from early childhood through adult- hood. While there is less evidence for females, the studies indicate that father-absence during early child- hood has a negative effect upon sex-role development among adolescent and adult females (Hetherington, 1972; Jacobson & Ryder, 1969). Unfortunately, no previous study had systematically considered the effects of paternal avail- ability during childhood upon the development of sex-roles among individuals from intact families. Moreover, no previous study considered the stability of the effects of paternal availability during childhood upon individu- als from intact families, and there was a need to con- tribute to the study of the stability of the effects of this variable. While the evidence indicates that high levels of paternal nurturance and availability during childhood have greater positive effects upon positive adjustment among young adult males and females, it was necessary to consider the effects of combinations of these variables upon sex- role development among adolescent and young adult indivi- duals from intact families. Although Fish & Biller (1973) supported the relationship between paternal nurturance 32 during childhood and adjustment among father-absent college females and Reuter & Biller (1973) showed that both paternal nurturance and availability were important for positive adjustment among college males, no previous study considered the effects of combinations of paternal nurturance and availability during childhood upon sex- role development among adolescent and young adult indivi- duals from intact families. This study considered some of the issues that have not been studied in previous research. The study con- sidered the effects of paternal nurturance and avail- ability during childhood, upon sex-role development among adolescent and young adult college students from intact families. This research also contributed to the study of the stability of the effects of paternal nurturance and paternal availability. Several hypotheses follow and these hypotheses con- sidered some of the unresolved issues that were previously described. Hypothesis 1: Adolescent and young adult males with perceptions of having had a highly nurturant father during child- hood will have higher Masculinity scores than will adolescent and young adult malesvfidfli perceptions of having had a low nurturant father during childhood. 33 Hypothesis 2: Adolescent and young adult females with perceptions of having had a highly nurturant father during child- hood will have higher Femininity scores than will adolescent and young adult females with perceptions of having had a low nurturant father during childhood. Hypothesis 3: Adolescent and young adult males with perceptions of having had a combination of high paternal nur- turance and high paternal availability during child- hood will have higher Masculinity scores than will adolescent and young adult males with perceptions of having had other combinations of paternal nurturance and paternal availability during childhood. Hypothesis 4: Adolescent and young adult females with perceptions of having had a combination of high paternal avail- ability and high paternal nurturance during childhood will have higher Femininity scores than will adoles- cent and young adult females with perceptions of having had other combinations of paternal nurturance and paternal availability during childhood. Hypothesis 5: Adolescent and young adult males with perceptions of having had low paternal nurturance and high paternal availability during childhood will have lower Masculinity 34 scores than will adolescent and young adult males with perceptions of having had other combinations of paternal nurturance and paternal availability during childhood. . Hypothesis 6: Adolescent and young adult females with perceptions of having had low paternal nurturance and high paternal availability during childhood will have lower Femininity scores than will adolescent and young adult females with perceptions of having had other combinations of paternal nurturance and paternal availability during childhood. Summary This chapter contained the Introduction, Review of Literature, and the Statement of the Problem and the Hypotheses. The Review of Literature included the major theories of sex-role development and it considered the relevant empirical work on parent-child relationships and on father-absence,along with other research. The Problem considered some aspects of the father-child rela- tionship and father-availability during childhood and sex-role development among college students from intact families. Finally, six hypotheses were presented. CHAPTER II METHOD This chapter includes aspects of the methodology that were used for this study. First, there is a dis- cussion of the subjects that participated in the study and this is followed by descriptions of the instruments. The sections on procedure, statistics, and research de- sign follow thereafter. Subjects The subjects of the study were 68 male and 72 female undergraduates at a large midwestern university. These individuals ranged from 17 through 23 years of age and all of them were from intact families. It was indicated that there was a need for research on individuals with these characteristics. Earlier it was reported that a variety of variables affected sex-role development and some of these variables included socioeconomic status, race, siblings, and surrogate fathers. These variables were controlled by restricting the sample and while this method of control had the effect of 35 36 restricting the generalization of the findings, it had the advantage of being more efficient than other proce- dures. In order to control for socioeconomic status, all subjects were from middle and upper—middle class families. Paternal occupations and paternal education were used to assess socioeconomic status. The United States Department of Labor's Dictionary of Occupational Titles was used to code paternal occupations. While numerically low codes indicated professional or high socioeconomic status occupations, the numerically high codes were characteristic of manual or low socioeconomic status occupations. The subjects reported their parents' occupations and educational attainments. Since the litera- ture indicated that differences were apparent between low and middle socioeconomic status families, the low socio- ecnoomic status subjects were eliminated from the analysis. In this study, low socioeconomic status individuals were defined as those individuals that reported having fathers with no more than a high school education and having fathers that were employed in low status occupations or unemployed. In order to control for race, all subjects were White. Moreover, restricting the sample to first- born and only children insured that older sibling surrogates were unavailable to the subjects. Finally, the study con- trolled for some surrogate father figures by restricting the sample to individuals without older males (aside from 37 fathers) living with them. Descrippions of the Instruments Personal Background Questionnaire In order to control for a variety of variables that affect sex-role development, it was necessary to collect information about the backgrounds of the participants. To this end, a revision of Jones (1975) questionnaire was used (See Appendix C). While the participants did not provide their names, they were asked to respond to a number of items related to racial, academic, socio-economic status, and family variables, among others. The ques- tionnaire required less than ten minutes for completion. A Parent-Child Relations Questionnaire (PCR) The PCR.II is used to measure parental behavior towards young children, as experienced by the child. The instru- ment is usually administered to adolescents and young adults who complete the instrument with reference to their own childhood. The original questionnaire, the PCR I, was de- scribed by Roe & Siegelman (1963) and consisted of two forms that were used to measure perceptions of paternal and maternal behavior. In addition, the original instrument consisted of 10 subtests. A revised instrument, PCR II, was described by Siegelman & Roe (1978) and consists of 38 four forms, Father-Son, Father-Daughter, Mother-Son, and Mother-Daughter. Each of these forms measures the son's or daughter's perceptions of the mother's or father's behavior. The PCR II consists of five scales including Loving, Rejecting, Casual, Demanding and Attention. These categories relate to parent behavior as perceived by the child. The Loving and the Rejecting scales were used in this study. Loving parents were characterized by warmth, affection, helpfulness, reasoning with the child, and encouragement. Rejecting parents were described as un- available, ridiculing, inattentive to children, and not helpful. A factor analysis performed by the authors on the original PCR I considered undergraduate college students from CCNY and Harvard. The results of the factor analysis indicated that the five scales could be combined into three factors including Loving-Rejecting (LR), Casual-Demanding (CD), and Attention (AT). The PCR II is easily scored and administered. Each scale contains 10 items. Respondents are given four choices for each item including "Very true", "Tended to be true", "Tended to be untrue", and "Very untrue". These four choices are assigned point values including 4, 3, 2, and 1, respectively. The score for a variable is the total num- ber of points on the items that correspond with the vari- able. Factor scores are computed by subtracting one variable 39 from its polar opposite and adding 50. For example, the score for factor I (LR) is equal to the LOV score minus the REJ score plus 50. Administration time averages approximately 15 minutes. Siegelman & Roe (1978) also reported normative data on the PCR II. The data was collected in different geo- graphic regions of the United States, for black and white, ‘males and females. The authors reported that overall test-retest reliabilities for each of the variables and factors was high.with r ranging from approximately .77 to .95. For the purposes of this study, Factor I (LR) was used. This factor closely approximates the study's definition of nurturance~non-nurturance. The factor LR is preferred over either of the simple variable scores since the factor score has a higher test-retest reliability than either LOV or REJ scores (LR, r=.95; LOV, r=.94; REJ, r=.89). Paternal Availabilipy Scale (PAS) Reuter & Biller's (1973) paternal availability scale was used to ascertain the subjects' perceptions of the degree of paternal availability during childhood. The scale is a revision of one of the sub-scales of Winch's (1962) Family Life Inventory. The subjects were presented with ten items that related to paternal availability 40 during childhood and were asked to rate each item on a five point scale, ranging from very frequently to very seldom. The total score indicates the subjects' percep- tion of the degree of paternal availability during child- hood. Unfortunately, there is no information regarding the reliability and validity of this instrument. Bem Sex Role Inventory (BSRI) The BSRI was described by Bem (1974). The instrument contains both a malCulinity and a femininity scale. Each of these scales contains twenty personality characteris- tics. These items were selected as masculine or feminine based upon sex typed social desirability. In other words, a characteristic was judged to be masculine if it was judged to be more desirable in American society for a man than for a woman, while a characteristic was selected to be feminine if it was judged to be more desirable for a woman than for a man. In addition, the BSRI contains an Androgyny scale. Androgyny is the degree to which an in- dividual is both masculine and feminine. That is, the_ degree to which an individual is both instrumental and ex- pressive, both assertive and yielding. The BSRI Androgyny scale is useful in measuring the degree to which an indi- vidual is sex-typed or androgynous. A sex-typed individual behaves in a manner that is consistent with an internalized sex-role standaris. Sex-typed individuals usually suppress 41 behavior that is considered undesirable for his/her gender. The BSRI characterizes an individual as sex-typed or Androgynous as a function of the difference between his endorsement of masculine or feminine items. Thus, a sex- typed individual is characterized by a high difference score while an Androgynous individual is characterized by a difference score that is low. Finally, the BSRI con- tains a Social Desirability Scale that is neutral with re- spect to sex. Bem (1974) presents the method of item selection for the BSRI. Initially a list of 200 items was formulated by the author and several students. Items that were thought to be both positive in value and either masculine or feminine were included in the list. Items for the Social Desirability scale were selected in a similar fashion, with a total of 200 items being chosen on the basis of their gender specific independence or neutrality with respect to sex. Half of these Social Desirability items were positive in value while the remaining half were negative in value. The final list of items for the Mascu- linity and Femininity scales were selected on the basis of judgments of their desirability in American society for one sex as opposed to the other. The judges were Stanford University students. The judges were asked to rate each of the items on a seven-point scale ranging from 1 ("Not at all desirable") to 7 ("Extremely desirable"). Each 42 judge rated the desirability of each of the items for males or for females. Masculine characteristics were identified as those that were judged independently by both males and females to be significanlty more desirable for a man than for a woman (p <.05). Feminine items were identified as those that were judged independently by both males and females to be significantly more de- sirable for a woman than for a man (p <.05). From those characteristics that satisfied the cri- teria, 20 were chosen for the Masculinity scale, while 20 were selected for the Femininity scale. Neutral items were those that were judged independently by males and females to be no more desirable for one sex than for the other, and if male and female judges did not differ sig— nificantly in their overall desirability judgements of that item. Ten positive and ten negative items were selected from the list of items that satisfied the cri- teria for neutrality. These final twenty items composed the Social Desirability scale. The BSRI yields a Masculinity, Femininity, Androgyny and Social Desirability score for each respondent. The Masculinity and Femininity scores suggest the degree to which an individual endorses masculine and feminine char- acteristics as self-descriptions. The Masculinity score is the mean self-rating for all endoresed masculine items, while the Femininity scores equal the mean self-rating for 43 all endorsed feminine items. Both means can range from 1 to 7. The Androgyny score is the difference between a respondent's Masculinity and Femininity scores. The greater the absolute value of the Androgyny score, the more the person is sex-typed or sex-reversed with high positive scores indicating femininity and high negative scores indicating masculinity. Thus, a masculine sex- role represents simultaneous endorsement of masculine characteristics and rejection of feminine attributes,whi1e a feminine sex-role is represented by endorsement of feminine characteristics and rejection of masculine char- acteristics. Androgyny, on the other hand, increases as the absolute value approaches zero. Thus, an Androgynous individual is characterized by equal endorsement of both masculine and feminine characteristics. The Social Desirability score indicates the degree to which an in- dividual describes himself in a socially desirable direc- tion on items that are neutral with respect to gender. The Social Desirability score is computed by reversing the self raings on the ten undesirable items and then calculating the respondent's mean score over all twenty items. Thus, Social Desirability can range from 1 to 7, with l repre- senting a tendency to describe one's self in a socially undesirable direction and 7 representing a tendency to describe one's self in a socially desirable direction. 44 Bem (1974) reported the results of the psychometric analyses of the BSRI. The BSRI was administered to approximately 1000 students enrolled at either Stanford University or Foothill JuniOr College during the Spring, 1973. Measures of internal consistency indicated that all three scales were highly reliable. Coefficient alpha. was computed separately for Masculinity, Femininity, and Social Desirability scores of the subjects in each of the two normative samples using the procedure outlined by Nunnally (1967). In the Stanford sample: Masculinity 3=.82; Femininity 3:.80; Social Desirability 3=.75. In the Foothill sample: Masculinity a=.86; Femininity a =.82; and Social Desirability 3=.70. The reliability of the Androgyny difference score was .85. In addition, the author reported that Masculinity and Femininity were empirically independent. In other words, the structure of the BSRI does not constrain these indices and they are free to vary independently. The following correlation coefficients were reported for Masculinity and Femininity scales. Stanford male r=.ll; female r=-.14; Foothill male r=-.02; female r=-.07. Thus, there was no apparent relationship between Masculinity and Femininity. Bem (1974) also investigated the relationship of Masculinity, Femininity and Androgyny with Social Desir- ability. She found that while both Masculinity and 45 Femininity were correlated with Social Desirability, Androgyny was not correlated with Social Desirability. Thus, Androgyny was not measuring a general tendency to respond in a socially desirable direction. Instead, Androgyny was measuring a specific tendency to describe oneself in a manner that was consistent with sex-typed standards of desirable behavior for both men and women. Several test-retest reliability coefficients were also computed. Twenty-eight males and twenty-eight females were tested twice with a four week interval between testings. The findings indicated that all four scores were highly reliable over the four week period (Masculinity r=.90; Femininity, r=.90; Androgyny, r=.93; Social Desir- ability r=.89). Bem (1974) also computed norms for each of the scales. Males scored significantly higher than females on the Masculinity scale, while females scored significantly higher on the Femininity scale in both samples. In addition, with respect to androgyny, males scored on the masculine side while females favored the feminine side. While Stanford females scored significantly higher on Androgyny than Stanford Males, there was no significant difference between males and females at Foothill. A number of other studies have considered the BSRI. In general, the studies support Bem's (1974) finding of the independence of the Masculinity and Femininity scales. 46 Low intercorrelations between the Masculinity and Feminity scales have been reported by Gaudreau (1975), Waters & Pincus (1976), and Waters, op_al., (1977). BSRI Masculinity and Femininity scores will be used in the proposed study. The Masculinity and Femininity scores will represent the dependent variables for the hypotheses in this study. Procedure The procedure that follows was approved by the Uni- versity Committee for Research on Human Subjects and the University Committee on Institutional Research granted permission for the use of University student records. This insured the protection of the rights of the partici- pants and of the author. The instruments that were previously described were distributed to the subjects that were students in under- graduate classes at a large midwestern university. Prior to the distribution of the instruments, a letter of intro- duction and a statement of qualifications for participation in the study was read by the author or instructor. Those individuals that volunteered to participate and met the qualifications for participation were given a packet of materials which included: 1. A letter of introduction and instructions. 2. A 3x5 card including spaces for name and student 47 number along with a numerical code. (Com- pletion of this information was optional.) 3. A consent form. 4. A Personal Background Questionnaire. 5. A Parent-Child Relations Questionnaire. 6. A Paternal Availability Scale. 7. A Bem Sex-Role Inventory Copies of all the above mentionedraterials that were given subjects are contained in Appendix C. The subjects took the materials with them, completed them at a convenient time, and returned them to their instructors. Each page in the packet of materials was numerically coded so that each participant had a unique code. While these codes were used to identify and collate responses, the codes also protected the privacy of the individual participatns since no participant was required to indicate a name or a student number on any of the test materials. Many of the participants indicated their names and student numbers on the 3x5 cards, as requested, and this information was used to ascertain their ACT Composite scores from the University's records. When this information was not provided, the student's self-report of his ACT Compo- site score was used. When no ACT score was available for a participant and when a SAT score was available, a con- version table was used to equate the SAT score with the 48 corresponding ACT Composite score. Statistics and Research Design Before discussing the specific designs and statistical tests for each of the hypotheses, it is necessary to iden- tify the independent and the dependent variables. The independent variables included paternal nurturance during childhood and paternal availability during childhood. Paternal nurturance had two levels, high and low. The subjects that scored above the median on the PCR.were con- sidered high on paternal nurturance during childhood, while those subjects that scored below the median were considered low on paternal nurturance during childhood. Paternal availability also had two levels, high and low. While subjects that scored above the median on the PAS scale were considered high on paternal availability during child- hood, those subjects that scored below the median on the PAS scale were considered low on paternal availability during childhood. The dependent variables were Masculinity and Femininity. These variables were measured by the BSRIM scale and the BSRIF scale, respectively. Some of the research has indicated that intelligence affects sex-role development. In order to control for intelligence, the analysis of covariance was employed. This procedure effectively removes the error variance that is due to intelligence before the statistical test is 49 performed on the data. In this case, intelligence was the covariate (X) while sex-role score was the depen- dent measure (Y). While the ACT Composite scores were used as measures of intelligence, the sex-role scores, Masculinity for males and Femininity for females, were the dependent measures. The major weakness of the analysis of covariance is that its effects are reduced as the re- lationship between the variables declines and for this reason the analysis of variance procedure is used when the relationship between the variables is small (Cox, 1958; Kerlinger, 1973). Pearson correlation coefficients were calculated for the relationships between ACT scores and BSRIF scores; ACT scores and BSRIM scores; and, ACT scores and BSRIA scores among males and females, and these correlation co- efficients are reported in Tables I and II. Since all of the correlation coefficients were small and non-significant, the analysis of covariance procedure was disregarded in favor of the analysis of variance procedure. The hypotheses and the statistics and research design follow. The discussions of the assumptions and procedures for the power computations, analyses of variance, and the post-hoc tests, can be found in Glass & Stanley (1970), Kerlinger (1973), McSweeney & Busk (1976), and Runyon & Haber (1968). ACT ACT 50 TABLE I Pearson correlation coefficients and significance levels for ACT scores and BSRIF, BSRIM, and BSRIA scores for males. (Note: Since some subjects did not have ACT scores, N=64.) BSRIF BSRIM BSRIA 0.0080 -0.0923 0.0813 S=0.475 S=0.234 S=0.262 TABLE II Pearson correlation coefficients and significance levels for ACT scores and BSRIF, BSRIM, and BSRIA scores for females. (Note: Since some subjects did not have ACT scores, N=64.) BSRIF BSRIM BSRIA -0.0962 -0.0373 -0.0628 S=0.225 S=0.385 S=0.311 51 Hypothesis 1: Adolescent and young adult males with perceptions of having had a highly nurturant father during childhood will have higher Masculinity scores than will adolescent and young adult males with percep- tions of having had a low nurturant father during childhood. The test of this hypothesis required a comparison of the means of the Masculinity scores for the high and the low paternal nurturance groups for males. The two- way, fixed effects analysis of variance was appropriate and the appropriate test was for the Nurturance main effect. Since a difference of one-half unit between the means for the high and the low paternal nurturance groups was meaningful, and since there were 34 subjects in each group for a total of 68 subjects with alpha equal to .05, the power was .90. The analysis of variance F statis- tic (df = l, 64) was used and the decision to reject the null hypothesis in favor of Hypothesis 1 occurred when the obtained F—ratio exceeded 4.00. Hypothesis 2: Adolescent and young adult females with perceptions of having had a highly nurturant father during childhood will have higher Femininity scores than will adolescent and young adult females with per- ceptions of having had a low nurturant father during 52 childhood. The second hypothesis required similar statistical procedures as Hypothesis 1, but instead of using Mascu- linity scores as the dependent measures, the Femininity scores for females were used. Moreover, since a difference of one-half unit between the means for the high and the low paternal nurturance groups was meaningful, and since there were 36 subjects in each group for a total of 72 subjects with alpha equal to .05, the power was .91. The analysis of variance F statistic (df= 1,68) was used, and the decision to reject the null hypothesis occurred when the obtained F-ratio exceeded 4.00. Hypothesis 3: Adolescent and young adult males with perceptions of having had a combination Of high paternal nurturance and high paternal availability during childhood will have higher Masculinity scores than will adolescent and young adult makes with per- ceptions of having had other combinations of paternal nurturance and paternal availability during childhood. The test of this hypothesis requires the comparison of the mean of the high paternal nurturance-high paternal availability group with the average of the means for the remaining three groups. The hypothesis was directional and indicated that the mean Masculinity score for the 53 high paternal nurturance-high paternal availability group was meaningfully and statistically significantly greater than the average of the means for the remaining three groups. A difference of one-half unit between the largest and smallest means was meaningful and the statistical significance level was set at .05. The two-way analysis of variance was appropriate. Assuming that a difference of one-half unit between the smallest and the largest cell mean was meaningful since there were 17 subjects in each cell for a total of68 with alpha equal to .05, the power was .90. In this case the test of the Nurturance by Availability interaction was important. If the interaction was significant, then the Scheffe test was applied to the contrast 5 u = -X1 -22 -X3 + 3X4. If zero does not fall within the confidence interval, then it will be concluded that the high paternal nurturance-high paternal availability group differs significantly from the average of the remaining three groups. Hypothesis 4: Adolescent and young adult females with percep- tions of having had a combination of high paternal availability and high paternal nurturance during childhood will have higher Femininity scores than will adolescent and young adult females with 54 perceptions of having had other combinations of paternal nurturance and paternal availability. The statistical procedures for testing Hypothesis 4 were similar to those that were discussed for Hypothesis 3. In the case of Hypothesis 4, the Femininity scores for females were used instead of the Masculinity scores for males. Since there were 18 subjects in each cell, for a total of 72 subjects with alpha equal to .05, the power was .91. Hypothesis 5: Adolescent and young adult males with percep- tions of having had low paternal nurturance and high paternal availability during childhood will have lower Masculinity scores than will adoles- cent and young adult makes with perceptions of having had other combinations of paternal nur- turance and paternal availability during child- hood. The test of this hypothesis required the comparison of the means for the low paternal Nurturance - high paternal Availability group with the average of the means for the remaining three groups. Moreover, the hypothesis was di- rectional and indicated that the mean Masculinity score for the low paternal Nurturance - high paternal Availability 55 group was meaningfully and statistically significantly smaller than the average of the mean Masculinity scores for the other three groups. A difference of one-half unit between the largest and the smallest means was meaningful, the statistical significance was set at the .05 level. The two-way analysis of variance for fixed effects was appropriate. Assuming that a difference of one- half unit between the smallest and largest cell means was meaningful, and since there were 17 subjects in each cell, the power was .90. In this case, if the test of the Nurturance by Availability interaction was signifi- cant, the Scheffe test was applied to the contrast W: ~31 + 332 - i3 — 34. If zero does not fall within the confidence interval, then it will be concluded that the low-paternal Nurturance — high paternal Availability group mean differs significantly from the average of the means of the other three groups. Hypothesis 6: Adolescent and young adult females with percep- tions of having had low paternal nurturance and high paternal availability during childhood will have lower Femininity scores than will adolescent and young adult females with perceptions of having had other combinations of paternal nur- turance and paternal availability during childhood. 56 The statistical procedures for testing this hypothe- sis were similar to those that were used in Hypothesis 5. In this case, females' Femininity scores were used instead of males Masculinity scores. Since there were 18 subjects in each cell, for a total of 72 subjects with alpha equal to .05, the power was .91. Summary Chapter II considered the components of the research methods that were used in this study. There were descrip- tions of the subjects, the instruments, and the procedure, along with a final section that considered the statistics and research design that were used in this study. There were also descriptions of the research design and statistics for each of the six hypotheses. CHAPTER III RESULTS This chapter contains the results of the statistical analyses of the data that was collected for this disser- tation study. Each.hypothesis that was presented earlier was tested and the results of the tests are described. Following the descriptions of the tests of the hypotheses, there are descriptions of several additional analyses and these include some correlational analyses and analyses of other dependent measures of sex—role development. Finally, there is a summary of the chapter. Results‘of the Tests of the Hypotheses Hypothesis 1: Adolescent and young adult males with perceptions of having had a highly nurturant father during childhood will have higher Masculinity scores than will adolescent and young adult males with perceptions of having had a low nurturant father during childhood. 57 58 The 2 X 2 analysis of variance for males' BSRIM scores was used with N=68 and n=l7. An examination of Table III indicates that the Nur- turance by Availability interaction was significant and this finding cast doubt on the significance of the Nurturance main effect. According to Cox (1958) and Lindquist (1956), a main effect is not testable when the interaction is significant because the source of the apparent effect is uncertain. Hypothesis 1, therefore, was not testable. TABLE III Analysis of variance for males Masculinity (BSRIM) scores. (N=68, n=l7) Source of Sum of Mean F- Variation Squares of Square Ratio Significance Main Effects Nurturance 9.774 1 9.774 21.997 .001 Availability 0.188 1 0.188 0.424 .517 Two-way Interaction Nurturance X Avail. 2.440 1 2.440 5.491 .002 Error 28.436 64 0.444 Total 40.838 67 0.610 59 TABLE IV Descriptions of the sub-samples for males' BSRIM'scores (N=68, n=34). Standard Egogp Code Mean Deviation Variance §_ Entire Sample 4.9629 0.7807 0.6095 68 Nurturance Low 4.5838 0.7182 0.5159 34 Nurturance High 5.3421 0.6523 0.4255 34 Hypothesis 2: Adolescent and young adult females with percep- tions of having had a highly nurturant father during childhood will have higher Femininity scores than will adolescent and young adult females with perceptions of having had a low nurturant father during childhood. The 2 x 2 analysis of variance for females BSRIF was used with N=72 and n=l8. As in Hypothesis 1, the interaction effect was sig— nificant and the main effects were not testable. Hypothesis 2, therefore, was not testable. Hypothesis 3: Adolescent and young adult males with percep- tions of having had a combination of high paternal nurturance and high paternal availability 60 TABLE V Analysis of variance for females Femininity (BSRIF) scores (N=72, n=l8.) Source of Sum of Mean F- . Variation Squares of Square Ratio Significance Main Effects Nurturance 7.821 1 7.821 20.196 .001 Availability 0.104 1 0.104 0.267 .607 Two-way interaction Nurturance X Availability 2.057- 1 2.057 5.312 .014 Error 26.333 68 0.387 Total 36.315 71 0.511 TABLE VI Descriptions of the sub—samples for females BSRIF scores (N=72, n=36) Standard Group Code Mean Deviation Variance E Entire Sample 5.0607 0.7152 0.5115 72 Nurturance Low 4.7311 0.6754 0.4561 36 Nurturance High 5.3903 0.5983 0.3580 36 61 during childhood will have higher Masculinity scores than will adolescent and young adult ‘males with perceptions of having had other combinations of paternal nurturance and pater! nal availability during childhood. Once again, the analysis of variance procedure was used for males, Masculinity (BSRIM) scores with N=68 and n=17, and favorable results were obtained. The test for the Nurturance by Availability interaction effect was applied (df=l, 64) and the obtained F—ratio was 5.491. This F—ratio was significant with p<.022. (See Table III for the results of the analysis of variance for males Masculinity scores.) Moreover, Scheffé's test for com- plex comparisons was applied to the specified contrast; W= -Xlai2-§3 +3)?4 with df=l, 64 and the result indicated that the hypothesis was supported at the 3:0.05 level. (See Table VII for the results of the Scheffe tests.) Hypothesis 4: Adolescent and young adult females with.per— ceptions of having had a combination of high paternal nurturance and high paternal avail— ability during childhood will have higher Femininity scores than will adolescent and young adult females with perceptions of having had other combinations of paternal nurturance and paternal availability during childhood. 62 The test of Hypothesis 4 required the application of the analysis of variance procedure and the Scheffe pro- cedure for females' Femininity scores with N=72 and n=l8. The test for the Nurturance by Availability interaction effect was applied (df=l, 68), and the obtained F-ratio was 5.312. This F-ratio was significant with p<.024. Scheffé's test for complex comparisons was applied to the specified contrast T= -Xl-X -X3 +35?4 with df=l, 68 and the result indicated that the hypothesis was supported at the 3=0.05 level. (See Table VII for the results of the Scheffe tests.) Hypothesis 5: Adolescent and young adult males with percep- tions of having had low paternal nurturance and high paternal availability during childhood will have lower Masculinity scores than will adoles- cent and young adult males with perceptions of having had other combinations of paternal nur- turance and paternal availability during childhood. In the case of Hypothesis 5, the analysis of variance procedure for males' Masculinity scores was used in con- junction with the Scheffe test for complex comparisons and favorable outcomes were obtained. Previously, it was re- ported that the Nurturance by Availability interaction was significant and the Scheffe test was applied to the 63 contrast W= -X1+ 3X2-X3 +33?4 . The Scheffe test indicated that the Masculinity scores of the males in the low- Nurturance - high Availability group were significantly lower than were the Masculinity scores of the other three males groups combined and this result was supported at the 8=0.05 level. (See Table VII for the results of the Scheffe tests.) Hypothesis 6: Adolescent and young adult females with per- ceptions of having had low paternal nurturance and high paternal availabiltiy during child- hood will have lower Femininity scores than will adolescent and young adult females with per- ceptions of having had other combinations of paternal nurturance and paternal availability during childhood. The test of Hypothesis 6 required the use of the analysis of variance procedure for females' Femininity scores in conjunction with the Scheffe test for complex comparisons and the results indicated support for Hypothe- sis 6. It was reported in Table V that the Nurturance by Availability interaction effect was significant and Scheffé's procedure was performed using the specified 2-23-X4. The difference between the mean for the low-Nurturance-higheAvailability group and contrast W= -il +32 the mean for the other three groups combined was significant 64 at the 3:0.05 level. (Table VII includes the result of this test.) TABLE VII Results of the Scheffe tests for the complex comparisons that were used in Hypotheses 3 through 6. Significance Hypothesis Effect (Units) Level 3 2.4847 0.05 4 2.1461 0.05 5 -2.0634 0.05 6 -1.8428 0.05 Additional Analyses Pair Comparisons for BSRIM Scores Amopg Males Previously, it was reported that there was a signifi- cant Nurturance by Availability interaction effect for BSRIM scores and it was necessary to make a number of post- hoc comparisons to interpret the meaning of the interaction. To this end, six pair comparisons of cell means were for- mulated and tested by the Tukey procedure at the 3:0.05 level and the results of these tests indicated that while 65 three of the tests of the pair comparisons were signifi- cant, the remaining three tests of pair comparisons failed to reach statistical significance at the 3=0.05 level. The findings showed that males in the high-Nur— turance-high-Availability group were significantly higher on Masculinity scores than either the males in the low- Nurturance-low-Availability group. Moreover, males in the high-Nurturance-low-Availability group scored sig- nificantly higher on Maculinity than did males in the low-Nurturance-high-Availability group. The results of these post-hoc analyses are reported in Table VIII. Pair Comparisons for BSRIF Scores Amonngemales Similarly, the finding of a significant Nurturance by Availability interaction effect for BSRIF scores among females required additional tests for interpreting the meaning of the interaction and Tukey tests were performed on six pair comparisons. The results of these Tukey tests indicated that while three of the differences were signifi- cant, the remaining three were not significant at a=:os and while the significant comparisons among the females corresponded with the significant comparisons among the males, the non-significant comparisons among the females corresponded with the non-significant comparisons among the males. The results of the pair comparisons for the BSRIF 66 scores among females appear in Table IX. TABLE VIII Results of the Tukey Tests of BSRIM Scores Among Males (N=68, n=l7). Difference Difference Comparison* (Units) (SD) Result 21 - 22 0.2736 0.3504 Not Sig.,a=.05 i3 - i1 0.3794 0.4859 Not Sig.,3=.05 x4 - 23 0.4841 0.6200 Not Sig.,3=.05 23 - 22 0.6529 0.8363 519., =.05 i4 - 21 0.8635 1.1060 319., a=.05 24 - 22 1.1131 1.4257 Sig., a=.05 * These are comparisons of cell means. X relates to low- Nurturance-low-Availability. X is low-Nurturance-high- Availability. X is high-Nurturance-low-Availability and X4 is high-Nurtufance-high-Availabiltiy. Correlational Analyses The correlational analyses were performed to ascertain the relationships among all of the variables in the study and the results of these analyses indicated that a number of significant relationships existed among several of the variables. The Pearson correlation coefficients were cal- culated for all of the possible pair relationships of variables and these results are reported for males and females 67 TABLE IX Results of the Tukey Tests of BSRIF Scores Among Females (N=72, n=18). Difference Difference Comparison (Units) (SD) Result 21 - 22 0.2622 0.3666 Not 319., at a=.05 i3 - 31 0.3211 0.4489 Not Sig., at a=.05 i4 - 23 0.4319 0.6038 Not Sig., at a=.05 i3 - 22 0.5833 ' 0.8155 319., a=.05 i4 - 21 0.7350 1.0276 Sig., a=.05 i4 - 22 0.9972 1.3942 319., a=.05 in Tables X and XI, respectively. Among males, four significant Pearson correlation co- efficients were obtained (p<.05). Significant relation- ships were found between Nurturance (PCR) and Masculinity (BSRIM) and between Nurturance (PCR) and Androgyny (BSRIA). In the former case, r=0.6627 (p<.001) while in the later case r=-0.6293 (p<.001). Two additional findings indicated that Femininity (BSRIF) was significantly, positively correlated with Androgyny, with r=0.5548 (p<.001) and that Masculinity was significantly, negatively correlated with Androgyny with r=-0.9076 (p<.001). None of the other correlation coefficients approached significance. 68 TABLE X Pearson Correlation Coefficients for Males (Coefficient/Cases/Significance) ACT PCR PAS BSRIF BSRIM BSRIA ACT 1.0000 0 .001 PCR -.1953 1.0000 64 0 .061 .001 PAS .0538 .1144 1.0000 64 68 0 .336 .176 .001 BSRIF .0080 -.1652 .0052 1.0000 64 68 68 0 .475 .089 .483 .001 BSRIM -.0923 .6627 .0920 -.1627 1.0000 64 68 68 68 0 .234 .001 .228 .092 .001 BSRIA .0813 -.6293 -.0816 .5548 -.9076 1.0000 64 68 68 68 68 0 .262 .001 .254 .001 .001 .001 Five significant relationships were found for females. There were significant positive relationships between Nur- turance and Femininity with r=0.5328 (p<.001) and between Nurturance and Masculinity with r=0.5252 (p<.001). More- over, significant positive relationships existed between Femininity and Masculinity with r=0.6319 (p<.001) and be- tween Femininity and Androgyny with r=.4635 (p<.001). Finally, there was a significant negative relationship 69 between Masculinity and Androgyny with r=-0.3923 (p<.001). None of the other correlation coefficients for females were significant. BSRIF BSRIM BSRIA Descriptive Statistics for The SPSS subprogram Condescriptive was used to ob- Pearson Correlation Coefficients for Females. TABLE XI (Coefficient/Cases/Significance) ACT 1.0000 0 .001 .1320 64 .149 .0827 64 .258 -.0962 64 .225 -.0373 64 .385 -.0628 64 .311 PCR 1.0000 0 .001 .0459 72 .351 .5328 72 .001 .5252 72 .001 .0345 72 .387 PAS BSRIF BSRIM 1.0000 0 .001 -.0720 1.0000 72 0 .274 .001 -.1803 .6319 1.0000 72 72 0 .065 .001 .001 .1194 .4635 -.3923 72 72 72 .159 .001 .001 Variables BSRIA 1.0000 0 .001 tain the descriptive statistics such as means, minimums, 70 maximums, variances, and standard deviations for variables PCR, PAS, BSRIF, BSRIM, and BSRIA for both males and fe- males and these values are presented in Tables XII and XIII, respectively. TABLE XII Descriptive Statistics for Variables Among Males (N=68). Standard Variable Minimum Mean Maximum ‘Variance Deviation PCR 41.000 65.897 79.000 89.736 9.473 PAS 21.000 34.956 50.000 57.088 7.556 BSRIF 3.200 4.619 5.400 0.159 0.398 BSRIM 3.000 4.963 6.600 0.610 0.781 BSRIA -2.300 -0.357 2.350 0.899 0.948 TABLE XIII Descriptive Statistics for Variables Among Females (N=72). Standard Variable Minimum Mean Maximum Variance Deviation PCR 43.000 68.403 80.000 89.737 9.473 PAS 18.000 35.319 50.000 55.882 7.475 BSRIF 3.000 5.061 6.550 0.511 0.715 BSRIM 2.500 4.577 6.250 0.483 0.695 BSRIA -0.800 0.481 1.650 -0.378 0.615 71 Self-Ratings of Males and Females on BSRI Masculine and Feminine Items, Respectively Tables XIV and XV show the mean self-ratings of males and females, respectively, on BSRI Masculine and Feminine items, respectively, and these self-ratings indicate that there were differences among the groups as a function of paternal nurturance and paternal availability during childhood. For example, males differed most on items such as 'acts as a leader', 'aggressive', 'athletic', 'compet- itive', 'dominant', 'forceful', and 'has leadership cap- abilities', with high-Nurturance-high-Availability males having substantially higher mean self-ratings than low- Nurturance-high-Availability group on items such as 'affec- tionate', 'compassionate', 'eager to soothe hurt feelings', 'feminine', 'sensitive to the needs of others', 'sympa- thetic', and 'understanding'. Analyses of Variance and Post-Hoc Procedures for Other Dependent Measures In addition to the previously reported analyses of variance, four additional analyses of variance and a number of post-hoc tests were performed on the data to measure the effects of paternal nurturance and paternal availability during childhood upon other measures of sex- role development. While the dependent measures for males 72 TABLE XIV Mean Self—Ratings of Males on Masculine Items As A Function of Father-Nurturance and Father- Availability (N=68). Masculine Lo Nurt. Lo Nurt. Hi Nurt. Hi Nurt. Items Lo Avail. Hi Avail. Lo Avail. Hi Avail. Acts as a Leader 4.0 . . Aggressive 4.3 . . Ambitious 4.8 . . Analytical 4.9 . . . Assertive 4.8 . . . Athletic 4.8 . . . Competitive . Defends own Beliefs 4.9 . . . Dominant 4.3 . . . Forceful 4.5 . . Has Leadership Abilities 4.5 . . Independent 4.9 . Individualis- tic 4.9 5.4 5.0 5.2 Makes Decisions Easily 4.8 . 5.0 . Masculine 4.8 . 5. . Self-Reliant 4.8 . 5. Self- Sufficient 4.8 4.7 5.0 5.3 Strong Personality 4.8 4.3 4.9 5.5 Willing to Take a Stand 4.9 4.3 5.1 5.7 Willing to Take Risks 5.0 4.8 5.0 5.5 73 TABLE XV Mean Self- Ratings of Females on Feminine Items As A Function of Father—Nurturance and Father-Availability (N=72). Feminine Lo Nurt. Lo Nurt. Hi Nurt. Hi Nurt. Items Lo Avail. Hi Avail. Lo Avail. Hi Avail. Affectionate 5.1 '4.4 5.3 5.7 Cheerful 4.8 4.3 5.4 5.5 Childlike 4.7 4.9 5.0 5.5 Compassionate 5.4 4.3 5.3 5.8 Does not use harsh language 4.4 4.4 5.0 5.3 Eager to soothe hurt feelings 5.1 4.4 5.4 Feminine 5.1 4.6 5.4 5.7 Flatterable 4.8 4.8 5.1 5.5 Gentle 4.9 4.7 5.2 5.5 Gullible 4.7 4.7 5.0 5.5 Loves . Children 4.9 4.6 5.4 5.6 Loyal 4.8 4.8 5.1 Sensitive to the needs of others 5.1 4.4 5.4 5.9 Shy 4.8 4.9 5.1 Soft spoken 4.8 4.8 5.0 5.5 Sympathetic 5.2 4.4 5.3 5.8 Tender 4.8 4.7 5.0 5.6 Understanding 5.3 4.5 5.6 5.8 Warm 5.0 4.7 5.3 5.6 Yielding 4.5 4.7 5.1 5.4 74 included BSRIF scores and BSRIA scores, the dependent measures for females were BSRIM scores and BSRIA scores and the results of these analyses are reported and tabled in the two sections that follow. Analyses of Variance and Post-Hoc Procedures for BSRIF Scores and BSRIA Scores Among Males Two analyses of variance were performed for males' BSRIF scores and BSRIA scores and the results indicated that while there were no significant differences for the main effects or for the interaction effect in the case of the BSRIF scores, there were some significant findings for the BSRIA scores and these results are reported in Tables XVI and XVII. The analysis of variance for the BSRIA scores showed that the Nurturance by Availabiltiy interaction was significant with an obtained F-ratio of 5.953 (df=1, 64; p<.017). Since the interaction was sig- nificant, the apparent significance of the Nurturance effect was doubtful. Tables XVI and XVII follow and these tables summarize the results of the analyses of variance for BSRIF scores and BSRIA scores, respectively, among males. In order to determine the meaning of the significant interaction of Nurturance and Availability for BSRIA scores among males, it was necessary to perform several post-hoc procedures including Tukey tests of the six pair 75 TABLE XVI Analysis of Variance of BSRIF Scores for Males (N=68, n=l7). Source of Sum of Mean F- Variation Squares of Square ratio Main Effects Nurturance 0.017 1 0.017 0.105 Availability .000 l .000 0.001 Two-way inter- action Nurt.X Avail. 0.204 1 0.204 1.250 Error 10.415 64 0.163 Total ‘ 10.636 67 0.159 TABLE XVII Analysis of Variance of BSRIA Scores Males (N=68, n=l7). Source of Sum of Variation Squares Main Effects Nurturance 9.488 Availability 0.235 Two-way inter- action Nurt.X Avail. 4.300 Error 46.233 Total 60.256 g; 64 67 Mean F- Square ratio 9.488 13.134 0.235 0.326 4.300 5.953 0.722 0.899 Significance 0.747 0.971 0.268 for Significance 0.001 0.570 0.017 76 comparisons of means and two Scheffe tests of complex comparisons of means and the results of these post-hoc procedures indicated that while two pair comparisons of means were significant, the two complex comparisons of means were also significant. *The low-Nurturance-low- Availability group by 0.8647 unit and this difference was significant at 3=.05. The low-Nurturance-high-Avail- ability group differed significantly (a=.05) from the high-Nurturance-high-Availability group by 1.2500 units. Scheffé's procedure was used to test the significance of the two complex comparisons and both of the complex comparisons were significant at a=.05. The first con- trast compared the mean for the high-Nurturance-high- Availability group with the average of the means of the other three groups. The contrast was W= -X1-X2-X3+3X4 and the result indicated that the mean for the high- Nurturance-high-Availability group was significantly greater than the average of the means of the other three groups. The second contrast compared the mean for the ' low-Nurturance-high-Availability group with the average of the means of the remaining three group. The contrast was Y=-Xl+3X2-X3-X4 and the result indicated that the mean for the low-Nurturance-high-Availability group was sig- nificantly less than the average of the means for the remaining three groups. A summary of the results of the post-hoc procedures is reported in Table 18. 77 TABLE XVIII Results of the Post-Hoe Procedures for BSRIA Scores Among Males (N=68, n=l7). Comparison Difference Test Result 22 - 21 0.3853 Tukey Not Sig., at a=.05 i3 - 22 0.2441 Tukey Not 519., at a=.05 X4 - X1 0.8647 Tukey Significant at a=.05 i4 - 22 1.2500 Tukey Significant at a=.05 x3 - 22 0.6294 Tukey Not 819., at a=.05 i4 - 23 0.6206 Tukey Not 519., at a=.05 il-iz-x3+3i4 -2.7353 Scheffe Significant at a=.05 il+322-i3-i4 2.2647 Scheffe Significant at a=.05 Analyses of Variance and Post-Hoc Procedures for BSRIM Scores and BSRIA Scores Among Females The results of the analysis of variance for BSRIM scores and BSRIA scores among females indicated that there were some Significant effects in both analyses. While the analysis of variance of BSRIM scores resulted in a sig- nificant Nurturance main effect, none of the other sources of variation in this analysis approached significance. Moreover, the test for the NurturanCe main effect (df=1,68) yielded an F-ratio of 24.782 (p<.001) and this indicated that the difference of 0.7122 unit or 1.0247 standard 78 deviations between the high and the low Nurturance groups that favored the high-Nurturance group was meaningfully significant. While the analysis of variance of the BSRIA scores showed that the test of the Nurturance by Avail- ability interaction yielded an F-ratio of 6.315 (df=1,68; p<.014), none of the other sources of variation were significant. In order to interpret the meaning of the interaction of Nurturance and Availability a number of post-hoc procedures were performed on the BSRIA data and the results of these analyses indicated that there was one significant complex comparison of cell means. Scheffé's procedure was applied in conjunction with the contrast W =Xl-X2-X3-3X4 and this contrast compared the mean BSRIA. score for the high-Nurturance-high-Availability group with the average of the mean BSRIA scores for the remain- ing three groups. The finding indicated that the mean BSRIA score for the high-Nurturance-high-Availability group was significantly higher than the average of the BSRIA score means for the remaining three groups and the difference was 0.8989 unit or 1.4629 standard deviations. Tables XIX,XX, and XXI follow and these tables contain the results of these tests. 79 TABLE XIX Analysis of Variance for BSRIM Scores Among Females (N=72, n=18). Source of Sum of Mean F- Variation Squares of Square ratio Significance Main Effects Nurturance 9.138 1 9.138 24.782 .001 Availability 0.083 1 0.083 0.226 .636 Two-way inter- action Nurt.X Avail. 0.000 1 0.000 0.001 .977 Error 25.073 68 0.369 Total 34.295 71 0.483 TABLE XX Analysis of Variance for BSRIA Scores Among Females (N=72, n-18). Source of Sum of Mean F- Variation Squares df Square ratio Significance Main Effects Nurturance 0.032 1 0.032 0.090 .765 Availability 0.344 1 0.344 0.968 .329 Two-way inter- action Nurt.X Avail 2.247 1 2.247 6.315 .014 Error 24.197 68 0.356 0.356 Total 26.821 71 0.378 80 TABLE XXI Results of the Post-Hoc Tests for BSRIA Scores Among Females (N=72, n=18). Comparison Difference Top; Results 22-'1 0.2150 Tukey Not Sig., at a=.05 23-21 0.3955 Tukey Not Sig., at a=.05 21-24 0.0962 Tukey Not Sig., at 3=.05 24-22 0.3112 Tukey Not Sig., at a=.05 XZ-XB 0.1805 Tukey Not Sig., at 3=.05 24-23 0.4917 Tukey Not Sig., at a=.05 ~21-22-23+3x4 0.8989 Scheffe Significant at a=.05 -il+iz-i3-i4 —.3456 Scheffé Not Sig., at a=.05 Additional Descriptive Statistics The tables that follow contain the descriptive sta- tistics for BSRIF, BSRIM, and BSRIA scores for males and for females. 81 TABLE XXII Descriptive Statistics for BSRIF Scores Among Males (N=68, n=l7). Group Entire Group Nurturance Availability Availability Nurturance Availability Availability Availability Availability Code Mean 4.6188 Low 4.6029 Low 4.5500 High 4.6559 High 4.6347 Low 4.6912 High 4.5782 Low 4.6206 High 4.6171 Standard Deviation 0.3984 0.4589 0.4724 0.4531 0.3335 0.2313 0.4112 0.3732 0.4278 TABLE XXIII Variance 0.1588 0.2106 0,2231 0.2053 0.1112 0.0535 0.1691 0.1393 0.1830 34 17 17 34 17 17 34 34 Descriptive Statistics for BSRIM Scores Among Males (N=68, n=l7). Standard Group Code Mean Deviation Variance §_ Entire Sample 4.9629 0.7807 0.6095 68 Nurturance Low 4.5838 0.7182 0.5159 34 Availability Low 4.7206 0.4157 0.1728 17 Availability High 4.4471 0.9227 0.8514 17 Nurturance High 5.3421 0.6523 0.4255 34 Availability Low 5.1000 0.4161 0.1731 17 Availability High 5.5841 0.7615 0.5799 17 Availability Low 4.9103 0.4526 0.2048 34 Availability High 5.0156 1.0134 1.0270 34 82 TABLE XXIV Descriptive Statistics for BSRIA Scores Among Males (N=68, n=l7). Group Entire Sample Nurturance Availability Availability Nurturance Availability Availability Availability Availability Code Low Low High High Low High Low High Standard Mean Deviation -0.3574 0.9483 ‘0.0162 0.9973 -0.l765 0.6897 0.2088 1.2235 -0.7309 0.7374 -0.4206 0.5365 -l.0412 0.7932 -0.2985 0.6209 -0.4l62 1.1972 TABLE XXV Variance 0.8993 0.9947 0.4757 1.4969 0.5438 0.2878 0.6291 0.3855 1.4333 N— 68 34 17 17 34 17 17 34 34 Descriptive Statistics for BSRIF Scores Among Females (N=72, n=18). Group Entire Group Nurturance Availability Availability Nurturance Availability Availability Availability Availability Code Low Low High High Low High Low High Mean 5.0607 4.7311 4.8622 4.6000 5.3903 5.1833 5.5972 5.0228 5.0986 Standard Deviation 0.7152 0.6754 0.5976 0.7386 0.5983 0.3581 0.7198 0.5121 0.8788 Variance 0.5115 0.4561 0.3571 0.5456 0.3580 0.1282 0.5181 0.2622 0.7724 36 18 36 18 36 36 83 TABLE XXVI Descriptive Statistics for BSRIM Scores Among Females (N=72, n-18). Standard Group Code Mean Deviation Variance Entire Sample 4.5771 0.6950 0.4830 Nurturance ' Low 4.2208 0.6191 0.3833 Availability Low 4.2528 0.5183 0.2687 Availability High 4.1889 0.7200 0.5184 Nurturance High 4.9333 0.5792 0.3354 Availability Low 4.9694 0.4612 0.2127 Availability High 4.8972 0.6893 0.4751 Availability ' Low 4.6111 0.6049 0.3659 Availability High 4.5431 0.7820 0.6116 36 18 18 36 18 18 36 36 .TABLE XXVII Descriptive Statistics for BSRIA Scores Among Females Standard Group Code Mean Deviation Variance Entire Sample 0.4804 .0.6146 0.3778 Nurturance Low 0.5019 0.4532 0.2054 Availability Low 0.6094 0.4517 0.2041 Availability High 0.3944 0.4409 0.1944 Nurturance High 0.4597 0.7483 0.5600 Availability Low 0.2139 0.5625 0.3164 Availability High 0.7056 0.8417 0.7085 Availability Low 0.4117 0.5413 0.2930 Availability High 0.5500 0.6808 0.4634 84 SummaEy This chapter contained the results of the tests of the hypotheses and of several additional analyses. The findings indicated support for most of the hypotheses. A number of correlation coefficients were reported and some of these indicated that significant relationships existed among several of the variables. Additional analyses of variance of other dependent variables demonstrated that there were some significant main effects and interactions. CHAPTER IV DISCUSSION Chapter IV includes the discussions of the results that were presented in Chapter III along with some dis- cussions of other issues. The results supported four of the hypotheses with.some more support from the add- itional findings and this information is included in separate discussions of each of the main effects and the two-way interaction. The information in Chapter III in! dicated that intelligence had little relationship with sex-role develOpment among college students and this finding is discussed following the discussion of the in- teraction. These discussions are followed by sections that consider the limitations of the study, the conclusions, and some recommendations for future research. Paternal Nurturapce During Childhood and Sex-Role Development Although it was not possible to test the first two hypotheses, this study indicated that paternal nurturance is necessary for sex-role development among male and female college students. Males and females that reported having 85 86 highly nurturant fathers during childhood scored higher on measures of Masculinity and Femininity, respectively, than did males and females that reported having less nur- turant fathers during childhood. Additional support is provided by the findings of significant positive rela- tionships between PCR Love-Reject factor scores and BSRIM scores for males and between PCR Love-Reject factor scores and BSRIF scores for females. Moreover, males in the high-Nurturance group had more negative BSRIA scores than did the males in the low-Nurturance group. This result indicates that the males that reported having highly nur- turant fathers during childhood had more masculine scores than did the males that reported having less nurturant fathers during childhood. The results of this study have supported the role theorists' position that while fathers facilitate instru- mental behaviors in their sons, they encourage the develop- ment of expressive behaviors in their daughters and this effect is apparent upon observing and comparing the BSRIM responses of the high-Nurturance males with the BSRIM re— sponses of the low-Nurturance males and after making simi- lar observations and comparisons for BSRIF scores among females. Males in the high-Nurturance groups made higher self-ratings on all but one of the twenty Masculinity scale items with the high-Nurturance males scoring sub- stantially higher than the low-Nurturance males on such 87 instrumental items as leadership, aggression, ambition, athletism, competition, dominance, and forcefulness. It is interesting that there was a small difference favoring the low-Nurturance males on 'individuality' and this suggests that the low-Nurturance males are at least as, and perhaps more individualistic than the high-Nurturance males. Females in the high-Nurturance groups rated them- selves higher on all Femininity scale items than did fe- males in the low-Nurturance groups with the high-Nurturance females scoring substantially higher than the low-Nurturance females on such expressive items as affection, compassion, sensitivity, and sympathy. There is evidence for the stability of the effects of paternal nurturance during childhood upon sex-role development. The previously discussed research indicated that paternal nurturance during early childhood facilitated masculinity in young males and femininity in young females and the present study has shown that adult males and adult females with perceptions of having a highly nurturant father during childhood had higher Masculinity and Femininity scores, respectively, than did males and females that re- ported having less nurturant fathers during childhood. Apparently, early paternal behavior toward the child has a lasting effect upon the child's sex-role development. 88 Paternal Availability During Childhood and Sex-Role Development The results of this study indicated that father avail- ability alone during childhood did not affect sex-role development among college students. While there were no significant differences between the high-Availability and low-Availability male groups on Masculinity scores there were also no significant differences between the high- Availability and the low-Availability female groups on Femininity scores. Moreover, while there were no sig- nificant differences between the high-Availability and the low-Availability male groups on BSRIF scores and on BSRIA scores, there were also no significant differences between the high-Availability and the low-Availability female groups on BSRIM scores and on BSRIA scores. While the finding of no differences in sex-role de- velopment as a function of paternal availability alone during childhood seems to conflict with the findings of the research on the effects of father-absence upon sex-role development, the apparent conflict is resolved after con- sidering the differences in the early family environements of the populations. The father-absence studies compared individuals from intact families with individuals from single parent families and these two kinds of families provide different environments for children. Moreover, 89 the research on father-absence usually compares the sex- role development of individuals from homes where the father is usually present from birth until the time of the study with the sex-role development of individuals whose fathers terminally departed at some time before the collection of the data. The present study considered only individuals from intact families and these individuals had access to both parents from birth until at least age sixteen. Conceivably, the differences in sex-role development between father- present and father-absent individuals reflect differences between intact families and single parent families while the lack of differences among individuals with highly available fathers and individuals with less available fathers reflects the inefficacy of the father's availability alone in facilitating sex-role development. Paternal Nurturance and Paternal Availability During Childhood and Sex-Role Development The results indicated that a father who is both highly nurturant and highly available will be more effective in facilitating sex-role development among his children than will a father with any other combination of Nurturance and Availability. Apparently, fathers that are both highly nurturant and highly available provide both a high quality and a high quantity of nurturance toward their children. 90 Moreover, the higher availability of these fathers during childhood serves to compliment the effects of paternal nurturance during childhood and this has the most en- hancing effects upon sex-role development among their children as adults. On the other hand, a less nurturant and highly available father will have the most damaging effects upon the sex-role development of his children. The fathers that are both less nurturant and highly available provide a higher quantity of non-nurturant relationships for their children during childhood and these experiences have a negative effect upon sex-role development which is appar- ent among their children as adults. Moreover, the higher availability of these fathers serves to intensify the effects of paternal non-nurturance during childhood and this has the most damaging effects upon sex-role develop- ment among their children as adults. The significant differences of several of the pair comparisons for males and females on Masculinity and on Femininity scores, respectively, provides additional sup- port for the interaction effect and also demonstrates the differences in the magnitudes of the effect for varing combinations of paternal nurturance and paternal availability during childhood for both males and females. The greatest difference was for the comparison of the high-Nurturance- high-Availability group with the low—Nurturance-high- 91 availability group. Although there was less difference between the highONurturance-high-Availability group and the low-Nurturance-low-Availability group, the least Sig- nificant difference was found for the comparison of the high-Nurturance-low-Availability group with the low-Nurtur- ance-high-Availability group. The results of the tests among the Androgyny scores indicated a significant Nurturance by Availability inter- action effect for both males and females and the results of several post-hoc tests reflect differences in the mag- nitudes of the effects in some cases. The significance of two complex comparisons for males were consistent with the results of two hypothesized comparisons and these re- sults supported the hypotheses that highly nurturant and highly available fathers are more effective in facilitating Masculinity in their sons than are fathers with other com- binations of Nurturance and Availability. Moreover, while the comparisons for males of the high-Nurturance-high- Availability group with the low-Nurturance-high-Availability was significant and yielded the greatest effect, the comparison of the high-Nurturance-high-Availability group with the low-Nurturance-low-Availability group was also significant and produced the second largest effect. Al- though none of the other differences for the pair com- parisons among males were significant, differences were apparent and the magnitudes of the effects were in the same order as the magnitudes of the effects for the Mascu- linity pair comparisons. Among females, the tests for 92 effects upon Androgyny produced a significant difference for one complex comparison and this showed that highly nurturant and highly available fathers were more effec— tive in facilitating the development of a feminine sex- role among their daughters than were fathers with different combinations of Nurturance and Availability during their children's early years of life. Although the remaining complex comparisons were not statistically significant, differences were apparent and the order of magnitudes of these differences were dissimilar to the order of the magnitudes of the corresponding differences among males. The dissimilarity between the results for the male and for the female groups on Androgyny may be attributed to the differences between the male group and the female group with respect to the results of the analyses of variance for Femininity and for Masculinity scores. There were no effects on Femininity scores among males, but females Femininity scores were affected by both_Nurturance and by the interaction of Nurturance and Availability. Males' Masculinity scores were affected by Nurturance and the interaction of Nurturance and Availability; females' Masculinity scores were affected by Nurturance. The dif- ferential trends with respect to the effects of Nurturance and the interaction of Nurturance and Availability upon males' and females' Masculinity and Femininity scores and 93 the fact that the Androgyny measure is a difference score between one's Femininity score and one's Masculinity score may be responsible for the dissimilar trends that were observed for Androgyny. Intelligence and Sex-Role Development The results of this study indicated that there was little relationship between ACT Composite scores and any of the measures of sex-role development for both males and females and these findings are consistent with some of the previous research. Kohlberg (1966) reported that there was a small negative relationship between IQ and measures of sex-role among college students and this study demonstrated a similar relationship with a similar sample. Apparently, while intelligence has little effect upon sex-role development among college students, this study does not permit any generalizations beyond this group. College students are not representative of the population of late adolescents and young adults since college students must meet an admissions standard and this standard usually assures that most college students are at least of aver- age intelligence. Moreover, college students differ from the population of late adolescents and young adults on a number of other relevant characteristics such as race and socio-economic status. 94 Limitations of the Study Although the results of this study supported the hypotheses and the role theorists' position, there were some limitations and criticisms that apply to the study and the first limitation is a result of the study's sample. The participants in the study included undergraduates from one large midwestern university and these individuals are not representative of the population of adolescents and young adults. A number of qualifications which were necessarily imposed upon the sample had the effect of re— stricting the external validity or the generalizability of the findings. As a result of this procedure, the re- sults of the study are generalizable to white, middle and upper-middle class, first-born, American, college students. Judging from the response rate that was obtained during the collection of the data, fewer than twenty percent of the college population meet all of these qualifications. The findings of this study may also be restricted to the population of individuals that were reared during the 1950's and the 1960's. The participants perceptions of their relationships with their fathers and their fathers' availability represented reports of experiences that occurred during those two decades. The accuracy of adults as reporters of their early childhood experiences must be questioned. The present study used£1retrospective method to ascertain information about 95 some aspects of the father-child relationship during early childhood among adolescent and young adult college students. It is possible that these individuals may not have accurately reported their fathers' behaviors toward them as children. These possible inaccuracies may be attributed to the participants' memories of early child- hood experiences which may have been affected by the passage of time. One must also consider the instrumentation of the study. The series of questionnaires that were used were restricted in the nature and in the number of items that were available and although the items in the quesionnaires required numerical ratings, checks, or Short answers, these brief responses may not accurately represent the father-child relationship, and the father's availability. Moreover, since the amount of time that was available for responding was limited, the number of items that were in- cluded in the questionnaires were limited and this may have affected the reliability of the scores. The final criticism concerns the study's failure to control for a variety of surrogate father figures outside the home that may have been available to the participants. While the study effectively controlled for the presence of surrogate fathers within the home by restricting the sample to first born and only children and by not considering individuals with older males (aside from the father) at 96 home, it was not possible to control for the effects of older adult males that many of the participants may have contact with throughout their lives. Conclusions of A number of conclusions can be made on the results this study and these conclusions follow. Paternal nurturance during childhood is important for sex-role development among adults. The effects of paternal nurturance toward children during childhood are intensified by high paternal availability during childhood. Fathers that are both highly nurturant and highly available to their children during childhood are more effective in facilitating sex-role development among their children than are fathers that are low on nurturnace and either highly available or low on availability. Fathers that are low on nurturance and highly available to their children during childhood have a damaging effect upon sex-role development among their children. 97 Recommendations for future research The previous research and the findings of this dissertation indicate that there is a need for additional research. Longitudinal research could consider the stability of the effects of the father-child relationship and father availability during childhood upon sex role development. The research could examine a wide group of individuals at several times over their life spans and assess the effects of early father-child relations and father availability upon sex-role development. Future research should treat gender as an independent variable. Studies could form groups on combinations of paternal nurturance and paternal availability during childhood and compare males' sex-role development with females sex-role development. Recently, the concept of androgyny has been discussed in the literature, but there have not been any studies that consider the effects of the father upon the develop- ment of androgyny. Future research could consider the effects of paternal nurturance and paternal availability upon the development of androgyny. There is also a need for studies that consider the effects of surrogate father-child relationship and surrogate father availability during childhood upon 98 sex-role development. These studies should consider surrogate fathers in the forms of older siblings, other relatives, or any other male adults that are available to children. The roles of parents and particularly the role of the mother in the family may be changing and these changes indicate that researchers should examine the effects of both parents upon sex-role among children growing up today. While the earlier research indicated that fathers usually had greater influence than did mothers upon the sex-role develOpment of their children, these findings are based upon data for an earlier period and the research on contemporary families may provide different findings. It is also necessary to study the utility of the existing instruments that measure parent-child relation- ship, parent availability, and sex-role development. These variables are affected by social changes and the instruments that are designed to assess these variables may not be accurate in the future. APPENDICES APPENDIX A Letters of Permission MICHIGAN STATE UNIVERSITY 99 OFFICE OF INSTITUTIONAL RESEARCH EAST LANSING ° MICHIGAN ' 48824 ADMINISTRATION BUILDING August 11, 1978 Mr. Edward T. Manzitti Doctoral Candidate Educational Psychology Desk 4MM - Erickson Hall Campus Dear Mr. Manzitti: As Chairman of the University Committee on Release of Confidential Information, I am responding to your letter of August 1, 1978 in which you request access to students' SAT quantitative and verbal scores or ACT scores in connection with a research study conducted as part of your doctoral requirements. You will note that copies of this letter to you are going to Assistant Registrar Phyllis Wilkie and to Prof. Arvo Juola as well as to your adviser. I am sending these two persons a copy of your letter so that they will be able to relate my response to your original request. Acting for the Committee, I approve of your request for the use of these test scores. The conditions which you have set up for their use seem adequate to safeguard the information and the approval is contingent upon your following these procedures. I believe that Prof. Juola can arrange for you to have access to these scores when you reach that stage. However, it is probable that in getting to the combination of data that you need you will need to talk with both Assistant Registrar Wilkie and Prof. Juola. Sincerely yours, ./ Paul L. Dressel, Chairman Committee on Release of Confidental Information cc: Phyllis Wilkie Arvo Juola Louise Sause MICHIGAN STATE UNIVERSITY 10° L UNIVERSITY COMMITTEE ON RESEARCH INVOLVING EAST LANSING ' MICHIGAN ° 45824 NUMAN SUBJECTS IUCRIHS) 258 ADMINISTRATION“ BUILDING August IS, l978 (SIT) SSS-IIEO Dr. Louise Sause Educational Psychology #59 Erickson Hall Dear Dr. Sause: Subject: Project by Edward T. Manzitti Entitled "Father-Child Relationship 5 Father-Availability During Childhood s Sex-Role DeveIOpment Among College Students From Intact Families" ' The above referenced project was recently submitted for review to the UCRIHS. We are pleased to advise that this review indicated that the rights and welfare of the human subjects appear to be adequately protected and the Committee, therefore, approved this project at its meeting on August lh, l978. In approving this preposal, however, you should be advised that a condition of approval is that the consent form specifically identifies the project by title so that the limits of the subjects' participation are clearly defined. Please provide this office with a c0py of the revised consent form so that our records can be completed. Thank you for bringing this project to our attention. If we can be of any future help, please do not hesitate to let us know. ‘ Sincerely, 5.324;“. Henry E. Bredeck Chairman, UCRIHS ’}v3—-/7 (, f'li // /h7 DU .. I jms cc: Edward T. Manzitti APPENDIX B Instructions for Instructors and for Subjects 101 To the Instructors: Here are the questionnaires that relate to the study that considers the ways that fathers behave toward their children. the amount of time that fathers spend with their children. and the self-descriptions of these children as adults. In order to control for a variety of variables that may affect the results it is necessary to restrict the sample. Please distribute the questionnaires only to students that meet all of the following qualifications: 1. Between 16 and 25 years of age. 2. White and a native of the U.S.A.. 3. First-born or an only-child. h. From an intact family. (Must have lived with both natural parents from birth until age 16.) Please note that there are two forms. one for females and one for males. The female form is marked 'P' and the male form is marked 'M'. The students should follow the instructions that are included with the questionnaires and they should complete the questionnaires whenever they have the time. The questionnaires can be completed within twenty minutes and should be returned to you during the following class. Please hold both the completed and the spare questionnaires for me. I will get them at a time that is convenient for both of us. If you have any questions you can contact me at 3h9-526h. Thank you for helping me. Sincerely. J. ‘vétz; Edward T. Manzitti ‘ Doctoral Candidate Educational Psychology 102 To the Prospective Subjects: This study is being conducted by Edward T. Manzitti. a doctoral candidate in Educational Psychology at Michigan State University. The study considers the ways that fathers behave toward their young children. the amount of time that fathers spend with their young children. and the self-descriptions of these children as adults. In order to control for a variety of variables that may affect the results. it is necessary to restrict the sample. If you want to participate in the study. you must meet all of the following qualifications: 1. You must be between 16 and 25 years of age. 2. You must be White and a native of the U.S.A.. a. You must be first-born or an only-child. . You must be from an intact family. That is. you must have lived with both of your natural parents from birth until the time you were 16. As a subject in this study. you will be asked to answer a variety of questions that appear in the questionnaires that are supplied to you. The questions require short answers. checks. or numerical ratings. as the instructions specify. Take as much time as you need to respond. but you probably won't need more than twenty minutes to finish. All responses will be confidential. Please 93 933 write your name on any of the question sheets. but write your name and student number on the 3x5 card. Notice that there are two forms. one for males and one for females. Hales should use the form marked ‘fl' and females should use the form marked 'P'. Be sure to read. sign. and date the Consent Porn. When you are finished. please return all of the materials intact to your instructor or the investigator. as directed. You can find out more about the study by contacting the investigator at the Dept. of Ed. Psych. office. fourth floor. Erickson Hall. M.S.U.. A summary of the results of the study will be available at the same office when the study is completed sometime in the winter. Thank you for your help. Encorely .%j‘a% Edward T. Manzitti Doctoral Candidate Educational Psychology APPENDIX C Consent Form and Instruments 103 Michigan State University Consent Form Project: Father-child relationship and father-availability during childhood and aspects of affective development among college students I have freely consented to take part in a scientific study being conducted' by: Edward—1L_Jummriftie under the supervision of: m Academic Title: Ihggggllpr- The study has been explained to me and i understand the explanation that has been given and what my participation will involve. I understand that I am free to discontinue my participation in the study at any time without penalty. I understand that the results of the study will be treated in strict confidence and that I will remain anonymous. Within these restrictions. results of the study will be made available to me at my request. i understand that my participation in the study does not guarantee any beneficial results to me. . " i understand that, at my request. I can receive additional explanation of the study after my participation is completed. Signed Date 104 PERSONAL BACKGROUND QUESTIONNAIRE 1. Name (395 :ggniggg) 2. A89"____. 5. Date of birth 4. Place of birth 5. Have you ever lived outside the USA? HOW long? Your age then 6. Year in collecie, Fresh. S0ph. Jr. Sr. 7a. Grade point average 7b. SAT scores: Verb. __ Quant. 7c. M/ TO'mL fro/3’5 8. Your marital status Single Married Divorced/ Separated Widowed 9. Race or ethnic group ___white ___Plack (non-Spanish surname) ___Black (Spanish surname) ___Chicano _Oriental American Indian ___Other (Explain) 10. Marital status of parents Mother (check all that apply) Divorced or separated Married to your father 11. ’12. 13. 14. ’15. 105 Remarried (year remarried ) Widowed (year husband died ) Single (never married) Other (Explain) Father (check all that apply) ___pivorced or separated _Married to your mother ___Remarried (year remarried ) ___Widowed (year wife died ) ‘__‘Single (never married) ___Cther (Explain) Father's Occupation (Please be specific) Mother's Occupation (Please be specific) Father's Education (Mark last grade completed) 4—8 9-41 Completed 3.8. Some college Bachelor's degree Some graduate work Master's degree Doctoral degree Mother's Education (Mark last grade completed) ___1+-8 ___9—’l’l __Completed 3.8. __Some college ___Pachelor's degree\ ___Some graduate work .__;Master's degree ___Doctora1 degree If you have any brothers or sisters please indicate sex and ages below: (oldest to youngest) ’l. Sex _M F Age 2. 5. 106 4. 5. 6. 7. .. (If you need additional space, please use the back of this page. 16. Are there any other adults living with you? Yes No If yes, indicate the sex and ages below: ’1. Sex __M ___F Age 2. 3. (If you need additional space, please use the back of this page. . 17. If your parents are married to each other, please answer the following. Do not answer if your parents are divorced or separated. A. Was your father in the home all the time (except for short trips for no more than two weeks at a time for vacation or business purposes) when you were growing up? ___Xes ___No If your answer is "no,” please explain on the back of this page. B. If your father was not in the home consistently, how old were you when he left home? Age 107 C. wa much contact did you have with your father after he left home? daily weekly monthly eveiiSB "“' mon every 6 months once a year other (explain 108 Huber ‘80 Date PCR 01151310811113! II FATHER-DAUGHTER Siegelm-Roe Here are 50 etetenente which deecribe different me that fathers act towards their daughters. Reed each eteteunt carefully and think heaven it deecrihee hmyour father acteduhileyou were growing”. Makemciany aboutthetinebefonycuwere 12. Beforeeechetctenent therearefcur linee. These muumdvm mgmmumgmmum;mm. Putanlon thelinethetindicateehcwtrueycuthinkeechetetemntmcfyour' father. For ample, ifyournemcryiethet your father often let you off eeeymenyeudidecmthingwrong,yonuculdmrktheiteueetonoue: ENDED TENDED VERY tube tube VERY T3118 T3113 mm: UMBER Hyfether mthingmng IIEE ll’é . . llll llll llll llll 3% PCR QUETIONNAIRE II TENDED to be vmu UNTFUE UNTRUE 109 FATHER-mm My father 1. 2. 3. h. 5. 6. 7. 8. 9. ’10. 11. 12. 13. 1h. 1§e 15. 17. 1e. 19. was genuinely interested in w affairs mashednehardenoughwhsn Inisoe- havedtonakesurexuouldnotdoit again ' spoiled u let us know I was not wanted set very few rules for no discussed what was good about m be- haviorandhslpedtonake clear the desirable consequences of my actions nods it clear that he was boss relaxedrulssandregulationsasa remrd wastoobusytoanswermyqusstions gavemasmchfreedonastanted made no feel wanted and needed neverletnegetamwithbreahng arule rewarded me by letting me off some of my regular chores didnotspendawnoretinsuithne thanhehadto letneoffeasywhenldidsonething wrong nadene feeluhatIdidwas important spanked or whipped me as punishant gave me candy or ice cream as a reward did not taloe me into consideration in making plans TENDED tebe TRUE 110 PCR QUESTIONNAIRE II TENDED to be UNTRDE I I II I I I III I IIIII I Iga FATHER-DAUGHTER My father 20. 21 . 22. 23. 2h. 25. 26. 27. 28. 29. 30. 31 . 32. 33- 3h. 35. 36. 37. did not want no to play rough outdoor games for fear I might be hurt . talked to me in a warm and affection- ate way . demanded unquestioning respect made others give in to me ridiculed and made fun of me let me stay up as late as I liked tried to help me when I was scared or upset punished me by sending me out of the room or to bed gave no special attention as a reward acted as if _ I did not exist let me do pretty much what I wanted to do respected my point of view and en- couraged me to express it punished me by being more strict about rules and regulations let me go to parties or play with others norethanusualasareward complained about me did not object to aw loafing or dayb dreaming tried to help me learn to live com- fortably with myself wanted to have complete control over my actions TEIIIDEDTEIIHJED PCR QUESTION II tobe tobe van: mans Myfather TRUE mm 38. 39. 1.0. In. 142;. 1:5 . I46. ’47. 1:8. 1:9. 111 FATHER-mm ‘ rewardednebygivingnenonsyorin- creasingnvallowance ignoredneaslongasIdidnotdo anythingtodisturbhin did not object when I was late for meals encouragednetobringfriendshons andtriedtonakethingspleasant forthen taughtnethatheknewbestandthat I must accept his decisions wantednetohaveatleastaslarge anallowanceaswfriends didnottrytohelpmelearnthings waseasywithns madeit easyformetoconfideinhim expected prompt and unque stioning obedience gavenenewthingsasareward, such 3815078 believed a child should be seen and not heard did not bother much about enforcing rules 112 Date PCR ammonium II FATHER-80H Siegelman-Roe Here are 50 statements winch describe different ways that fathers act towards their sons. Read each statement carefully endthinkhowwenitdescribeshowyourfatheractedwmleyou weregroungup. Tuskespecianyeboutthetdnebeforeyou werelz.' Before eachstetuent there are'fourlines. Thesearelae ummm3mmmumgmwummgm ' W. Putanlonthelinethatindicateehowtrueyouthink eaohstatenentwasofyourfather. Porexauple,ifyournemryisthetyourfatheroftenlet youoffeesyvmsnyondidsouthingwrou,youwouldnarkthe itanasfollews: mm mm vm tube tobe mu mm mm mm mm: Hyfather __ __ __ _X__ l. neverletneoffeasywhen Ididsonethingvn-ong lléé PCR QUETIONNAIRE II mm TENDED tabs tobe vm TRUE mm UNTRUE 113 EATER—SON My father 1. 2. 3. h. 6. 7. 8. 9. 1o. 11. 12. 13. 1h. 15. 16. 17. 18. 19. was genuinely interested. in w affairs punishednehardenoughwhenlmbe- havedtonakesureIwouldnotdoit again triedtogetneeverythinnganted letneknowIwasnotwanted letneependwallowanceanyweyI liked made me feel what I did was important tookmymtcysandplaythingswhen Iwasbad spoiledne wastoobusytoanswerwquestions set very few rules for no nadsnefeelwantedsndneeded nadsitclearthathewasbcss gavenenewbooks or recordsqu didnotspendmnoretinewithne thanhehadto gaveneasnuchfreedonaslwanted talked to no in a warm and affection- ate way would not let neplay with other childrenwhen I was bad praised me before w playmates paid no attention to what I doing in school to be to'be TEE WIEUE mans PCR QUESTIONNAIRE II TENDED V311! 114 FATHER-SON My father 20. 21. 22. 23. 2h. 25. 26. 27. 28. 29. 31 . 32. lotneoffeasywhenIdidsonothing wrong triedtoholpnewheanasacared orupsot spankedorwhippednoaspunishnont gavenocanchoricecreanasa did not take no into consideration innakingplans gave no a choice of what to do when- ever it was possible respected w point of view and on- couragod no to express it frightened or threatened no when I -didwrong nade others givein to no complained about no letnedoprottynuchwhatlwanted todo nadenafoolproudwhenIdidwoll denandedumuestioningrespect gave no special attention as a reward didnottrytoholpnelearnthings didnottollnowhattinetobohono whoanontcut reasoned with no and explained possible hamful consequences when I did wrong punished no by being more strict about rules and regulations VERY ma otasnomum n MEDTENDED tobe tobe TRUE 01mm VERY 115 FATHER-SON My father 38. 39. hOe In. h2. 1.13. 115. 116. 117. hugged no, kissed no, potted no when I was good iguorodnoaslongasIdidnotdo enythingtodisturbhin did not object when I was late for meals was willing to discuss regulations withne, andtookwpoint ofview into consideration in making then taughtnethathoknowbestandthat I must accept his decisions gevenonewthingsaearoward,such "to?! ‘paidnoattentiontono waseasywithne nadeitoasyfornetoconfidoinhin expected prompt and unquestioning obedience hatedtorefusenoamthing avoidodwconpaw did not bother much about enforcing rules 116 For the following question please check the rating that indicates to what extent each iten a plied to your father during the years when you were a child grow ng up. ' Very Very Frequently Frequently Sonotinee SeldosJ Seldos 130ne for lunch. Away free hole and children on the Iwoekends. Out in the evening at least two nights a week. hone afternoons ‘hen children cone Ihose free school. way fron hone and children for sake and sonths at a tine. breakfast 1th fanily and hildren. hildren at least {hissed supper with c two nights a week. iway free hone for 1days at a tine. Is all day with anily and children. ut in the evening t least four ghts a week. 117 On the following page, you will be shown a large number of personality char- acteristics. we would like you to use those characteristics in order to describe yourself. That is, we would like you to indicate, on a scale from 1 to 7, how true of you these various characteristics are. Please do not leave any character- istic unmarked. Example: sly Mark a 1 if it is NEVER OR ALMOST NEVER TRUE that you are sly. Mark a 2 if it is USUALLY NOT TRUE that you are sly. Mark a 3 if 1: is sous'rmgs 3m: INFREQUENTLY runs that you are sly. Mark a 4 if it is OCCASIONALLY TRUE that you are sly. Mark a 5 if it is OFTEN TRUE that you are sly. Mark a 6 if it is USUALLY TRUE that you are sly. Mark a 7 if it is ALNAYS 0R ALMOST ALWAYS TRUE that you are sly. Thus, if you feel it is sometimes but infrequently true that you are "sly,” never or almost never true that you are "malicious," always or almost always true that you are "irresponsible," and often true that you are "carefree," then you would rate these characteristics as follows: Sly 3 Irresponsible [ 7 Malicious I an“... J5 118 Strong personality Eager to soothe hurt feelings Loyal Conceited Unpredictable Dominant Forceful Soft-spoken Feminine Likable Masculine 1 2 3 l. 5 6 "7 ,2 l l l L l l l _7 r I l I 1 I NEVER 0R USUALLY SOMETIMES BUT OCCASIONALLY OFTEN USUALLY vALWAYS OR ALMOST NOT INFREQUENTLY TRUE TRUE TRUE ALMOST NEVER TRUE TRUE TRUE ALWAYS TRUE Self reliant Reliable Harm Yielding Analytical Solemn Helpful Sympathetic Willing to take a stand Defends own Jealous beliefs Tender Has leadership Cheer f u]. ab 1]. it 1.8 Pr iendly Moody Sensitive to the Aggressive needs of others Independent Gullible Truthful Shy Inefficient Willing to take risks Conscientious ACCO II a leader _ Understanding Athletic Childlike Secretive Affectionste Adaptable Makes decisions Theatrical easily Individualistic Augertivg Compassionate Does not use harsh language Flatterable Sincere Unsystematic Competitive Loves children Tactful Ambitious Gentle Conventional BIBLIOGRAPHY 119 BIBLIOGRAPHY Aberle, D. & Naegele, K. Middle class father's occupa- tional role and attitude toward children. American Journal of Orthopsychiatry, 1952, 22, 366-378. Altus, W.D. The broken home and factors of adjustment. Psychological Reports, 1958, i, 477. Bach, G.R. Father-fantasies and father-typing in father separated children. Child Development, 1946, 11, 63-80. Bandura, A. & Walters, R.H. Social learningand person- ality development. New York: Holt, Rinehart and Winston, 1963. Barclay, A.G. & Cusumano, D. Father-absence, cross-sex identity, and field-dependent behavior in male adolescents. Child Development, 1967, 28, 243-250. Bem, S. The measurement of psychological androgyny. Journal of Consulting and Clinical Psychology, 1974, 32, 155-162. Biller, H.B. A note on father absence and masculine development in young, lower class Negro and White boys. Child Development, 1968, 39, 1003-1006. Biller, H.B. Father-absence and personality development of the male child. Developmental Psychology: 1970, 2, 181-201. Biller, H.B. Father-absence, maternal encouragement, sex-role development in kindergarten age boys. Child Development, 1969, 2Q, 539-546. Biller, H.B. Father, child, and sex-role. Lexington, Mass.: Lexington Books, D.C. Heath & Co., 1971(a). Biller, H.B. The mother-child relationship and the father- absent boy's personality development. Merrill-Palmer Quarterly, 1971(b), 11, 227-241 120 Biller, H.B. Paternal deprivation. Lexington, Mass.: Lexington Books, D.C. Heath & Co., 1974. Biller, H.B. The father and personality development: Paternal deprivation and sex-role development. In M. Lamb (Ed.) The role of the father in child development. New York: John Wiley & Sons, 1976. Biller, H.B. & Bahm, R.M. The father-daughter relation- ship and the personality development of the female. Journal of Genetic Psychology, 1970, 114, 79-93. Blood, R.O. & Wolfe, D.M. Husbands and wives: The_ dynamics of married living. New York: Free Press, 1960. Bowerman, C.E. & Elder, G.H. Jr., Variations in adoles- cent perception of family power structure. American Sociological Review, 1964, 29, 551-567. Broadbeck, A.J. Learning and identification: IV. Oedipal motivation as a determinant of conscience development. Journal of Genetic Psychology: 1954, 84, 219-227. Bronson, W.C. Dimensions of ego and infantile identifica- tion. Journal of Personality, 1959, 31, 532-545. Brown, D.G. Sex-role development in a changing culture. Psychological Bulletin, 1958, 55, 232-241. Carlsmith, L. Effect of early father-absence on scholastic aptitude. Harvard Educational Review, 1964, 34, 3-21. Cottrell, L.S. The adjustment of the individual to his age and sex-roles. American Sociological Review, 1942, 1, 617-620. Cox, D.R. Planning of experiments. New York: Wiley & Sons, Inc., 1958. Deutsch, M. Minority group and class status as related to social and personality factors in scholastic achieve- ment. Mono raphs of the Society for Applied Anthropology, 1960, 2, l- 2. Deutsch, M. & Brown, B. Social influences in Negro-White intelligence. Journal of Social Issues, 1964, 20, 24-35. 121 Distler, L.S. Patterns of parental identification: An examination of three theories. Unpublished doctoral dissertation, University of California, Berkeley, 1964 . In H.B. Biller, Paternal deprivation. Lexing- ton, Mass.: Lexington Books, D.C. Heath & Co., 1974. Fish, K.D. & Biller, H.B. Perceived paternal relation- ships and college females personal adjustment. Adolescence, 1973, 5, 415-420. Gardner, L.P. An analysis of children's attitudes toward fathers. Journal of Genetic Psychology, 1947, 15, 3-28. Gaudreau, P. Bem Sex-Role Inventory validation study. Paper presented at the meeting of the American Psychological Assocation, Chicago, IL. , 1975. Glass, G.V. & Stanley, L.C. Statistigal method§5in educa- tion andpsychology, Englewood Cliffs, N.J.: Prentice-Hall, 1970. Goodeneough, E.W. Interest in persons as an aspect of sex-differences in the early years. Genetic Psychology Monographs, 1957, 55, 287-323. Gray, S.W. Perceived similarity to parents and adjustment Child Development, 1959, 55, 91-107. Hartnagel, T.F. Father-absence and self-conceptions among lower class White and Negro boys. Social Problems, 1970, 55, 152-163. Helper, M.M. Learning theory and the self-concept. Journal of Abnormal and Social Psychology, 1955, 55, 184-194. Herzog, E. & Sudia, C.E. Bo s in fatherless families. Washington: Office 0 C ild Development, 1970. Hetherington, E.M. Effects of father-absence on personality development in adolescent daughters. Developmental Psychology, 1972, 1, 313-326. Jacobson, G. & Ryder, R.G. Paternal loss and some char- acteristics of the early marriage relationship. American Journal of Orthopsychiatry, 1969, 55, 779-787. Johnson, M.M. Sex-role learning in the nuclear family. Child Development, 1963, 54, 319-333. 122 Jones, H.B. Father-absence during childhood, maternal attitudes toward men, and the sex-role development of male college students. 'Unpublished doctoral dissertation, Michigan State University, 1975. Kagan, J. The concept of identification. Psychological Review, 1958, 55,-296-305. Kerlinger, F.N. Foundations of behavioral research. New York: Holt, Rinehart, and Winston, Inc., 1973. Landy, F., Rosenberg, B.G., & Sutton-Smith, B. The effect of limited father-absence on the cognitive and emotional development of children. Child Development, 1969, 59, 941-944. Leonard, M.R. Fathers and daughters. International Journal of Psychoanalysis, 1966, 51, 325-333. Lindquist, E.F. Design and analysis of experiments in psychology and education. Boston: Houghton Mif lin Co., 1956. Lynn, D.B. The_father: His_£ole in child development. Monterey, Ca.: Brooks-Cole Publishing Co., 1974. Lynn, D.B. & Sawrey, W.L. The effects of father-absence on Norwegian boys and girls. Journal of Abnormal and Social Psychologym 1959, 55, 258-262. Kohlberg, L. A cognitive-developmental analysis of children's sex-role concepts and attitudes. In E.E. Maccoby (Ed.), The development of sex differences. Stanford, CA.: Stanford University Press, 1966, 82-173. Maccoby, E.E. & Jacklin, C.N. The psychology of sex differences. Stanford, CA.: Stanford University Press, 1974. McCord, J., McCord, W., & Thurber, E. Some effects of paternal absence on male children. Journal of Abnormal and Social Psychology, 1962, 53, 361-369. McKinley, D.G. Social class and family life. New York: Free Press, 1964. McSweeny, M. & Busk, P. Notes for Education 969B. Un- published manuscript, Michigan State University, East Lansing, MI., 1976. 123 Mowrer, G.H. Learning theory and personality dynamics. New York: Ronald Press, 1950. Mussen, P.H. Some antecedents and consequences of masculine sex-typing in adolescent boys. Psychological Monographs, 1961, 55, No.2 (Whole No. 506). Mussen, P.H. & Distler, L. Masculinity, identification, and father-son relationships. Journal of Abnormal and Social Psychology, 1959, 55, 350-356. Mussen, P.H. & Rutherford, E.F. Parent-child relationships and parental personality in relation to young chil- dren's sex-role preferences. Child Development, 1963, 55, 589-607. Nunnally, J.C. Psychometric theory. New York: McGraw- Hill Company, 1967. Osgood, C., Suci, G., & Tanenbaum, P.H. The measurement of meaning. University of Illinois Press, 1957. Parsons, T. Family structure and the socialization of the child. In T. Parsons and R.F. Bales (Eds.), Family, socialization, and interaction process. Glencoe, ILL.: Free Press, 1955, 25-131. Parsons, T. Social structure and the development of per- sonality: Freud's contribution to the integration of psychology and sociology. Psychiatry, 1958, 55, 321-340. Payne, D.B. & Mussen, P.H. Parent-child relations and father- identification among adolescent boys. Journal of Abnormal and Social Psychology, 1956, 55, 358-362. Reuter, M.W. & Biller, H.B. Perceived paternal nurturance- availability and personality adjustment among college males. Journal of Consulting and Clinical Psychology, 1973, 55, 339-342 Roe, A., & Siegelman, M. A parent-child relation ques- tionnaire. Child Development, 1963, 55, 355-369. Runyon, R.P., & Haber, A. Fundamentals of behavioral statistics. Reading, Mass.: Addison-Wesley, Inc., 1968. Santrock, J.W. Paternal absence, sex-typing and identifi- cation. Developmental Psychology: 1970, 5, 264-272. 124 Sears, P.S. Doll-play aggression in normal young chil- dren: Influence of sex-age, sibling status, and father's absence. Psychological Monographs, 1951, 55, No.6, (Whole No. 323T} Sears, P.S. Child rearing factors related to playing of sex-typed roles. American Psychologist, 1953, 8, 431. Sears, R.R., Pintler, M.H., & Sears, P.S. Effect of. father separation on pre-school children's doll play aggression. Child Development, 1946, 55, 219-243. Sears, R.R. Identification as a form of behavior develop- ment. In D.B. Harris (Ed.), The concept of develop- ment. Minneapolis, MN.: University of Minnesota Press, 1957, 149-161. Sears, R.R., Maccoby, E.E., & Levin, H. Papterns of child rearing. Row, Peterson, Inc., 1957. Siegelman, M., & Roe, A. Parent-child relations question- naire II: manual. Unpublished manuscript, The City College of New York, N.Y., 1978. Sopchak, A.L. Parental "identification" and "tendency toward disorders" as measured by the MMPI. Journal of Abnormal and Social Psychology, 1952, 55, 159-165. Tasch, R.J. The role of the father in the family. Journal of Experimental Education, 1952, 55, 319-361. United States Department of Labor Dictionary of ocoupational titles. Washington: United States Government Publica- tions, 1965. Waters, C.W., & Pincus, 8. Sex of respondent, respondent's sex-role, self-concept, and responses to the Sensation- seeking scale. Psychological Reports, 1977, 55, 567-570. Waters, C.W., Waters, L.K., & Pincus, 8. Factor analysis of masculine and feminine sex-typed items from the Bem Sex-Role Inventory. Psychological Reports, 1977, 55, 567-570. Whiting, J.W.M. Sorcery, sin, and the superego: A cross- cultural study of some mechanisms of social control. In M.R. Jones (Ed.), Nebraska symposium on motivation. Lincoln, NEB.: University of Nebraska Press, 1959. 125 Whiting, J.W.M. Research on mediation and learning by identification. In I. Isco and H.W. Stevenson (Eds.), Personality development in children. Austin, TX.: University of Texas Press, 1960. Winch, R.F. Identification agd its familial determinants. New York: Bobbs-Merrill Press, 1962.