4273"“ "ll ‘Ilf"l‘ x-‘Iin‘IIIIi‘II-IIIIIf?!SIEIIIEII PERSONAL. FAMILY AND INSTITUTIONAL FACTORS ASSOCIATED WITH ATTITUDES TOWARD WOMEN'S ROLES AMONG ; FRENCH-CANADIAN DDLLEGE STUDENTS Dissertation for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY MONIQUE C. PROULX 1975 AH“ EIJ [,IBKi/ik 1' D-viiciiigan State University This is to certify that the thesis entitled Personal, Family and Institutional Factors Associated with Attitudes Toward Women's Roles Among French—Canadian College Students presented by Monique C. Proulx has been accepted towards fulfillment of the requirements for Ph .D . degree in Family Ecology Major professor Date August 13, 1976 0-7 639 i “8:80! i ' ox slunEIiv it i i mum mums ‘ simmer. Inc-am II ABSTRACT PERSONAL, FAMILY AND INSTITUTIONAL FACTORS ASSOCIATED WITH ATTITUDES TOWARD WOMEN'S ROLES AMONG FRENCH-CANADIAN COLLEGE STUDENTS BY Monique C. Proulx This study was designed to explore sex-role attitudes among French-speaking college students of New Brunswick, Canada. An objective was to identify factors that are associated with students' attitudes toward women's roles. An attitudinal scale, developed for the present study was used to determine the respondents' positions on a continuum of traditionalism-liberalism with regard to views of appropriate roles for women. The sample consisted of 234 male and female college freshmen enrolled in a degree program in three French-speaking institutions. Chi-square tests, Pearson's product moment corre- lation and multiple correlation analyses were performed to study the relationships between the students' attitudes toward women's roles and selected personal, family, and institutional factors. Substantial differences of attitudes between men and women were observed in this sample; women students' Monique C. Proulx attitudes were significantly more liberal than men's. The relationships between attitudes toward women's roles and the independent variables were further explored by controlling the variable of sex. Different patterns of association were found between males and females. The hypotheses related to personal variables were supported except those pertaining to age and employment experience; the other variables were: sex, anomia, self- actualization, academic performance, residential inde- pendence, choice of academic field, and marriage plan. The family variables, such as father's occupation, occupational status of the mother, mother's level of educa- tion, family income, and family size were not significantly associated with the dependent variable. Father's education was significantly associated with daughter's attitudes. Institutional factors, such as size of institution attended, and urbanization level of region of origin were significantly associated with attitudes toward women's roles. Although the subjects coming from larger home towns tended to be more liberal than those coming from smaller towns, no significant association was found between stu- dents' attitudes and size of home town. The multiple regression procedure performed separately for the male and female subsamples generated a different combination of variables as best predictors Monique C. Proulx of attitudes toward women's roles. Size of institution attended, self-actualization, field of study, and resi- dential independence were the best predictors of the men's attitudes. Marriage plan, size of institution attended, and self-actualization were the best predictors of the women's attitudes. Implications for future research concerned cross— cultural and inter-generational comparisons, and changes in men's roles as well as changes in wOmen's. Propositions were offered pertaining to sex-role socialization, and educational effort toward a de-emphasis of sex-role stereotypes. PERSONAL, FAMILY AND INSTITUTIONAL FACTORS ASSOCIATED WITH ATTITUDES TOWARD WOMEN'S ROLES AMONG FRENCH-CANADIAN COLLEGE STUDENTS BY \ a 4—! Monique c: Proulx A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Family Ecology 1976 ACKNOWLEDGMENTS My gratitude goes to several peOple who have contributed to my professional development and to the completion of this study. I am particularly indebted to Professor Jean Schlater, academic advisor and dissertation director, for her assistance and her unfailing availability during the course of my studies, and her patience in intro- ducing me to the subtleties of the English language. I want to express my sincere thanks to Professor Beatrice Paolucci for her interest and support throughout my program of studies. Her inspiring teaching and her professional involvement were influential in my professional development. Thanks are due to Professor Denton Morrison who was particularly helpful in development of the research design. To Drs. Vera Borosage and Jane Oyer who served as committee members, my special appreciation for their interest and encouragement. I am very grateful to Dr. Mary Andrews, who introduced me to the language of computer science and assisted in the analysis of the data. I am also grateful to Drs. Norma Bobbitt and Linda Nelson for their careful reading of the manuscript. ii Special appreciation goes to my colleagues of the Université de Moncton, particularly to Dr. Marielle Préfontaine, director of the Ecole des Sciences domestiques, whose interest and encouragement contributed to the completion of this study. To the staff members of St. John's Student Center, East Lansing, whose friendship has been a great support during my studies, I express my deepest appreciation. I am also grateful to the members of the Institute of Notre Dame du Bon Conseil de Montreal for their support and encouragement to complete my doctoral program. Part of this study was supported by a grant from Le Conseil des Arts du Canada. iii TABLE OF CONTENTS LIST OF TABLES O O O 0 O O O O O O O O O O O O 0 LIST OF FIGURES . . . . . . . . . . . . . . . . . CHAPTER I. INTRODUCTION . . . . . . . . . . . . . . Statement of the Problem . . . . . Research Objectives and Hypotheses Definitions . . . . . . . . . . Assumption . . . . . . . . . . The Setting . . . . . . . . . . . . II. CONCEPTUAL BASIS . . . . . . . . . . . . Role and Sex-Role Concepts . . . . . Definition of Roles . . . . . . . Sex-Role Concepts . . . . . . . Theories of Sex-Role Socialization . Personal Factors . . . . . . . . Family Factors . . . . . . . . . Societal Factors . . . . . . . . Conceptual Models of Women‘s Roles . III 0 REMTED RESEARCH O O C O O O Q 0 O O O O Sex-Role Attitudes . . . . . . . . . Trends in Sex-Role Attitudes . . Male-Female Differences in Sex-Role Attitudes . . . . . . . . . . . . Factors Influencing Sex-Role Attitudes IV. DESIGN OF THE STUDY . . . . . . . . . . . Population and Sample . . . . . . . . . Data Collection . . . . . . . . . . . . Development of Measures . . . . . . . . iv Page vi viii \lO‘O‘NH H 13 13 15 18 18 20 24 30 30 30 31 35 42 42 43 45 CHAPTER Measure of Attitudes Toward Women's Roles . . . . . . . . . . . . . . Measures of Personal Characteristics Measures of Family Characteristics . Measures of Institutional Factors . . Statistical Treatment of the Data . . . V. THE FINDINGS O O O O O O O O O O O O O 0 0 Description of the Sample . . . . . . . . Results of Data Analysis . . . . . . . . WOmen' 3 Role Attitude Scores . . . Personal Characteristics and Attitudes Toward Women' s Roles . . . . . . Family Characteristics and Attitudes Toward Women's Roles . . . . . . Attitudes Toward Women' s Roles and Institutional Factors . . . . . . . Summary of Results . . . . . . . . . . . VI. DISCUSSION OF FINDINGS AND LIMITATIONS OF THE STUDY 0 O O O O I O C O C O C O O 0 Discussion of Findings . . . . . . . . . Personal Factors . . . . . Family Characteristics and Attitudes Toward Women's Roles . . . . . . . Institutional Factors and Attitudes Toward Women's Roles . . . . . . . Limitations of the Study . . . . . . . . VII. SUMMARY AND IMPLICATIONS . . . . . . . . . Summary . . . . . . . . . . . . . . . . Implications for Future Research . . . . APPENDIX A. QUESTIONNAIRE, LETTER OF INTRODUCTION, AND LETTER OF RECALL . . . . . . . . . . . . . B. PERCENTAGE DISTRIBUTION OF RESPONSES ON ATTITUDINAL ITEMS BY SEX AND SIZE OF INSTITUTION ATTENDED . . . . . . . . . . . BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . Page 45 49 55 59 61 61 65 67 83 87 97 100 100 100 106 108 110 113 113 116 119 137 138 Table 1. 10. 11. 12. 13. LIST OF TABLES Descriptive Statistics for WOmen's Role Attitude scale 0 O O O O O O O O O O O O O 0 Frequency Distribution of Raw Scores on the Women's Role Attitude Scale . . . . . . Frequency Distribution of Categorical Scores on the WOmen's Role Attitude Scale . Description of the Sample of Students . . . Association Between Attitudes Toward WOmen's Roles and Sex . . . . . . . . . . . . . . . Pearson Product Moment Correlation Coef- ficients Representing the Interrelationships Among WOmen's Role Attitudes and Selected Personal Variables . . . . . . . . . . . . . Association Between Age and Attitudes Toward WOmen's Roles . . . . . . . . . . . . . . . Association Between Attitudes Toward WOmen's Roles and Residential Independence . . . . . Association Between Attitudes Toward WOmen's Roles and Employment Experience . . . . . . Association Between Attitudes Toward Women's Roles and Sex-Bound Choice of Academic Field Association Between Attitudes Toward WOmen's Roles and Marriage Plan . . . . . . . . . . Summary Table of Statistical Tests Based on Contingency Tables for Selected Personal Characteristics of Students . . . . . . . . Association Between Attitudes Toward WOmen's Roles and Family Characteristics . . . . . . vi Page 49 50 51 66 69 73 74 76 78 79 81 82 85 Table 14. 15. 16. 17. 18. 19. 20. 21. 22. Summary of Statistical Tests for Family Characteristics by Sex and Total Sample . Association Between Attitudes Toward Women's Roles and Size of Institution Attended 0 O O O O O O Q 0 O O O O O O O 0 Association Between Attitudes Toward WOmen's Roles and Urbanization Level of the Subject's Region of Origin . . . . Association Between Attitudes Toward WOmen's Roles and Size of Home Town . . . Association Between Attitudes Toward Wbmen's Roles and Sex by Institution Attended 0 O O I I O 0 O O O O O O O 0 O 0 Summary Table of Statistical Tests for Institutional Factors . . . . . . . . . . Multiple Regression Analysis Predicting Attitudes Toward WOmen's Roles from Independent Variables for Male Subsample . Multiple Regression Analysis Predicting Attitudes Toward Women's Roles from Independent Variables for Female Subsample Percentage Distribution of Responses on Attitudinal Items . . . . . . . . . . . . vii Page 86 88 9O 91 92 93 94 95 Figure 1. LIST OF FIGURES Geographical Location of Institutions in the Sample . . . . . . . . . . . . . Urban Centers in the Atlantic Provinces Frequency Distribution of Scores on the WOmen's Role Attitude Scale . . . . . Respondent's Position on the Continuum of Traditionalism-Liberalism, by Sex viii Page 68 68 CHAPTER I INTRODUCTION Statement of the Problem Until recently, the available data on family life have presented French Canadians with strongly conservative views about women and their roles in the family and in society. Rocher (1962) pointed out that family life in French Canada has been profoundly revolutionized and that these images of women appeared to be different from reality. The Report of the Royal Commission on the Status of Women (1970) threw some light on the contradictions between the traditional conceptions of women as represented in popular imagination and the new type of French—Canadian women. A well documented study by Carisse (1975) revealed the exis- tence, among this group, of a number of innovative women who are creating models in all facets of social life. The WOmen's International Year provided opportunities for the expression of opinions concerning women's roles and made more obvious the diversity of views and attitudes related to this theme. An inventory of materials on Canadian women published between 1950 and 1972 indicated a sustained interest in the status of women (Eichler, 1973). Few studies conducted in Canada pertained to factors associated with variations of attitudes toward women's roles. Moreover, only one title on that long list concerned New Brunswick women and a few more were about women in the other Atlantic Provinces. More information is necessary on a population often neglected on the national scene: the Acadians, a French-Canadian group living in a predominantly English- speaking Province. This population, seldom heard about, is the most numerous group of French Canadians living out- side the Province of Quebec. They form a subcultural group with a history, a folklore and traditions of their own. Because of the extremely limited published literature on women in this particular group of Canadians, this study hoped to provide pertinent data about young Acadians, their families, and their ideas of appropriate roles for women. Research Objectives and Hypotheses The general purpose of this study is to investigate college students' attitudes toward women's roles. More precisely, it is designed to seek answers to the following questions: 1. Is there an indication of a liberalization of attitudes toward women's roles among French-Canadian college students in New Brunswick? 2. What are the factors that account for variations in attitudes toward women's roles: a. Are there selected personal characteristics, such as sex, age, academic performance, and life experiences associated with variations in attitudes toward women's roles? b. Are certain family characteristics associated with attitudes toward women's roles among college students? c. Are there selected institutional factors such as: size of the institution, size of home town, and urbanization level of region of origin asso- ciated with attitudes toward women's roles? 3. Which of these variables are most powerful in predicting college students' attitudes toward women's roles? The French-speaking college students in New Brunswick are found on three campuses located in the three regions of the Province with a concentration of the French- speaking population. Thus, the sample for this study was drawn from these institutions. It is hoped that the information gained in this study will be useful in helping young men and women to better understand their reciprocal expectations and to cooperate in creating satisfying life styles in their adult life. The regional comparisons made possible by the inclu- sion of the three academic institutions might be informative on regional characteristics that need to be taken into consideration in educational and social planification. The hypotheses of the study are the following: 1. A significant association exists between a student's position on the traditionalism—liberalism continuum regarding attitudes toward women's roles and personal characteristics: 1.1 A more liberal position is positively associated with feminine sex. 1.2 A more liberal position is positively associated with a higher degree of self—actualization. 1.3 A more liberal position is negatively associated with a higher degree of anemia. 1.4 A more liberal position is positively associated with higher academic performance. 1.5 A more liberal position is positively associated with older age. 1.6 A more liberal position is positively associated with residential independence from parents. 1.7 A more liberal position is positively associated with residential independence from school authorities. 1.8 A more liberal position is positively associated with employment experience. 1.9 A more liberal position is negatively associated with sex—bound choice of academic field. 2. A significant association exists between a student's position on the traditionalism-liberalism continuum regarding attitudes toward women's roles and some characteristics of the family: 2.1 A more liberal position is positively associated with the social prestige of the father's occupation. 2.2 A more liberal position is positively associated with employment experience of the mother. 2.3 A more liberal position is positively associated with higher education of the mother. 2.4 A more liberal position is positively associated with higher education of the father. 2.5 A more liberal position is negatively associated with larger size of the family. 2.6 A more liberal position is positively associated with higher family income. 3. A significant association exists between a student's position on the traditionalism-liberalism continuum regarding attitudes toward women's roles and selected institutional factors: 3.1 A more liberal position is positively associated with a larger institution. 3.2 A more liberal position is positively associated with a more urbanized region of origin. 3.3 A more liberal position is positively associated with a larger size of home town. Definitions Attitudes toward women's roles are defined as the evaluative responses made by the students about statements pertaining to behaviors generally attributed to women. Operationally, they are defined by the student's position on a continuum of traditionalism—liberalism with respect to women's roles. The position is determined by the score made on a scale of attitudes toward women's roles based on theoretical definitions that are elaborated upon in Chapter II. Self-actualization is defined as the orientation of a person who is more fully functioning and lives a more enriched life than does the average person (Maslow, 1962: Shostrom, 1965). Two dimensions of this concept as opera- tionalized by Shostrom and used in the present study are further explained in the description of instruments. Anomia is defined as a degree of interpersonal alienation, social malintegration and psychic isolation (Srole, 1956). Assumption The Women's Role Attitude Scale can adequately reveal differences among individuals in their attitudes toward women's roles. The Setting In order to provide background information for the interpretation of the findings, a brief historical note about the academic institutions included in the present study is presented along with the general characteristics of the geographical regions in which these institutions are located (see Figures 1 and 2). In Canada, the Université de Moncton is the only French-speaking university outside the province of Québec. This institution resulted from the fusion of four small universities; the oldest one, founded by the Holy Cross Fathers, had been in existence since 1898, and was located in the southeast region of New Brunswick, near the City of Moncton. This institution was the nucleus from which the Université de Moncton emerged in 1963; the others were transformed into colleges and participated in the formation of the new university through an affiliation system. Grad- ually these colleges ceased to function as separate entities to be fully integrated to the major institution. The total enrollment is now in the order of 5,000 students. Half of these are full-time students distributed in seven "Facultés" or “Ecoles”: Administration, Arts, Education, Home Eco- nomics, Nursing, Science, and Social and Behavioral Sciences. The other half of the student body is under the Continuing BRUNSWICK *7 wmun'nc V r*\ / , ' \uL ,‘ N 0 VA //’ / ‘ o 4 t W 1 /i 0‘ /.///’///i ‘m- 1' W ' Figure 1. Geographical Location of Institutions in the Sample. ca C3 .erihook indamu rand Fal . 14143 ' 01161 Number of persons 200000 100000 50000 10000 19663 Figure 2. Urban Centers in the Atlantic Provinces. Source: Alan G. Macpherson, Studies in Canadian Geography: The Atlantic Provinces (Toronto: University of Toronto Press, 1972). 10 Education Service which offers both degree and non-degree programs. The major campus is located in Moncton, a city at the core of a metr0politan area of about 60,000 population. It is the second city in size after Saint John in New Brunswick, and fifth in the Atlantic Provinces. The Moncton area is considered an important focus of Acadian culture and has a population that is more than one-third French in origin. The economic structure of Moncton is based on transportation and retail trade. The employment figures based on the 1966 Census of Canada showed that Moncton registered the highest percentage of labor force in both transportation and retail trade among the 25 Canadian cities in the same pOpulation range (Macpherson, 1972). Outside the Moncton area, it is only since 1972, after the work of a provincial commission for the planning of higher education, that the other institutions were affiliated with the Université de Moncton. Three regional colleges came under its authority and offered both technical training and programs leading to a bachelor's degree. Two of these, still functioning under the affiliation system and offering degree programs, were included in the present study, one in the northwest region and one in the northeast region of the Province. 11 The northwest campus is located in Edmunston, a city of about 12,000 population, located 300 miles from Moncton, and at the frontiers of Quebec and Main. This region has the highest concentration of the French-speaking population of New Brunswick. Over one-quarter of the total employment in Edmundston is in manufacturing activities (Macpherson, 1972). The city is surrounded by a prosperous agricultural area. The college enrollment for 1974-75 was about 500 students distributed in the fields of Humanities, Sciences, Administration, and in Nursing Techniques and Secretarial Techniques. The College draws its population almost totally from within the northwest area. The Uni- versité de Moncton did not recruit more than 10 percent of its total enrollment (for the Moncton campus) for the same year, from the northwest area. The campus of the northeast area is situated in Shippegan, a small town on the sea shore, with a population of about 2,000. The economy of the area is based mainly on fishery, agriculture, and mining. The total enrollment for 1974-75 was about 150 students, with a predominance of women. They were divided among three fields: Humanities, Adminis- tration, and Secretarial Techniques. The totality of the student body came from within the northeast area. The same year, the Université de Moncton recruited almost 40 percent of its enrollment (for the Moncton Campus) from the north- east area. 12 Although these three institutions operate under the same authority, each campus tends to function with as much autonomy as possible and in relative isolation from each other. Enormous physical distances between the campuses, and the lack of a rapid transportation system prevent interaction among these schools except for the insti- tutional meetings. This investigator was on the faculty of the Ecole des Sciences domestiques, Université de Moncton, for two years (1974-1976). One of her assignments had been to teach one summer course at the college of Edmundston. CHAPTER II CONCEPTUAL BASIS The conceptual framework used in the present study is one of socialization into sex roles. The discussion in this chapter revolves around this framework and is divided into three sections: (1) role and sex-role concepts are discussed; (2) some theories of sex-role socialization are reviewed: and (3) the conceptual models for the operation- alization of attitudes toward women's roles are presented. Role and Sex-Role Concepts Definition of Roles The study of social role acquisition, and especially of sex-role acquisition, occupies an important place in the literature pertaining to socialization. Socialization refers basically to the process of learning new roles, a process that continues throughout the life cycle. Cogswell (1968) defined roles as behavior associated with any type of social identity and suggested that a socialization frame of reference can be used to study how individuals become occupants of roles. 13 14 Biddle and Thomas (1966) surveyed the use of the concept of role in the social science literature. Summarizing their review, they proposed that "the most common definition is that role is the set of prescriptions defining what the behavior of a position member should be" (p. 4). This definition contains three elements-- prescription, position, and behavior-~that are emphasized in most discussions of roles. Roger Brown's (1965) defi- nition of role refers to prescriptive behaviors: "Roles in society are prescribed actions rather than persons" (p. 152). These requirements are attached to one or another of the recognized positions or statuses in the society, such as husband, son, employee, and male adult, and can be called role prescriptions (Brim, 1966). Brim's formulation points to the function of role as a position in a social structure, that is also reflected in Lennard and Bernstein's (1969) definition: "The concept of role refers to a configuration of acts that accompany status occupancy (for example, hus- band or therapist)" (p. 43). Jacobson (1966) defined role as a culturally ascribed pattern of behavior including duties, expected or required of persons behaving in special situations. In explaining children's learning and perform- ing roles, Harocks and Jackson (1972) also stressed the behavioral aspect of role: "a child can only take a role 15 after he has some cognitive basis for defining a role, such as what a mother does, what a teacher does, and what a good boy does not do" (p. 86). Biddle and Thomas (1966) pointed out some distinc- tions to be made among a set of concepts used in relation to roles: these terms are conceptions, perceptions, and expec- tations. Each concept implies cognitions about roles, but with a different connotation in regard to time: role con- ceptions are not bounded by time, while role perceptions refer to the past and role expectations to the future. The same authors mentioned that role prescriptions can be overt and as such refer to norms, while overt prescriptions refer to demands or explicit expression of a prescription. Sex-Role Concepts I The literature on sex roles contains a variety of concepts that are not always used consistently from one researcher to another. It seems appropriate to discuss these concepts to clarify their meaning in relation to the present study. One basic process in the task of learning one's social role is sex-typing. Mischel (1966) defines sex— typing as the process by which the individual acquires sex-typed patterns of behaviors and learns to discriminate between sex-typed behavior patterns, then to generalize 16 from these learning experiences and finally to perform sex-typed behaviors. Mischel (1970) defines sex-typed behaviors as "those that are less expected and sanctioned when performed by one sex, and in contrast are considered to be more appropriate when manifested by the other sex" (p. 4). His definition does not imply that these patterns are immutable, and all acquired during childhood. Instead, they are seen as susceptible to change throughout people's lives. Aggressive behavior typically attributed to males and dependent behav- ior typically attributed to females are two instances of sex-typed behaviors that have received much research attention. In addition to actual average differences in patterns of behaviors, the sexes are also characterized by broad categories they generate about traits they are expected to have. These are called sex—role stereotypes, which are defined as expectations and typical behaviors supposedly displayed by members of each sex (Mischel, 1970). Another concept sometimes used to indicate what the behavior of each sex should be is sex-role standards, defined by Kagan (1964) as expectations about how the sexes should behave. A process critical to the organization of sex-role attitudes is the development of sex-role identity, which 17 according to Money and Hampson (1957) should happen before the age of three or four. Sex-role identity refers to the degree to which an individual regards himself as masculine or feminine (Kagan, 1964). Biller (1972) uses the expres- sion sex-role orientation to designate what Kagan calls sex-role identity: "Sex—role orientation is related to the way an individual basically views himself in terms of maleness and/or femaleness" (p. 75). Sex-role orientation is used by Brogan and Kutner (1976) to refer to normative conceptions of appropriate behavior for males and females. What Brogan and Kutner studied might have been better termed sex-role standards or sex—role conceptions. 'Sex-role identity needs to be distinguished from sex-role identification, sex-role adoption, and sex-role preference. Sex-role adoption refers to an individual's masculinity and/or feminity as viewed by members of his society (Biller, 1969),while sex-role identification means the internalization of the role typical of a given sex in a particular culture and the unconscious reactions character- istic of the role (Lynn, 1969). In turn, sex-role prefer- ence refers to the desire to adOpt the behavior associated with one sex or the other or the perception of such behavior as desirable (Lynn, 1969). 18 Theories of Sex-Role Socialization The importance of socialization processes in the development of sex-role concepts is well established (Sears, 1965: Mischel, 1966). A large body of theories and research illustrates the complexity of these processes and suggests that several categories of factors account for the differ- ences in conceptions and attitudes related to sex roles. The theories most relevant to the objectives of the present study are briefly reviewed in an attempt to bring into focus three categories of factors to be investigated: (1) personal factors in Kohlberg's theory of sex-role development; (2) family factors in social learning and identification theories; and (3) societal factors that are viewed as important in the socialization process in general and in formation of sex—role concepts and attitudes. Personal Factors Personal factors are emphasized in Kohlberg's cognitive developmental theory of sex—role identity (1966). This theory is based on the Piagetian notion of cognitive growth as a developmental process in which basic changes or qualitative differences in mode of thinking lead to transformed perceptions of the self and the social world. Kohlberg assumes that basic social attitudes are not patterned directly by either biological instincts or 19 arbitrary cultural norms, but by the child's cognitive organization of the social world along sex-role dimensions. The development of basic universal conceptions of gender and role in children occurs in three different stages. First, there is the development of constant gender categories; second, the development of awareness of genital differences; and third, the development of diffuse masculine- faminine stereotypes based largely on the connotations of non-genital body imagery (Kohlberg, 1966). Kohlberg maintains that these concepts and identi- ties generate masculine—feminine values, on the basis of the assumption that the child engages in spontaneous evaluations of self-worth and the worth of others, and that the child has natural tendencies to ascribe worth to self, to seek worth, to compare self-worth with that of others, and to evaluate others. Kohlberg challenges the old notion that the child's judgments of self and others are internalizations of parental-cultural judgments. He admits that judgments of value by others as well as associated rewards are influential factors in determining the child's valuing, but they are not sole nor basic in origin. The interest of this theory lies in the fact that it raises questions with regard to some other well known and accepted theories of sex-role develOpment. Kohlberg's theory provides a basis 20 for including personal variables as partial explanations of differences in attitudes toward women's roles in the present study. Family Factors Family factors are more central in the theories of identification and of social-learning. The identification theory, also called the psychoanalytic theory, is based on Freud's concept of identification (1949). Because of the constant association of the mother with rewarding experi- ences of infancy, the child develops a strong love for her. Both boys and girls form their first identification with the mother. The girl retains her initial identification with the mother, while the boy has to shift his to a masculine identification if he is to develop a normally masculine personality. Mowrer has a slightly different hypothesis called developmental identification (1960). He maintains that love and affection for the model are the principal factors instigating identification. If the child has rewarding interactions with a parent, that parent's behavior acquires positive value. The child being dependent on the nurturant parent, feels frustrated when he is absent. By performing some of the acts ordinarily executed by that parent, the child is able to provide himself with the rewarding feelings 21 associated with the parent's presence. According to this hypothesis, the mother is also the first source of identi- fication for both boys and girls. The shift to father- identification is based on love and affection for the father. This generally occurs at the age of three or four when, Mowrer says, the good father tends to take his son on, to accept responsibilities for him in a way that he does not do for his daughter. In social-learning theory, sex-typed behaviors are defined as behaviors that typically elicit different rewards for one sex than for the other. They have consequences that vary with the sex of the performer. Sex-typing is defined by Mischel as "the process by which the individual acquires sex-typed behavior patterns; first he learns to discriminate between sex-typed behavior patterns, then to generalize from these specific learning experiences and finally to perform sex-typed behaviors" (1966, p. 57). Mussen explains the principle of generalization which is particularly relevant to this process. It states that when a response has been learned to one stimulus, it is likely to occur in response to other similar stimuli (1969). In the learning of sex-typed behavior for example, if the boy is rewarded for expression of aggression toward other children in his own backyard, he is likely to behave aggressively in his interactions with children in other 22 situations. As the cognitive development progresses, the child forms concepts and attaches labels to objects and events. These labels may then serve as the bases for further generalization. Imitation as a process of learning sex-typed behaviors is another concept that has retained attention. A vast amount of research supports the idea that by observ- ing a model's behavior, the child may acquire sex-typed responses. Bandura (1963) and his colleagues have found that powerful and nurturant models were more likely to be imitated than models lacking these qualities. They con- ducted an experiment showing that cross-sex imitation occurs and is facilitated when the opposite-sex model has greater power than the same-sex model. Brim (1960) also reported some evidence of learning by imitation. His research supported the hypothesis that cross-sex siblings have more traits of the opposite-sex than have the same-sex siblings. The two theories briefly reviewed, identification and social-learning, underline the role played by parents in the formation of sex-role concepts. They suggest that the family characteristics can be viewed as another source in explaining variations in attitudes toward women's roles. 23 Societal Factors Societal factors are also considered important in the formation of sex-role concepts (D'Andrade, 1966). Their influence is mediated through the family in early childhood, but as the child becomes less dependent upon his parents, the community, the school, and the peer groups are other agents sharing with the family the socializing function. According to Brim and Wheeler (1966), some of the most significant aspects of sex—typing and socialization are affected by extra-familial variables and societal influences that occur outside the immediate family and throughout the life cycle. Several studies have explored the impact of the school environment in orienting the values and attitudes of college students. Feldman and Newcomb (1972) reviewed research on college effects and formulated several general- izations. They suggested that peers seemed to play a key role in personality change within the college environment. They also stressed that colleges tend to attract students having distinguishable sets of characteristics which result in a certain degree of homogeneity within colleges. The major impact of the college experience, according to these researchers, is to accentuate the characteristics of an individual that selectively propel him toward particular educational settings. The conditions for campus-wide 24 impact appear to have been most frequently provided in small residential four-year colleges. Feldman and Newcomb con- cluded their study by writing that "processes of attracting and selecting students are interdependent with processes of impact" (p. 218). These findings suggest the introduction of insti- tutional factors as a third group of variables in this study of college students' attitudes toward women's roles. Conceptual Models of Women's Roles The contemporary scene is witnessing the production of innumerable works related to the status and roles of women. A common characteristic of these writings is a tendency to structure opinions and attitudes along "tra- ditional" versus "modern" or "liberal" conceptions of women's roles. A basic assumption underlying this con- ceptualization seems to be that the norms that regulate women's roles have their origin in two different cultures, one inherited from an agricultural society and the other evolving from an industrial and post-industrial world. Carisse (1975), in her study of innovative women in Quebec, analyzed the major writings about women repre— senting both cultural heritages. The first profile emerging from the works in the traditional vein (Deutsch, 1945; Von Le Fort, 1948) portrays the ideal woman as a mother before 25 everything else; her "feminine" activities are mainly oriented toward the expression of her maternal "instinct" and those that are apt to promote the blossoming of her "feminine essence." This ideal creature is "other oriented" in the sense that she was born for the service of others and she can only find her pleasure in someone else's pleasure. She is not seen on the academic scene nor in the political arena and is devoided of any desire for self-expression. This caricatural model does not exist in its pure form in real life; it illustrates, however, the internal consistency of the cultural orientation of those who subscribe to it, to whatever extent. The radical form of the other orientation found expression through the writings of a generation of feminists with Kate Millett (1970) and Greer (1971) as forerunners. A milder version of this orientation had its origin in France with de Beauvoir's Le deuxieme sexe (1949) echoed in the United States in Betty Friedan's The Feminine Mystique (1963) and followed by Sullerot (1966) and Grégoire (1970). A model of the "modern" woman resulted from Carisse's analysis of these works. The woman of this new era claims the right to exist for herself and no longer seeks her identity through a husband or a child. She requests the right to self-expression and self-development, the right to enter into relationship with men with equality of status, 26 in marriage as well as in professional or socio-political activities, the right to be considered as a first-class citizen, with full rights and responsibilities. Contrarily to her traditional homologue, this ideal "modern" woman is self-oriented, in the sense that she has an existence of her own, and as such, has a right to self-actualization. This generation of women is often seen as revolutionary because they go further than feminists of earlier periods. They reject old stereotypes imposed upon women, they go beyond denouncements of inequalities and claims of equal status for men and women. Not only do they explain the origin of these social and cultural disparities, but they seek solutions at the society level that call for transformations of structures as well as ideologies (Carisse, p. 65). Between the two poles of this continuum of traditionalism-modernity, there is obviously a middle- of-the-road position where a large proportion of women are likely to be found, with all the tensions created by the convergence of two sets of opposing values. This position is held by women who try to compromise between the two orientations, and is well exemplified by the dual-career family phenomenon (Rapoport and Rapoport, 1971). Carisse's conceptualization of women's roles is in agreement with several other researchers who studied 27 sex-roles, in Canada (Hobart, 1972); in the United States (Steinman, 1963; Kammeyer, 1964: Epstein and Bronsfat, 1972; Lipman and Blumen, 1973; Osmond and Martin, 1975; Brogan and Kutner, 1976); in Finland (Haavio-Mannila, 1967); and in Sweden (Dahlster, 1962). Steinman's study is, among several, a good illustration of a similar conceptualization. Steinman (1963) used a continuum that she labelled "traditional- liberal." The traditional concept of the feminine role was defined as the concept held by a woman who conceives of her- self as the other, the counterpoint of the man and children in her life. The liberal concept of the feminine role is that concept held by the woman who seeks fulfillment through her own accomplishments. In her study, the traditional woman was called "other-oriented" while the liberal woman was called "self-oriented." Osmond and Martin (1975), in their conceptualization of sex—roles insisted on the degree of stereotyping of sex roles as the criterion that differentiates between tradi- tional and modern attitudes. In the operationalization of the concepts and presentation of results, they emphasized the equality-inequality dimension of sex roles, although they claimed not to subscribe to this conceptualization. This distinction seemed to be a taxonomic problem: the equality-inequality dichotomy applied to sex roles appears to be one of the features of stereotypic thinking. 28 In the present study, the concept of traditionalism versus liberalism as used by Carisse underlies the opera— tionalization of attitudes toward women's roles. There is an apparent incongruence in using such dichotomization of concepts and orientations, particularly when stereotyping and tendency to think in dichotomous style are seen as part of the definition of one term of the dichotomy. This feeling may be somewhat reduced by viewing the opposite terms as two ends of a continuum on which individuals occupy different positions. It may also be useful to state there is no moral evaluation attached to these positions, such as modernism is "good" or "bad." The terms traditional and liberal are used in a sociological meaning; either orientation may be functional or dysfunc- tional for different individuals because of complex factors that are beyond the domain of the present study. One more commentary is in order to explain the rationale for using the concept of attitudes toward women's roles as the focus of the present study. Attitudes are viewed as the predispositions of an individual to respond evaluatively to some object or issue. Attitude is a useful concept, because it serves as a simple, manageable repre- sentation of something quite complex: a brief summary of what has gone before in the individual's experience that may affect his present behavior (Hollander and Hunt, 1967). 29 Attitude is also seen as a neutral concept which permits virtually any and all content to be contained in it (Hollander and Hunt, 1967). Katz (1959) mentioned that opinion is the verbal expression of attitudes, that atti- tudes have both an affective, or feeling core of liking and disliking, and cognitive, or belief elements. Attitudes are also considered as a range of positions on a continuum. Women's roles constitute an issue that elicits affective and cognitive reactions. Attitudes toward women's roles seem to be an appropriate expression in the context of the present study. CHAPTER III RELATED RESEARCH The following chapter contains a review of research concerning sex roles and is divided into two parts: (1) studies dealing with sex-role attitudes; and (2) studies dealing with the factors influencing attitudes toward women's roles. In both instances, more attention will be given to studies conducted among college students. Sex-Role Attitudes Trends in Sex-Role Attitudes Some researchers have attempted to assess the degree of change in attitudes toward women's roles. Komarovsky compared the results of two studies of women's role attitudes among undergraduate women conducted on the same campus, one in 1943 and the other in 1971. She found that the proportion of career oriented women had remained constant between the two periods studied. The same researcher (1973) had been interested in the attitudes of male college students toward future wives‘ occupational roles. Their attitudes were heavily traditional; one-fourth of the sample were outrightly traditional while only 7 3O 31 percent expressed feminist attitudes. Komarovsky also reported that 30 percent of these men expressed feelings of intellectual insecurity and strain in their relationship with their female friends of superior intellectual ability. This researcher concluded her study by stressing ambiva- lences and inconsistencies in attitudes toward women's roles. Parelius (1975) studied changes in sex-role attitudes of college women between 1969 and 1973. Her data showed that a substantial shift toward feminism occurred over that period. Changes toward the marital and maternal roles were less pronounced than those related to occupational activities and equal rights and duties for both sexes. These women, while rejecting economic depen- dence and unalieviated household responsibilities, showed positive attitudes toward marriage and motherhood. Similar findings were reported by Epstein and Bronsfat (1972) when they compared female college freshmen of the year 1970 with those of the year 1965. Male-Female Differences in Sex-Role Attitudés A number of studies have dealt with sex differences in sex-role concepts and attitudes. Hartley (1968) con- ducted a series of studies to investigate children's conceptions of sex roles. She compared boys and girls 32 of three age levels on several aspects of sex-roles. There were 47 boys and 110 girls with working and nonworking mothers. She found that a large proportion of children see the roles of men and women clearly and traditionally differentiated. She observed a tendency in boys to be more sensitive to female appropriate behaviors than girls are to male appropriate role behaviors. Boys also perceived more participation of males in traditionally female role activ- ities than do girls. The proportion of girls who indicated that they planned to work was significantly higher than the proportion of boys who indicated that they might consent to their wives' working. Hartley's studies showed that the work status of the mother seemed to have an impact on boys different from that of girls. This supports the fact that the literature contains contradictions in directionality of impact. Haavio-Mannila (1967) compared men's and women's attitudes toward sex roles with a focus on women's roles, among two subcultural groups in Finland. The differences between men and women were not very great but showed that men were more traditional than women; they were more patriarchally oriented and wanted women to stay at home. More men than women agreed to leave the housework to women in all age groups, but the middle categories showed the greatest differences. Swedish-speaking respondents were 33 more modern than Finnish-speaking respondents, and this was especially true for Swedish-speaking women. The results of this study indicated that the traditional division of labor between the sexes persists in most families although two—thirds of the married women are working outside the home. There were also some evidences of the persistence of "double standards" with respect to norms prescribing the apprOpriate behavior for men and women: temperance, church attendance, marital fidelity and cultural interests were more often expected from women than from men. Hobart (1972) in an attitude study of French- and English—speaking Canadians, compared male and female stu- dents on the degree of egalitarianism versus traditionalism for several areas of marital role expectations. Women were found more egalitarian than men on the total score. Further sex comparisons between the two ethnic groups showed that the patterns vary by areas of role considered: English- speaking male sample members were more egalitarian than their female peers in the area of household management; in the area of the wife's role, including her authority, housework and gainful employment, the French-speaking respondents were found more egalitarian, but it was entirely due to the strong egalitarianism of the women. French- Canadian women appeared to be pioneers in advocating egalitarian and permissive orientations, in comparison 34 with English-speaking respondents, while French-Canadian men were the most traditional of all. Hobart suggested that these sex differences might be explained by the different stages of the feminine revolution, the movement being sufficiently advanced in the English-speaking pop— ulation to have affected men as well as women. These findings are in agreement with Haavio-Mannila's and point to the need to study cultural influences on the women's role in society. Osmond and Martin (1975) also compared male and female sex-role attitudes among college students. They reported that the sex-role issues over which men and women disagree the most were the extra-familial female roles and the stereotypes of both sexes. The males in the sample seemed to be least willing to accept women in leadership roles, while women appeared to be insistent that women should occupy these positions and can perform in them as well as men can. These researchers observed some inconsistencies in the males' attitudes: the same men who approved most sex-role stereotypes were at the same time supportive of social changes which would benefit women and unsupportive of women's occupying leadership positions outside the family. 35 Factors Influencing Sex—Role Attitudes This section is concerned with the factors that influence attitudes toward sex roles. The studies selected for this review are of two types: some are comprehensive and include a large number of variables, while others tend to focus on one or two specific factors. This discussion focuses on the variables that are most pertinent to the objectives of the present study. Several studies investigated the effect of the occupational status of the mother on the deve10pment of sex-role concepts and attitudes in her children. Hartley (1968) studied the influence of the employment of the mother on the vocational choice of the daughters. She reported that more daughters of nonworking mothers chose the vocation of "house-wife" for adulthood, and more daughters of working mothers mentioned professional and vocational choices and thought that they would continue to work after marriage. In a study of women's role orientation among college women in the Province of Quebec, Préfontaine (1969) observed that daughters of working mothers were more likely to be oriented toward professional roles while daughters of nonworking mothers were more inclined toward the family role. A study was conducted by Francis (1975) in New Zealand to investigate the effect of maternal 36 employment on the socialization of sex role and achievement. She found that daughters of working mothers perceived their mothers as being more capable than did daughters of unemployed women. The purpose of Tangri's study was to examine the relationship between role innovation among college women and family background, personality and college experience (1972). She reported data illustrating the role modeling effect of maternal employment on daughter's role orientation. Among the relationships studied, the strongest ones were with the present employment of the mother and the masculine character of her occupation. The data showed that role innovative women did not see themselves as masculine, nor did they reject the core female roles of wife and mother; they did expect to postpone marriage and have fewer children than more traditional women. In the same series of investigations, Vogel and her associates (1970) demonstrated the relationship between sex-role perceptions of college students and mother's employment status. The sex-role questionnaire used in this study was based on the concept of sex-role stereotypes. Stereotypic items determined a "competency" cluster sup- posedly characteristic of men and a "warmth-expressiveness" cluster supposedly characteristic of women. Daughters of employed women perceived smaller differences between men 37 and women than did daughters of homemaker women, on both clusters. Sons of employed mothers perceived smaller differences between men and women on the "warmth- expressiveness" cluster than did sons of homemaker mothers, but the perceptions of these two groups of men did not differ significantly on the "competency" cluster. Daughters of employed mothers tended to have a more positive perception of women than did daughters of homemaker mothers. The same sex-role questionnaire was used by Clarkson and his associates (1970) to investigate the relationship between self-perception in the context of stereotypic sex roles and the number of children a woman has. The findings indicated that women who hold relatively masculine self- concepts have significantly smaller completed families than women who hold more stereotypically feminine self-concepts. The subjects were Catholic mothers of male college students. The same researchers suggested on the basis of preliminary analyses of data on young college women, that college women with relatively high competency self—concepts perceived their ideal future family size as significantly smaller than college women who perceived themselves as relatively less competent. Meier (1972) studied the relationship of sex-role attitudes to parental characteristics. The results of his 38 study revealed a pronounced positive association of relatively egalitarian attitudes toward women's roles with salience of the mother in attitudinal socialization, the mother being college educated and the mother's involve- ment in occupational role outside the home. There was also a relationship between feminine social equality orientation and the sex of the student. Female undergraduates tended to have more egalitarian attitudes than their male peers, and this sex variation persisted at a stable level in all multivariate analyses, indicating its independence of the parental variables. These findings are somewhat at variance with Kutner and Brogan who used an approach similar to the one used by Meier. These researchers reported no relation- ship between sex-role orientation and salience of mother in attitudinal influence. Their data, however, paralleled Meier's findings about mother's educational attainment and employment (1976). In the 1972 survey of full—time freshmen college students, conducted by the American Council on Education and reported by Bayer (1975), data were obtained from 188,900 students in 373 institutions. Approximately 16 percent of the men and 9 percent of the women were labeled as sexist. The students who indicated that they strongly agreed to the statement that "the activities of married women are best confined to the home and family" were placed 39 in this category. A profile of the sexist student emerged from the study of several characteristics of these students as compared to the average freshmen. The sexist student was more likely to be somewhat older than the freshmen peer group, to come from lower socioeconomic families, to be nonurban in origin and to be less successful academically. Students who endorse traditional women's roles were more likely to subscribe to marrying soon and to raising a family as a primary life objective. They were also less likely to support equal opportunity in women's employment. Although the identification of sexist attitudes was based on one single item, and this seemed to be an index of questionable validity, these findings may suggest that there is a sub- stantial proportion of young adults who tend to subscribe to traditional roles. Hjelle and Butterfield studied women's attitudes toward their roles in relation to self-actualization (1974). Their subjects were single female full—time students between 17 and 20 years of age. Among the instruments used were Shostrom's Personal Orientation Inventory measure of self- actualization. The findings indicated that college-aged females holding pro—feminist attitudes vis—a-vis their social and sexual roles exhibited a higher degree of self- actualization than their counterparts expressing more traditional attitudes. These researchers suggested that 40 in the socialization experiences of the first group, early attempts to resist internalization of traditional behaviors were possibly met with parental approval. Kaley (1971) studied attitudes of professional men and women toward the married professional dual role. For the variables of sex and profession, she found significant differences within the group, but for age, education, race and professional experience, there were no significant differences. Hobart (1972) also studied a number of social and family characteristics in his study of sex-role attitudes among French- and English-speaking students in Canada. He found that whether a person adopts a more traditional or more egalitarian orientation appears to depend on person- ality or on a random combination of circumstances rather than upon the kind of predictors that sociologists usually find associated with attitudinal variables. This review of relatively recent studies of sex-role attitudes points to a certain number of constants: (1) There are indices of the existence of clearly defined sex— role stereotypes for men and women; (2) men and women are perceived as having different traits and masculine traits are more often perceived to be desirable than are feminine traits; (3) women are faced with strain that results because different standards exist for adult behavior and for female 41 behavior; (4) men and women differ in their attitudes toward sex-roles and women seemed to be less traditional than men; and (5) among different factors associated with sex-role attitudes, sex is one of the most powerful predictors of attitudes and the social background characteristics are the least powerful in such prediction. CHAPTER IV DESIGN OF THE STUDY In this chapter, the population and the sample, the data collection procedures, the instruments used, and the statistical treatment of the data will be described. Population and Sample The population selected for the present study on attitudes toward women's roles was the French-speaking college students in New Brunswick, Canada. This population is concentrated in the three institutions described in Chapter I. These are located in the three geographical regions of this province where the French-speaking popu- lation is found. In order to obtain comparable groups of students, two criteria were used for eligibility: the sub- jects should be freshmen and enrolled in a degree program. Freshmen were selected because the number of students in the affiliated colleges is larger in the first year than it is in the more advanced years. Another reason for the selection of freshmen was to avoid the likelihood of reaching students who might have been over-used as 42 43 research subjects, as can be the case for the students in the more advanced years in a small university. A random sample of 20 percent of freshmen from the Moncton campus was drawn from the registrar's list of regular students enrolled in degree programs. All freshmen on this list (750 students) were assigned a number and a random table was used for the drawing until the total of 150 subjects (or 20 percent) was attained. In the affil— iated college of Edmundston all freshmen enrolled in a degree program were included which provided a list of 125 names. Using the same criteria for the selection of cases from Shippegan College, produced a sample of 30 subjects. Data Collection The data were collected during the months of April and May 1975. For the campus of Moncton, after obtaining the necessary authorization, the questionnaires were dis— tributed in the dormitories or mailed to the students living off campus. Prior to sending the questionnaire to the students in the two affiliated colleges, a personal contact was made with the authorities of these institutions to inform them of the purpose of the study and to solicit their cooperation. The student directory of the Edmundston campus was obtained which contained all needed information for contacting subjects through the mail. For the Shippegan 44 campus, the questionnaire was administered to the students within the college by a faculty member who volunteered to take this responsibility. A letter of introduction, different for each sex was attached to the questionnaire, explaining broadly the purpose of the study. The letter stressed that replies would be treated with confidentiality. A stamped envelope was provided to the subjects from the southeast and north- west campuses for the return of the questionnaire. About a month after the first distribution, a letter of recall was sent to the subjects who did not return the first one. Out of a total of 305 students drawn in the sampling procedure, 295 were contactable. Within a month and a half, 234 subjects, or 77 percent returned usable questionnaires. This was considered a high rate of return, considering the particular time of the year when students' addresses were likely to change; moreover a mail strike occurred within the data collection period. A certain number of subjects expressed their personal interest in the study by a short comment at the end of their responses, or a note accompany- ing the questionnaire, or even a personal appearance in the investigator's office. 45 Development of Measures The research design comprised attitudes toward women's roles as the dependent variable, and personal, family, and institutional factors as independent variables. The development of measures and the scoring procedures for these variables are now described. Measure of Attitudes Toward Women's Roles The instrument used for the measurement of attitudes toward women's roles consisted of 30 items on a Likert-type summated rating scale. This scale was constructed with the aim of distinguishing between subjects holding liberal atti- tudes from those holding traditional attitudes. A cursory review of available instruments suggested that selecting from these instruments the items that met the broad objec- tive of this study would be more appropriate than using one that had a narrower and more specific perspective. It was also felt that cultural differences imposed certain restric- tions on the use of instruments developed in another cultural context. The content of the scale was borrowed from several sources: Dunn's Marriage Expectation Inventory (1960) and a scale developed by Prefontaine (1969) provided approximately half of the items; the content of the other items was sug— gested by the current literature on the changing roles of women in society. 46 To increase the face validity of the items, two faculty members and three advanced students in the Ecole des Sciences domestiques at the Université de Moncton, were asked to react to the formulation of the statements. Their suggestions were used to clarify meanings and to eliminate items that raised doubts about their potential to measure attitudes towards women's roles. The original scale con- tained items pertaining to five different areas of women's roles: the familial, professional, intellectual, socio- political areas, and the division of responsibilities between men and women. This instrument was pretested on a sample of 45 male and female freshmen enrolled at the University of Moncton. The pretests were distributed in one Home Economics class and one Education class.- Based on the pretest, revisions were made to discard nondiscrim- inating items; a criterion of 80 percent or higher of sim- ilar responses was used as a basis for deletion of an item. In this process, most of the items related to the involve- ment of women in socio-political activities elicited a high degree of favorable attitudes among the respondents; this was interpreted as an indication of built-in social desir- ability and consequently, it was decided to omit these items from the scale. French and English versions of the final scale appear in Appendix A. 47 To evaluate the internal consistency of the scale, a test of reliability using the alpha model was performed. The coefficient alpha indicates the average interitem correlation of all items constituting the scale (Crano and Brewer, 1973). It also indicates that the individual items on the test were producing similar patterns of responses in different people. A reliability coefficient of .72 was obtained on the 30-item scale; it was decided to eliminate the items having an item-to-total correlation below .20. This criterion reduced the number of items to 18 and increased the reliability coefficient to .78. The scale used for the computation of the score for each respondent, and for the subsequent analyses, consisted of these 18 items that met the criterion. These are marked with an asterisk in the copy that is shown in Appendix A. Students were asked to express their agreement with the statements on a four-point continuum. A value of one was assigned to a response indicative of a traditional attitude, and a value of four was assigned to a response indicative of a liberal attitude. A recoding of the answers was necessary on the items for which a liberal attitude was indicated by an agreement with the statement. An alternance of items for which agreement meant a traditional attitude with items for which agreement meant a liberal attitude was observed to avoid the possibility of conveying a bias toward 48 one or the other orientation. The score for each subject was obtained by summing the values for the individual items, with possible scores ranging from 18 to 72. A higher score on the scale revealed the presence of liberal attitudes. These scores gave each subject a position on the continuum of traditionalism—liberalism with regard to attitudes toward women's roles. The distribution of scores on the women's role attitude scale was approximately normal since the mean and the median were almost identical. The negative kurtosis indicated a somewhat flatter than normal distribution, and the positive skewness indicated that the cases tended to cluster more to the left of the mean with most of the extreme values to the right (see Table 1). For some statistical procedures, the raw scores on the Women's Role Attitude Scale were transformed in cate- gorical scores (see Table 3). The cutting points for the grouping of the scores were selected on the basis of the median figure instead of the mean because of the skewness of the scale (Blalock, 1970). These points correspond to one-half a standard deviation on each side of the median. The lowest scores ranging from 30 to 40 formed the "tradi- tional" category; the middle scores ranging from 41 to 48 formed the "moderate“ category; the highest scores ranging from 49 to 67 formed the "liberal" category. The frequency 49 Table 1 Descriptive Statistics for WOmen's Role Attitude Scale Number Descriptive Statistic (N==234) Measure of Central Tendency Mean 45.8 Median 44.7 Measure of Dispersion Variance 64.69 Standard deviation 8.04 Standard error 0.526 Kurtosis -0.65 Skewness 0.32 Measure of Reliability Alpha coefficient 0.78 distribution of raw scores and of categorical scores are shown in Tables 2 and 3, respectively. Measures of Personal Characteristics Self-actualization. This aspect of personality was measured using two scales from the Personal Orientation Inventory. This instrument developed by Shostrom (1964) consists of 150 two-choice comparative judgments. It has been known as a valid measure of self-actualization. The scoring procedure of this test allows for a series of 10 50 Table 2 Frequency Distribution of Raw Scores on the Women's Role Attitude Scale Raw Score Number Percent 30 l 4.0 31 3 1.3 33 3 1.3 34 5 2.1 35 7 3.0 36 12 5.1 37 10 4.3 38 12 5.1 39 5 2.1 40 9 3.8 41 13 5.6 42 10 4.3 43 17 3.0 44 16 6.8 45 14 6.0 46 5 2.1 47 6 2.6 48 10 4.3 49 6 2.6 50 10 4.3 51 6 2.6 52 10 4.3 53 13 5.6 54 3.8 55 0.4 56 1.7 1.7 1.7 1.3 2.6 1.3 1.3 0.9 0.4 0.4 ox O HI—awwwmwaani-uo 51 Table 3 Frequency Distribution of Categorical Scores on the Women's Role Attitude Scale Categories Score Range Number Percent Traditional (1) 30-40 67 29 Moderate (2) 41-48 81 35 Liberal (3) 49—67 86 37 subscales from which two were selected. The Contact scale and the Synergy scale were used because the item content of these scales were deemed the most appropriate among the 10 with regard to the objectives of the study. The rationale for including a measure of self-actualization was based on a definition of self-actualization that includes dimensions of personal growth and autonomy that are often referred to in the discussion of women's liberation. The Capacity for Intimate Contact scale measures the ability to develop mean- ingful, contactful relationships with other human beings, unencumbered by expectations and obligations, the ability to express versus impress, being versus pleasing. The Synergy scale measures the ability to transcend dichotomies, to see. opposites of life as meaningfully related. A comparison between the scores found in this sample and those reported 52 by Jenkins (1970) in his study of entering college freshmen, male and females, showed similar results on the same scales. The mean on the Contact scale in the present study was 15.8 as compared to 15.6 in Jenkin's study. On the Synergy scale, the mean was 6.8, while it was 6.3 in the forequoted study. Anomia. A five-item scale of anemia developed by Srole (1956) was used for the measurement of this variable (Appendix A). Academic performance. The measure of academic performance was the subject's report of his latest grade point average. Students were asked to report their grade point average for the previous term of the school year, according to the following categories: 1. Less than 1 2. 1 to 1.4 3. 1.5 to 1.9 4. 2 to 2.4 5. 2.5 to 2.9 6. 3 to 3.4 7. 3.5 and over No attempt was made to check the accuracy of the subjects' responses against their official records. However, the frequencies observed in each category followed the general distribution pattern for the university as a whole. Thus, ‘the self-report measure was deemed sufficiently reliable for the purpose of the study. The responses were finally Classified in three groups: a low academic performance 53 corresponded to responses 1, 2, and 3; a medium academic performance to responses 4 and 5, and a high academic performance to responses 6 and 7. Age. The subjects indicated their age by checking one of the following categories: 1. 17 years 2. 18 years 3. 19 years 4. 20 years 5. 21 years 6. 22 years 7. 23 years and over Since the majority of the responses were in categories 2 and 3, the responses were classified in two groups: (1) under 19 years, and (2) 19 years and over. Residential independence. This variable was measured from the responses to this question: Where are you living during the school year? The subjects were asked to check one of four possible answers: With my parents In the student dormitory In my own apartment Other (specify) kwNI-J Response 1 was categorized as None, response 2 as Some, and responses 3 and 4 as More in relation to independence. Employment experience. Two questions were asked to evaluate the degree of employment experience the subjects had had: (1) Did you work during summer vacation last year? 54 (2) Did you have a part-time job during the school year? The response categories were "Yes" and "No" for both questions. A weight of zero was assigned to a negative answer on both questions; a weight of one was assigned to a positive answer on either question; a weight of two was assigned to a positive answer on both questions. Academic field. The students identified their academic field by checking one of the seven following categories: Humanities Administration Nursing Sciences Education . Home Economics . Social and Behavioral Sciences \JO‘U‘lubUJNl-J O The responses were subsequently grouped in two categories: (1) traditionally masculine fields: responses 2 and 4; (2) traditionally feminine fields: responses 1, 3, 5, 6, and 7. This classification was based on studies reported by Tangri (1972) and Ginzberg (1966). Tangri referred to teaching, nursing, and social work as "feminine" fields; Ginzberg reported a heavy concentration of college women in the humanities and listed education, social work, and public health among the classic female professions. This investigator's perception of the distribution of students in Moncton corresponded to this classification. 55 Measures of Family Characteristics Social prestige of father's occupation. The categories used to obtain information on the father's occupation were adapted from the Pineo and Porter's Index of Occupational Prestige. This index has been develOped on the basis of the findings of a large survey conducted in Canada (1964) and using the same methods as those used in the National Opinion Research Center (NORC) study in the United States (1963). The rankings on the Canadian index correlated .98 with the United States rankings. This index gives a score to each occupational title of the Census of Canada; these occupations are arranged in socioeconomic categories on the basis of the mean score of all titles in each category. In the pretest, on a free—response item on the father's occupation, several answers were difficult to give a score on this index. It was thus decided to use the occupational categories. Minor changes were introduced: the categories semi-professional and semi-skilled were not used, the title technician was added to the category cler- ical and sales, and the titles fisherman and logger were added to the farmer category. These changes were made to make it easier for the respondents to report accurately their father's occupation. The categories used were the following: mummbww 1" 0.00000 56 Professional (physician, lawyer, professor, etc.) Proprietor, manager of large enterprise Proprietor, manager of small enterprise Clerical, sales, technician Skilled worker (electrician, plumber, etc.) Farmer, fisherman, logger Unskilled worker Other (specify) The responses in the last category were interpreted and assigned to the proper category. For the purpose of data analysis, this information was reduced to four levels of occupational prestige: waI-J Responses 6 and 7 Response 5 Responses 3 and 4 Responses 1 and 2 Father's and mother's education. Information on father's and mother's education was obtained from responses to two identical questions: How many years did your father (mother) complete in school? The categories for responses were: U'IbUJNH 0000. 7 years or less Between 8 and 12 years l3, 14, or 15 years 16 years More than 16 years Three educational categories were derived from the responses: 1. 2. 3. Elementary education: response 1 Secondary education: response 2 Post-secondary education: responses 3, 4, and 57 Family income. The level of the family income was classified from the responses to the following question: What was the total income of your family last year? Five response categories were provided: U'I-waI-d 0000. Under $5,000 $5,000 to $7,999 $8,000 to $11,999 $12,000 to $15,999 $16,000 and over The responses were finally divided into three income groups: 1 2 3 High Low income group: responses 1 and 2 Middle income group: response 3 income group: responses 4 and 5 Employment experience of the mother. The measure of this variable employment. The their mother was time in the past one of the first was based on the history of the mother's subjects were asked to indicate whether currently employed, had been employed some or had never been employed. A response in two categories gave the mother a status of working mother; the other response, a status of nonworking mother. Size of the family. The responses to the question asking for the number of children in the family were classified in three categories: 58 1. Small family: 4 children or fewer 2. Average family: 5, 6, or 7 children 3. Large family: 8 children or more Measures of Institutional Factors Size of home town. The categories used to obtain this information were those of the Census of Canada: 1. Rural farm . Rural nonfarm Village with less than 2,500 population Town of 2,500 to 10,000 population City of 10,000 pOpulation and over Ulub-WN .0. These were further reduced to two categories: 1. Community of less than 2,500 population 2. Community of 2,500 and over Region of origin. This information was obtained by asking the subjects to identify the county of their home town, which was first classified according to the following categories: 1. Northeast 2. Northwest 3. Southeast 4. Other This classification was used mainly for descriptive purpose. For data analysis, only two categories were used: 1. Less urbanized region: categories 1, 2, and 4 2. More urbanized region: category 3 Campus size. The subjects also identified the institution attended. They were ranked according to the enrollment size, as described in Chapter I. 1. College of Shippegan 2. College of Edmundston 3. Université de Moncton 59 Statistical Treatment of the Data Nonparametric and parametric models were used for hypothesis testing. Contingency tables were constructed and chi-square tests were performed. A one-tailed test was used and an alpha level of .05 was set to determine levels of significance. The statistic Gamma and Contingency Coefficients were used to evaluate the strength of the asso- ciation between the dependent variable and the independent variables that were measured on ordinal and nominal scales, respectively. Gamma (G) is a coefficient of association between two sets of ordinal variables based on their mutual pre- dictability in terms of the relative number of agreements and inversions in the order of rankings. Gamma may vary from zero to one. It not only gives the strength of asso- ciation but also gives the direction of this association between two sets of variables. A negative sign indicates that the association is based on predominance and inver- sions, while a positive sign indicates the predominance of agreements (Freeman, 1965). The Coefficient of Contingency (C), also called Pearson's coefficient of mean square contingency, is a measure of association that can be applied to nominal scales possessing any number of categories. This statistic 60 goes to zero when there is no association, but its maximum value is .707. This limitation should be taken into con- sideration when interpreting C. Only tables having the same number of columns and rows can be compared, and the value of the coefficient is smaller when the tables are small (Blalock, 1970). A continuous score was available for the variables of self-actualization and academic performance. These variables were analyzed using a parametric statistic, the Pearson's correlation coefficient. Normality of distribu- tion was assumed, given the relatively large size of the sample (N = 234). To determine which variables were the best predictors of attitudes toward women's roles, a multiple regression analysis was used. The forward stepwise regres- sion procedure provides a means of choosing independent variables in an order that gives the best prediction equation possible. This procedure provides the multiple correlation coefficient (R2) and a partial F-test that shows whether the variable accounts for a significant amount of variation over that removed by variables previously in the regression (Nie et a1., 1970). The critical F-value of 2.5 was set for both entry into and deletion of a variable from the equation. CHAPTER V THE FINDINGS Description of the Sample The subjects of this study were freshman students in the French-speaking institutions of higher education in the Province of New Brunswick. They were all enrolled in a four-year program of studies leading to a bachelor's degree. The sample included 234 students, 47 percent of whom were males and 53 percent were females. Their ages ranged from 17 to 23, with a clustering of 81 percent in the categories of 18 and 19 years. A small minority of students were married and a large proportion expected to be married within one to five years. One-third of these students said they were uncertain as to whether they want to be married and another small percentage expressed their intention not to be married. For the most part, these students came from intact families; 20 of them had lost their father and two their mother, while eight reported that their parents were divorced or separated. These students belonged to fairly large families, when compared to the average family size in 61 62 New Brunswick, which is two children per family; 53 percent of the families had five children or more and 27 percent had eight children or more. It is known, however, that French ' Canadians in New Brunswick have larger families than English Canadians. The mother's participation in the labor force was about equal to the average participation rate of married women in Canada which was almost 40 percent in 1974; in this sample of women, 38 percent were employed outside their home at the time of the survey, while 36 percent were reported as having been out of the labor force all their lives. Information on the father's occupation was obtained for 225 of the 234 subjects, which was classified in four occupational groups of approximately equal size: the first category comprised 24 percent of the fathers and these were in professional or managerial occupations; the second was made up of another 24 percent who were technicians, or worked in clerical or sales occupations; the third group comprehended the 25 percent in skilled occupations: the fourth category, including farmers, loggers, fisherman, and unskilled workers, comprised 27 percent. The proportion of fathers in the higher occupational group is greater than the equivalent in the total population of Canada. According to the 1971 census, only 15 percent of the workers were in professional or managerial occupations. 63 The data on parents' education showed that 37 percent of the fathers had a schooling of elementary level and 30 percent of the mothers had this same level. While more fathers than mothers were in the lowest educational category, a greater proportion of fathers had an education of post-secondary level. A cross-classification of fathers' and mothers' schooling indicated that about 55 percent of the parents had an equivalent level of education, 25 percent of the mothers had an education of a higher level than the fathers, and 20 percent of the fathers had attained a higher level than the mothers. These different comparisons showed that, on the whole, educational attainment for the parents followed a distribution pattern between men and women similar to that found on the national scale. It was also observed that the pr0portion of fathers in the lowest category was approximately the same as that of the Atlantic Provinces, 37 percent against 36 percent, while the pro- portion of mothers was closer to the national figure, 30 percent as compared to 29 percent. On the other hand, a slightly greater proportion of fathers were in the highest educational category than the general population in Canada: 20 percent in the sample had a post-secondary education as compared to 18 percent for Canada and 16 percent of the Atlantic Provinces. 64 Information on the family income was obtained from 222 subjects out of the 234 in the sample. These data revealed that approximately one-third of these families had an income above the average figure in the Atlantic Provinces, which was $11,454 for the year 1974. At the same time a large proportion of students came from families in the low income category: 36 percent of them reported a family income below $8,000. This figure becomes even more meaningful when one is reminded of the large size of these families. In describing the size of communities from which the subjects came, it may be important to consider that almost half of the total population of New Brunswick lived in places of less than 1,000 persons (Macpherson, 1972). According to the 1971 Census, 43 percent of the population lived in rural areas but only 4 percent were considered as farm residents. Analyzing the rural component of New Brunswick, Macpherson (1972) made the hypothesis that the figures might simply mean an abandonment of farming while retaining the farmhouse as a family dwelling. On the urban scene, only two cities have populations above 25,000 persons. The families of the subjects were almost equally distributed among the four community types. One-fourth of the families lived in a rural area, although only 2 percent 65 were farm residents; there were 24 percent living in a village, 24 percent in towns of 2,500 to 10,000, and 27 percent in a city of 10,000 or more. A summary of the variables used to describe the sample of students is found in Table 4. Results of Data Analysis In this section the results of the study on atti- tudes toward women's roles are presented. These results are reported under the following subheadings: (1) Women's Role Attitude Scores; (2) Personal Characteristics and Attitudes Toward WOmen's Roles; (3) Family Characteristics and Attitudes Toward Women's Roles; and (4) Institutional Factors and Attitudes Toward Women's Roles. WOmen's Role Attitude Scores One objective of this study was to estimate the relative degree of liberalization of attitudes toward women's roles among college students. The scores on the attitudinal scale were used to give each respondent a position on a continuum of traditionalism-liberalism. The scores ranged from 30 to 67. The frequency distribution is illustrated in Figure 3. The procedure described in Chapter IV with regard to the grouping of scores in three categories was implemented (56 Table 4 Description of the Sample of Students Descriptive Variables Number of Students Percenta Sex: Male 111 47 Female 123 53 Age: 17 years 8 3 18 years 103 47 19 years 80 34 20 years and over 42 16 Grade Point Average: Less than 2 71 31 2 to 2.9 108 38 _73 and over 49 21 Marriage Plans: Already married 6 3 Within 1 or 2 years 21 9 Within 3 to 5 years 113 49 No plan to marry 11 5 Do not know 80 35 Institution’Attended: University of Moncton 124 53 College of Edmundston 89 38 College of Shippegan 21 9 Region of Origin: Northwest 93 40 Northeast 72 31 Southeast 48 20 Other 21 9 Type of Home Town: Rural 59 25 Village 56 24 Town of 2,500 to 10,000 55 24 _ City of 10,000 and over 62 27 Family Income: $4,000 to $7,999 82 36 $8,000 to $11,999 67 30 $12,000 to $15,999 49 22 _fi§16,000 and over 24 11 Family §ize: 1 to 4 children 85 37 5 to 7 children 84 36 8 to 10 children 45 20 __10 children and over 17 7 Father's Obcupation: Professional-managerial 52 24 Sales, clerical, technicians 50 24 Skilled workers 54 25 Farmers, loggers, fishermen 25 12 _fiynskilled workers 31;; 15 Father's schooling: Less than 8 years 87 37 8 to 12 years 95 41 13 to 15 years 27 12 16 years 5 2 More than 16 years 18 8 fiSther'E’SChoOling: Less than 8 years 70 30 8 to 12 years 118 51 13 to 15 years 31 13 16 years 6 3 _TMore than 16 years_i 7 3 Mother's Empléyment Experience: Presently employed 91 39 Employed in the past 84 36 Never been employed 58 25 aDue to rounding of the decimals, the percentages may not total 100. 67 and produced the results that are shown in Figure 4. A higher percentage of students were found in the liberal category than in the other two categories. The traditional category was the largest among the men with two-fifths of this group of respondents; the moderate category was next with almost one-third; and the liberal category was the smallest with about one-fourth of the men. The reverse proportion was observed among the female students. Almost one-half of the women were in the liberal category, while less than one-fifth were in the traditional category and about one-third in the moderate category. These results indicated that a wide range of attitudes exist among students, and there seemed to be a trend toward liberal attitudes among the women students. Personal Characteristics and Attitudes Toward Women's Roles A general hypothesis regarding the personal characteristics of the students is as follows: H1: A significant association exists between a student's position on the traditionalism- 1iberalism continuum regarding attitudes toward women's roles and selected personal characteristics. To study the association between attitudes toward women's roles and selected personal characteristics, nine subhypotheses were posed. No. of Subjects HUMAUIG‘QQO I Figure 3. Scores Frequency Distribution of Scores on the Women's Role Attitude Scale. Composite Female traditional moderate liberal Figure 4. Distribution of Respondents by Traditional, Moderate and Liberal Categories by Sex 69 A more liberal position is positively associated with the feminine sex. H1.1 Several studies indicated that sex is an important variable in predicting attitudes toward sex roles (Haavio- Mannila, 1967; Kaley, 1971). Some of the more recent studies, however, reported changes in the men's attitudes, at least in expressing their views on feminist issues (Bernard, 1974). As can be seen in Table 5, the trend observed in the earlier studies was still noticeable in this sample. There was a significant association between sex and atti- tudes toward women's roles. A higher percentage of women had liberal attitudes than men had, 45 percent as compared to 28 percent (C==.24; p==.0002). Hypothesis 1.1 was supported. Table 5 Association Between Attitudes Toward Women's Roles and Sex Attitudes Sex Traditional Moderate Liberal 2 i B i B. 1 Male (N = 111) 45 40 35 32 31 28 Female (N = 123) 22 18 46 37 55 45 70 The results of a substantial association between sex and attitudes toward women's roles suggested that different patterns of association between the dependent variable and the other independent variables might evolve if the data for male and female subsamples were analyzed separately. Thus, it was decided to perform separate tests on both male and female subsamples, in addition to the test performed on the total sample for all other independent variables. The results are presented together and combined in one table for the three tests performed on each inde- pendent variable. Three of the nine personal variables studied were measured with continuous scores and analyzed using a Pearson's correlation procedure. The results of these analyses are presented as a group. H1 2 A more liberal position is positively associated ' with self-actualization. H1 3 A more liberal position is negatively associated ’ with anomia. H1 4 A more liberal position is positively associated ° with academic performance. Two aspects of self-actualization were measured using two scales from Shostrom's Personal Orientation Inventory (1963): the Capacity for Intimate Contact scale-- subsequently called Contact——and the Synergy scale described in Chapter IV. The Pearson's correlation coefficient between 71 Contact and attitudes toward women's roles was .20. This moderate association, significant at the .001 level, indi- cated that college students who scored high on the Contact scale tended to exhibit more liberal attitudes toward women's roles than those who scored low on this scale. A weaker association (r==.09; p==.081) was found between Synergy and attitudes toward women's roles. Men and women were compared on the scales of self-actualization. It was found that the association held for both groups on the Con- tact scale (Male: r= .22; p= .01; Female: r= .25, p=.01). The association between Synergy and attitudes toward women's roles slightly increased for the males (r==.l4) while it decreased for females (r==.05). Therefore, H1.2 was partially supported. An ambiguous pattern of association was found between anomia and attitudes toward women's roles. The direction of association differed between males and females. When the data were analyzed as a whole, the Pearson corre- lation coefficient suggested that there was no association between these two variables (r==.03). The same statistic performed on each subsample resulted in a positive asso— ciation (r==.14; p==.017) in the female sample, while in the male sample it was negative (r==-.05) although non— significant. Consequently, H1 3 was not supported. 72 Academic performance was measured by the latest grade point average as reported by the subjects. The Pearson's correlation coefficient between academic per- formance and attitudes toward women's roles was .11 (p==.044) for the total sample. The pattern of association was similarly positive but still lower when males and females were considered separately; the coefficients were .06 and .08, respectively. Therefore, H1.4 was weakly supported. A correlation matrix in Table 6 shows the inter- relationships among these variables. It can be observed that the relationship between the two scales of self- actualization is substantially stronger than the relation- ship between Synergy and the WOmen's Role Attitude Scale. This may suggest that the content of the items on the Synergy scale might be less relevant to women's roles than is the case for the Contact scale, since the rela- tionship between the WOmen's Role Attitude Scale and the Contact scale was relatively in the same range of strength. H1 5 A more liberal position is positively associated with older age. The variable of age was examined in relation to attitudes toward women's roles. Table 7 shows that younger students were more liberal than older students; 42 percent 73 Table 6 Pearson Product Moment Correlation Coefficients Representing the Interrelationships Among Women's Role Attitudes and Selected Personal Variables Women's Grade Role Point Attitudes Contact Synergy Anomia Average Women's Role Attitudes: Females 1.00 .25** .05 .14 .06 Total 1.00 .21*** .09 .04 .11* Contact: Males 1.00 .14 -.00 .11 Females 1.00 .45*** -.O7 -.05 Total 1.00 .32*** .04 —.14* Synergy: Males 1.00 -.23 .13 Females 1.00 -.13 .10 Total 1.00 -.18* -.ll* Anomia: Males 1.00 -.15* Females 1.00 -.13 Total 1.00 -.14* Grade Point Average: Males 1.00 Females 1.00 Total 1.00 *p = .05. **p = .01. ***p = .001. 74 Table 7 Association Between Age and Attitudes Toward WOmen's Roles Attitudes Age Level Group Traditional Moderate Liberal 2. 2. 2. E. 2. ‘3 Male 18 39 13 28 15 23 Younger Female 8 12 25 39 32 49 Total 26 23 38 34 47 42 Male 27 41 22 34 16 25 Older Female 14 25 20 35 23 40 Total 41 34 42 34 39 32 Males: x2=o.92; p=.315; G=-.09. Females: x2=3.15; p=.103; G=-.22. Total: x2=3.79; p=.075; c=-.21. of the students under 19 years of age were liberal in their attitudes as compared to 32 percent for the students of 19 years and over. Although there was a tendency for these variables to be negatively associated, contrarily to the hypothesis, this association was not statistically significant. H A more liberal position is positively associated 1'6 with residential independence from parents. H1 7 A more liberal position is positively associated ° with residential independence from school authorities. 75 An index of residential independence was used to divide the respondents in three categories: category None corresponded to the students living with their parents (36 percent), category Some contained students living in the college dormitories (34 percent), and category More included the students living in their own apartments (25 percent). Table 8 shows that the percentage of students with liberal attitudes is highest when students are most independent, i.e., when they live in independent housing. Among students living with their parents, 33 percent demonstrated liberal attitudes, as compared to 35 percent of those living in the college dormitories, and 45 percent of those living in inde- pendent housing. Chi-square tests were performed on each possible combination of groups. A significant association was found between residential independence and attitudes toward women's roles in all comparisons. The strength of this association was greatest when category None and category Mg£g_were compared (Gamma==.41). Males and females were also compared on this variable. The asso— ciation was stronger among the males (Gamma==.36) than among the females (Gamma==.l4). Hypotheses 1.6 and 1.7 were definitely supported. 76 Table 8 Association Between Attitudes Toward WOmen's Roles and Residential Independence Attitudes Degree of Independence Group Traditional Moderate Liberal P. i 2 i a 1 Male 25 57 10 23 9 21 None Female 11 28 10 26 18 46 Total 36 43 20 24 27 33 Male 17 40 13 30 13 30 Some Female 8 17 20 43 18 39 Total 25 28 33 37 31 35 Male 3 13 9 21 8 36 More Female 3 8 14 42 18 50 Total 6 10 26 45 26 45 None and Some: x2= 5.24; p= .0035; G= .17. None and More: x2=18.37; p< .0001; G= .41. Some and More: X2 = 6.67; = .017; G= .28. Total: x2=18.95; p= .0004; G= .27. Male: X2=11-73: p= .0095; G= .36. Female: X = 6.73; p=.075; G=.l4. 77 H1.7 A more liberal position is positively associated with employment experience. An index of employment experience was developed to assess the extent to which the subjects had been gainfully employed. It was expected that the more experience they had, the more liberal their attitudes would be. The results were the Opposite: the highest percentage of liberal stu- dents was found among students without any work experience, and this was observed on both subsamples, as Table 9 shows. The number of male students without employment experience was too small to meet the minimal requirement for using a chi-square test to determine the significance level. A similar test among the female students failed to detect any significant association. Hypothesis 1.7 was not supported. H1 8 A more liberal position is negatively associated with sex-bound choice of academic field. The data on choice of academic field were cross- tabulated and tested to determine the existence of a relationship between attitudes toward women's roles and choice of academic field. A higher percentage of subjects with liberal attitudes was found among the students in the traditionally feminine fields, regardless of the sex of subjects (p==.0001; C==.26). In this particular sample, 78 Table 9 and Employment Experience Association Between Attitudes Toward WOmen's Roles Attitudes Degree of Experience Group Traditional Moderate Liberal B i B i B 1 Male 3 37 2 25 3 37 None Female 5 15 ll 33 17 51 Total 8 2O 13 32 20 49 Male 27 34 27 34 25 32 Some Female 13 17 30 40 32 43 Total 40 26 57 37 57 37 Male 15 62 6 25 3 13 More Female 4 27 5 33 6 40 Total 19 49 ll 28 9 23 Females: x2= 1.63; p=.400; c=-.14. Total: x2=ll.48; p= .010; G=-.30. this could mean that, since more women than men were in the feminine fields (81 women vs. 40 men), and women were found in general more liberal than men, this association could be explained by the presence of more women in these fields. In the data analysis performed on separate male and female subsamples, this association held; i.e., students in the feminine fields had more liberal attitudes than students in the masculine fields, but it was found significant in the male subsample only (p==.002; C==.30). These findings, 79 reported in Table 10,suggest that the men who crossed the sex frontier to choose traditionally feminine fields were significantly more liberal in their attitudes toward women's roles than their peers who chose the traditionally masculine fields. The reverse was not true of the women in this sample, although the association between liberal attitudes and feminine field was not found significant (p==.06; C: .18). Table 10 Association Between Attitudes Toward WOmen's Roles and Sex-Bound Choice of Academic Field Attitudes Academic Field Group Traditional Moderate Liberal B 1 El 1". E i . . Male 31 43 28 39 13 18 grad1§T°na11Y Female 10 25 16 40 14 35 asc“ lne Total 41 37 44 39 27 24 . Male 14 36 7 18 18 46 gzggiiignally Female 10 12 30 37 41 50 9 Total 24 2o 37 31 59 49 Male: =10.9; p= .002; C= .30. 2 X Female: X2= 4.09; p=.06; c= .18. x2 =16.70; p= .0001; c= .27. 80 No hypothesis was formulated regarding the subjects' orientation toward marriage. However, the data obtained for descriptive purposes about their orientation toward marriage, allowed exploration of the relationship between this vari- able and attitudes toward women's roles. For this analysis, the responses on the question, "Do you plan to be married?" were used. The responses were grouped in two categories: (1) Yes, for the subjects who indicated that they were already married, would marry within one year or two, or within three to five years; and (2) No, for the subjects who said that they did not plan to marry or that they did not know. Table 11 shows that 47 percent of the students who had no plan for marriage within five years had liberal attitudes as compared to 31 percent of those who were already married or planned to be married within five years. The chi-square test on the total sample revealed a signif- icant association (p==.0016; C==.21). This association remained significant for the female subsample (p==.001; C==.30), but not for the male subsample. These results suggested that postponing or eliminating marriage may have a different meaning for women than for men. The results on personal characteristics are summarized in Table 12. 81 Table 11 Association Between Attitudes Toward Women's Roles and Marriage Plan Attitudes Marriage Plan Sex Traditional Moderate Liberal a i B 3.5. E 1 Male 32 47 18 26 18 26 Yes Female 18 25 29 4O 25 35 Total 50 36 47 34 43 31 Male 13 31 16 38 13 31 No Female 2 4 17 35 30 61 Total 15 17 33 36 43 47 Male: 2= 2.96; p= .11; c--- .21. X Female: x2=12.46; p= .001; c= .30. x Total: 2 = 11.41; p= .0016; c= .21. Summary Table of Statistical Tests Based on Contingency Tables 82 Table 12 for Selected Personal Characteristics of Students variable Group x p G C Sex Total 15.50 .0002 -- .24 Male .92 .315 -.09 -- Age Female 3.15 .100 -.22 -- Total 3.79 .076 -.20 -- Male 11.73 .0095 .36 -- Residential Independence Female 6.73 .075 .14 -- Total 18.95 .0004 .27 Malea -- -- -- -- Employment Experience Female 1.63 .400 -.14 —- Total 11.48 .0105 -.30 -- Male 10.90 .002 -- .30 Academic Field Female 4.09 .060 -- .18 Total 16.70 .0001 -- .27 Male 2.96 .110 -- .16 Marriage Plan Female 12.46 .001 -- .30 Total 11.41 .0015 -- .21 aNumber of cases too small for chi-square requirement. 83 Family Characteristics and Attitudes Toward’WomenTs Roles A general hypothesis was formulated and six subhypotheses were posed with respect to family char- acteristics. They will be discussed as a group, since the statistical procedures were identical and the results are similar. H - A significant association exists between a student's 2 position on the traditionalism-liberalism continuum regarding attitudes toward women's roles and family characteristics: H2 1 A more liberal position is positively associated ' with the social prestige of the father's occupation. H2 2 A more liberal position is positively associated with employment experience of the mother. H2 3 A more liberal position is positively associated ' with father's education. H2 4 A more liberal position is positively associated ' with mother's education. H2 5 A more liberal position is negatively associated ° with size of the family. H2 6 A more liberal position is positively associated with family income. The data for each family characteristic were cross— tabulated and tested for significance level by chi-square procedures. Contrary to the hypotheses, family factors did not appear to be associated with attitudes toward women's roles. The variable of father's occupation prestige, mother's employment experience, mother's education, family 84 size and family income were not found significantly associated with the dependent variable. The only variable that was significantly associated with attitudes toward women's roles was father's education (p==.0035; G==.l7). When the variable of sex was controlled, it appeared that this association held among the female subjects only (Female: p=.0015; G=.23; Male: p=.25; G=.l4). The data for each family variable were examined with the sex variable controlled. These tests, except in the case of the father's education, as already mentioned, did not reveal any significant association. The test on employment experience of the mother showed that daughters of working mothers were more likely to have liberal attitudes than did daughters of nonworking mothers, 50 percent as compared to 34 percent, respectively; the reverse trend was observed among the males: 33 percent of nonworking mothers' sons had liberal attitudes as compared to 14 percent of the working mothers' sons. These results are summarized in Tables 13 and 14. Hypotheses 2.1, 2.2, 2.4, 2.5, and 2.6 were not supported. Hypothesis 2.3 was partially supported. 85 Table 13 Association Between Attitudes Toward Women's Roles and Family Characteristics Attitudes Family Characteristics Traditional Moderate Liberal B i n 35. 2 3; Father's Occupation: Professional-managerial 14 27 16 31 22 42 Technical-clerical 13 26 17 34 20 40 Skilled 19 35 17 32 18 33 Farmer, fisherman, logger, and unskilled 16 29 24 43 16 29 Mother's Occupational Status: Working 37 25 54 36 58 39 Nonworking 30 36 26 31 28 33 Father's Education: Elementary 25 29 40 46 22 25 Secondary 26 27 31 33 38 40 Post-secondary 15 30 9 18 26 52 Mother's Education: Elementary 26 37 22 31 22 31 Secondary 33 28 42 36 20 36 Post-secondary 7 16 16 36 21 48 Family Size: Small 29 34 23 27 33 39 Medium 24 29 33 39 27 32 Large 14 23 24 39 24 39 Family Income: Low 25 31 29 35 28 34 Medium 18 27 23 34 26 39 High 20 27 26 36 27 37 86 Table 14 Summary of Statistical Tests for Family Characteristics by Sex and Total Sample Variables Group x p G C Prestige of father's Male 3.07 .39 .02 -- occupation Female 8.89 .08 .18 -- Total 4.16 .32 .10 -- Mother's employment Male 2.22 .16 -- 14 status Female 3.64 .08 -- 17 Total 3.10 .105 -- 11 Father's education Male 3.34 .25 .14 -- Total 13.93 .0035 .17 -- Mother's education Male 2.33 .34 .13 -- Female 4.65 .16 .24 -- Total 6.50 .08 .21 -- Family size Male 5.10 .135 .15 -- Female 1.67 .39 .00 -- Total 4.53 .17 .06 -- Family incomea Total 0.45 .49 .04 -- aSeparate tests were not performed on the subsamples, given the probability observed on the total sample. 87 Attitudes Toward WOmen's Roles and Institutional Factors A third group of variables was related to institutional factors. The general hypothesis regarding these variables was as follows: H3: A significant association exists between a subject's position on the traditionalism-liberalism continuum regarding attitudes toward women's roles and institutional factors. Three subhypotheses were posed: 3 1 A more liberal position is positively associated ' with the size of the institution attended. H3 2 A more liberal position is positively associated ’ with the urbanization level of the region of origin. 3.3 A more liberal position is positively associated With the size of home town. A series of chi—square tests were performed on each variable for the sample taken as a whole and subsequently with the sex variable controlled. Institutional factors appeared to be better predictors of attitudes toward women's roles than were family factors. Table 15 shows that 25 percent of the students in the smaller institutions had liberal attitudes toward women's roles as compared to 49 percent of the subjects in the larger institution. The chi-square test revealed a significant association between these variables (p<:.0001; G==.48). The comparison of attitudes between males and females in relation to size of institution shows that the association between attitudes 88 Table 15 Association Between Attitudes Toward Women's Roles and Size of Institution Attendeda Attitudes Size of Institution Group Traditional Moderate Liberal 2. i 2 1 P. 1 Male 31 54 18 31 8 14 Smaller Female 14 25 22 39 20 36 Total 45 4o 40 35 28 25 Male 14 26 17 31 23 43 Larger Female 8 11 24 34 38 54 Total 22 18 41 33 61 49 Male: X2 = 13.63; p= .0005; G= .53. Female: x2 = 9.71; p= .005; G= .35 Total: X2 = 26.06; p= .0001; G= .48 aSince the females' institutions, the data were of cases. attitudes were similar in the smaller combined because of the limited number 89 toward women's roles and the size of the institution was stronger in the case of the male students than it was for the female students (Male: p==.0005; G==.53; Female: p= .005; G= .35). A significant association was found between atti- tudes toward women's roles and the level of urbanization of the subject's region of origin when the data for the total sample were analyzed (p==.003; G==.39). This asso- ciation weakened somewhat when the data were analyzed with the variable of sex controlled (Male: G==.31; Female: p==.10; G==.37). The number of male subjects in the more urbanized category was too small to meet the minimal requirements for the chi-square test. Table 16 shows that 50 percent of the male subjects issued from the more urbanized area had liberal attitudes as compared to 24 percent for those coming from a less urbanized area. The equivalent figures for the female subjects were 60 percent versus 38 percent. The third institutional variable studied was the size of home town. Table 17 shows that 50 percent of the subjects coming from larger towns had liberal attitudes as compared to 36 percent of those coming from smaller towns. A weak but still significant association was found between this variable and attitudes toward women's roles. The association was even weaker when the sex variable was controlled. 90 Table 16 Association Between Attitudes Toward Women's Roles and Urbanization Level of the Subject's Region of Origin Attitudes Urbanization Level Group Traditional Moderate Liberal 2 3; fl 1 E i L 88 Male 39 42 32 34 22 24 ufibaniz a Female 19 21 37 41 35 38 8 Total 58 32 69 37 57 31 M Male 6 33 3 l7 9 50 °Ee . ed Female 3 10 9 3o 18 60 “r anlz Total 9 19 12 25 27 56 Female: x2= 4.53; p=.05; G=.37. Total: x2=11.09; p=.0015; G=.39. 91 Table 17 Association Between Attitudes Toward Women's Roles and Size of Home Town Attitudes Size of Home Town Group Traditional Moderate Liberal B i B i P. 1 Male 21 38 22 39 13 23 Smaller Female 8 14 28 48 23 39 Total 29 25 50 44 36 31 Male 23 43 12 23 18 34 Larger Female 14 22 18 28 32 50 Total 37 31 30 36 50 43 Male: x2 = 3.75; p= .075; G= .03. Female: x2=5.08; p= .035; G= .06. Total: x2=8.23; p=.008; G=.O7. 92 As both size of campus and sex were two variables found strongly associated with the dependent variable, further exploration of this association was pursued by comparing men and women when controlling the variable of institution. Although women were found more liberal than men in the larger institution, the association between the variable of sex and women's role attitudes was stronger in the smaller institution (p==.01; G==.33) than it was in the larger institution (p==.05; G==.30) (see Table 18). These comparisons were possible between two institutions only, since the subjects on the third (smallest) institution were all female. The results of the analyses for institutional factors are shown in Table 19. Table 18 Association Between Attitudes Toward Women's Roles and Sex by Institution Attended Attitudes Sex Institution Traditional Moderate Liberal 2 i E 1 E i M 1es Larger 14 26 17 31 23 43 a Smaller 31 54 18 32 8 14 F 1 Larger 8 12 24 34 38 54 ema es Smaller 9 28 12 38 11 34 Larger: x2=4.53; p= .05; C= .18. Smaller: x2=7.32; p=.01; C= .27. 93 Table 19 Summary Table of Statistical Tests for Institutional Factors 2 Variable Group x p G Male 13.63 .0005 .53 Size of institution Female 9.71 .001 .34 Total 26.06 <.0001 .39 Male 5.52 .03 .31 Urbanization level Female 4.53 .05 .37 Total 11.109 .003 .39 Male 3.75 .08 .03 Size of home town Female 5.08 .03 .06 Total 9.42 .008 .07 The last statistical procedure used in this study was a forward stepwise regression analysis. This technique was used to find the best equation to predict attitudes toward women's roles, and to determine the contribution of each of the variables in this equation. The factors were entered into regression according to their degree of correlation with the dependent variable. An F-ratio of 2.5 was set in order to delete the variables that did not add substantially to prediction accuracy, once certain other independent variables were included. Tables 20 and 21 summarize the results of the regression techniques for the male and female subsamples separately. 94 Table 20 Multiple Regression Analysis Predicting Attitudes Toward WOmen's Roles from Independent Variables for Male Subsample Stepwise Regression Independent Correlation Beta F-Statistic Additive Variables Coefficient (Elasticity) (Probability) R2 Campus size .35 .18 13.138 .121 (.10) (.000) Self-actualization .22 .26 6.279 .055 (Contact scale) (.22) (.014) Field of study .23 .32 7.0091 .058 (.04) (.009) Age -.10 -.16 2.906 .023 Size of home town .07 .29 2.596 .020 Residential .16 .38 3.958 .030 independence (.14) (.104) Urbanization of .23 .22 2.697 .020 region of origin (.01) (.104) Marriage plan -.10 -.16 3.380 .024 (.02) (.069) Multiple R = .595 R2 = .35 F-Statistic = 6.249 p= .0002 Degrees of Freedom: 8 and 88 95 Table 21 Multiple Regression Analysis Predicting Attitudes Toward Women's Roles from Independent Variables for Female Subsample Stepwise Regression Independent Correlation Beta F-Statistic Additive Variables Coefficient (Elasticity) (Probability) R2 Marriage plan -.280 -.249 8.714 .078 (-.03) (.004) Campus size .254 .22 8.064 .058 (.07) (.010) Self-actualization .25 .22 5.781 .047 (Contact scale) (.12) (.018) Anomia .144 .16 3.323 .026 (.10) (.071) Father's .17 .14 2.559 .020 education (.05) (.113) Multiple R= .481 R2 = .231 F-Statistic = 5.900 p < .0001 Degrees of Freedom: 5 and 98 96 The best equation prediction of attitudes toward women's roles for the male sample included a set of nine variables. Four of these variables made a contribution to explained variance that is noteworthy. The size of the institution attended accounted for 12.1 percent of variation, followed by the choice of academic field which accounted for 5.8 percent of variation; self-actualization contributed another 5.5 percent and residential independence 3 percent. Marriage plan, age, size of home town, and urbanization of the region of origin met the criteria of inclusion into regression equation but they introduced unimportant changes in the Multiple R2. For the female subsample, the same criterion of inclusion into regression (F-ratio of 2.5) was used, and five variables met this requirement. The last variable that entered the regression equation for the male subsample was the first to enter for the female subsample. The best prediction equation would include the following variables: (1) marriage plan, which explained 7.8 percent of variation; (2) size of institution attended, which accounted for 5.8 percent; and (3) self-actualization, which accounted for 4.7 percent. The results of the regression analysis confirmed a previous observation that the pattern of association between the dependent variable and the independent variables was different between the male and the female subsamples. follows: 1. 97 Summary of Results The results of data analysis are summarized as Respondents' positions on the traditionalism- 1iberalism continuum. a. A larger percentage of respondents was found in the liberal category of attitudes: 37 percent as compared to 35 percent in the moderate category and 29 percent in the traditional category. The liberal category was the largest for the female subsample: 45 percent of the women students were in the liberal category, while 37 percent were in the moderate category and 18 percent in the traditional category. The traditional category was the largest for the male subsample: 40 percent of the men were in this category while 32 percent were in the moderate category and 28 percent in the liberal category. Personal characteristics and attitudes toward women's roles. a. The sex of the respondent was significantly associated with attitudes toward women's roles; female students held more liberal attitudes than male students. Younger students tended to have more liberal attitudes than older students. Students with higher academic performance tended to have more liberal attitudes than those with lower academic performance. Women with higher scores on anomia had signifi- cantly more liberal attitudes than women with lower scores. Men with lower scores on anomia tended to have more liberal attitudes. 98 The students who were residentially independent were significantly more liberal than those who lived with their parents or in the university residences. Employment experience was not associated with attitudes toward women's roles. The male students who chose a traditionally feminine field of study were significantly more liberal than those who chose masculine fields of study. The women who did not plan to marry within five years held more liberal attitudes than the women who were already married or who planned to marry within five years. Family characteristics and attitudes toward women's roles. a. The daughters of fathers with a high level of education held significantly more liberal atti- tudes than the daughters of fathers with a low level of education. There was no association between the sons' attitudes and the father's education. The other family variables-~social prestige of the father's occupation, mother's employment experience, mother's level of education, family size, family income--were not associated with attitudes toward women's roles. Institutional factors and attitudes toward women's roles. a. The students attending a larger institution had more liberal attitudes than those who attended a smaller institution. This association was stronger in the case of the male subjects. The students coming from a more urbanized area had more liberal attitudes than those coming from a less urbanized area. The students coming from a larger town tended to have more liberal attitudes than those coming from a smaller town. 5. 99 Best predictors of attitudes toward women's roles. a. Size of institution, self—actualization, field of study, and residential independence were the best predictors of the male subjects' attitudes toward women's roles. Marriage plan, size of institution, and self- actualization were the best predictors of the female subjects' attitudes toward women's roles. CHAPTER VI DISCUSSION OF FINDINGS AND LIMITATIONS OF THE STUDY Discussion of Findings The results of data analysis are discussed around the three major foci of investigation: (1) the association between personal characteristics of subjects and attitudes toward women's roles; (2) the association between the family characteristics of the subjects and attitudes toward women's roles; and (3) the association between institutional factors and attitudes toward women's roles. Personal Factors A measure of the student's attitudes toward women's roles was developed as a position on a continuum of traditionalism-liberalism. Selected personal factors were studied in relation to these attitudes. The results of these analyses suggested that the variable of sex was definitely associated with the dependent variable; i.e., attitudes toward women's roles. The female subjects had higher scores on the women's role attitude scale. Most studies have reported sex-related differences 100 101 in sex-role attitudes (Hartley, 1968; Haavio-Mannila, 1967; Hobart, 1972; Osmond and Martin, 1975). The smaller dif- ferences observed between men and women attending a larger institution than those found between men and women in a smaller institution supported Hobart's remark concerning men-women differences between English and French Canadians, as being related to differences in phase of attitude changes with respect to women's place. Similarly, it is more likely that the evolution of attitudes regarding women's roles is more advanced in a larger institutional setting, located in a more urbanized area, than it is in a smaller institution located in a less urbanized area. Therefore these findings corroborate those of earlier studies from yet another perspective. Self-actualization was positively associated with attitudes toward women's roles, as predicted. However, the results of a positive association between anomia and atti- tudes toward women's roles, while a weak but negative asso- ciation was observed between anomia and self-actualization, are difficult to explain. Conceptually, anomia should be related to attitudes in the same direction as it is to self-actualization. The data in this study did not permit clarification of these interrelationships. One possible explanation is that the most liberal women might view themselves as radical and perceive their attitudes as 102 socially deviant, explaining their high scores on anomia. At the same time their liberal attitudes are probably congruent with their psychological self, explaining their high scores on self-actualization. Discovering a negative association between age and attitudes toward women's roles was unexpected. Although the association was weak and below the .05 level of significance, the direction of the association was opposite to this study hypothesis. Bernard (1974) in reviewing a number of studies relating age to feminism, pointed out that the variable age is neither simple, nor unequivocal. Hypothesis 1.5 seemed to counter some research findings where liberalism was associated with younger age. The direction of this hypothesis was based on a developmental point of view; as adolescents mature, they are likely to become more autonomous and to depart from conformity to stereotyped views of sex roles. It was assumed that most students would still be in that phase of development, since age 30 seemed to be a turning point in the evolution of attitudes, as the studies reviewed by Bernard suggested. A contributing factor in the present study could be that the age range of college populations is limited. Because of the age distribution of this sample, the two age cate- gories were "under 19 years of age" and "over 19 years of age." Further investigation using a broader range of age, 103 including graduate students for instance, or using a sample stratified on the basis of age might result in different findings. Although a weak association was found between academic performance and attitudes toward women's roles, it nonetheless supported Bayer's (1975) findings that "sexist" students were more likely to have low academic performance. No other studies were found that included this variable. The results of a negative association between the existence of a plan to be married within five years and liberal attitudes seemed to be in agreement with Tangri's (1972) findings; role innovation among women students was associated with postponing marriage. The present study does not provide evidence that these women were eliminating marriage as a life style, nor that if they were, it was for feminist reasons; it only suggested that early commitment to marriage seemed to reveal more traditional attitudes, at least among women. It was suspected that the findings related to residential independence and to choice of academic field might have a common explanation, in relation to the concept of role innovation studied by Tangri (1972). In the present study, a significant association was found among the male subjects but not among the female subjects for both 104 variables. As male freshmen who crossed the sex frontier and chose feminine fields of study could be viewed as role innovators, freshmen who decided to have their own apart- ments and take on housekeeping responsibilities might be more innovative than the average male students. Only six students were married in the total sample, and it seemed that cohabitation was not a common pattern among freshmen on the largest campus studied. The Pearson's correlation coefficient between residential independence and choice of academic field was a negative one (-.14). Therefore this explanation does not hold for both variables. It can also be observed that male and female students on the larger campus are more likely to be residentially independent than those on the smaller campus (r==.45). The association observed between residential independence and attitudes toward women's roles might be a composite effect of being on the larger campus. The size of the sample did not allow pursuit of this investigation by controlling the variables of sex and institution at the same time. The results of a negative association between sex- bound choice of academic field among the male students suggested that the men who chose the tradiditonally feminine fields were more liberal than their peers who chose the traditionally masculine fields. These findings are similar to those reported by Tangri in her study on role innovation 105 among women. It would appear logical to find the same tendency among the female students in this study. One factor that may explain why the female students in the traditionally masculine fields were not significantly more liberal than those in the traditionally "feminine" fields is that in the smaller colleges, the only other options available besides Humanities are Administration and Science, which were defined as traditionally masculine fields.’ The women in the smaller colleges may not be role innovators by choice but merely by unavailability of other fields. A larger sample of women with equal opportunities for choice of academic fields would be necessary to explore this relationship further. A slight tendency to vary in opposite directions was observed for employment experience of the subjects and liberalism of attitudes. The rationale for the hypothesis that the association could be positive was that as young people enter the world of employment, they might become more aware of interesting options for women besides the traditional roles of wives and mothers. The truth might be that actual experience in the employment environment made them.more realistic about the problems that still exist for working women and contributed to modify their possibly idealistic views of women's roles. 106 Among the personal factors studied, the variable of sex was found strongly associated with the independent variable. For this reason it was felt necessary to control this variable for subsequent analyses. Two other personal factors were found to be good predictors of attitudes among male students: self—actualization (contact) and choice of academic field. Orientation toward marriage was included in the prediction equation but made a small contribution. For the female students, the best predictor among the personal factors was orientation toward marriage followed by self-actualization (contact). Family Characteristics and Attitudes Toward WomenTs Roles The family characteristics were not found to be particularly good predictors of attitudes toward women's roles among these college students. These general findings are in agreement with Hobart's observations among English- and French-Canadian students (1972). One general explanation could be that the French- Canadian population in New Brunswick is a rather homogeneous group with regard to socioeconomic and demographic attri- butes. The parents who are in the more prestigious occupa- tional categories are for the majority the first generation to achieve this level and their life styles are not basi- cally influenced by their social position. The variable of 107 family size, which is usually associated with employment status of the mother and attitudes in daughters, was not associated with attitudes. Having a large family in this sample, as among the French-Canadians of New Brunswick, was the norm rather than the exception, for people of this generation. Hence, it is not surprising that family size was not a good predictor for this population. The results of no association between mother's employment status and the subjects' attitudes was the most unexpected, since earlier studies have suggested the exis- tence of such an association, although there is no unanimity in this respect (Hoffman and Nye, 1974; Vogel, 1970). In the present study, mother's employment might be mainly a substitute or a supplement to the father's income, rather than a self-fulfilling activity for the mother, considering the high incidence of male unemployment in New Brunswick and the seasonal character of occupations such as fishery and agriculture. Although these findings do not support the broad hypothesis that women's roles attitudes are associated with family characteristics, it is not implied that the family does not influence the formation of sex-role attitudes. The parents' attitudes and behavior with regard to child rearing philosophy and practices, sex education, division of responsibility and authority in the family were not 108 investigated and these might be the most influential family factors. Institutional Factors and Attitudes Toward Women's Roles Two of the three institutional factors investi- gated were significantly associated with attitudes toward women's roles. The size of the institution appeared to be a powerful predictor of attitudes for both male and female respondents and even more so for the males. The regression analysis performed on the data for the male subsample indicated that the size of the institution was the best predictor of their attitudes toward women's roles. This factor, however, is a complex one, as was pointed out by Feldman and Newcomb (1972). The question is raised as to whether the students on the larger campus are more liberal because of the diversity of influences present on a larger campus, or are they on the larger campus because they are basically more liberal in their attitudinal make-up? The particularly traditional attitudes among the males in the medium-size institution is intriguing. It is difficult to impute these attitudes to the factor of institution solely; the sex variable accounted for their traditional attitudes since the female subjects from this institution were significantly more liberal. It was observed that fewer students from that area of the 109 Province are attracted by the larger institution. It is suspected that a good number of students from that area are attending institutions outside the Province, possibly in Quebec, since Laval University is located at an equiva- lent distance for them as is the Université de Moncton. Because of the attraction of a much larger university, as nearly located and with a wide range of academic fields, the more liberal students might be those who go outside the Province, explaining the more traditional attitudes found among the students, in the present study. These findings suggested that the size of the institution does have an impact on the students' attitudes. The students in a larger institution are exposed to more pluralistic norms and to a multiplicity of experiences that are not available on a small campus. This impact is pos- sibly compounded with the effects of the personality characteristics of the students who are attracted to a larger setting because they are seeking these kinds of experiences. Another institutional factor associated with liberal attitudes was the urban character of their region of origin. This relationship is difficult to separate from that between institution and attitudes, since in the medium- size college, 83 of the 89 subjects came from within the institution area. It was not possible either to compare 110 the subjects on that variable within the larger campus, since only 10 students from the northwest area came to the Moncton campus. This variable, however, entered into the regression equation for the male sample, and contributed 2.7 percent of the explained variance. The results of a weak association between size of home town and attitudes toward women roles indicated a tendency that was expected. This variable entered the regression equation and explained 2.6 percent of the vari- ance. Most communities were of small size in this sample, and about half of the subjects came from communities smaller than 2,500 p0pulation. The homogeneity of these communities may account for the finding that this variable was not a powerful predictor of attitudes toward women's roles. The impact of size of home town must be explored in a different setting to determine whether it is associated with attitudes toward women's roles. Limitations of the Study Instrument. The instrument used for the measurement ‘of the dependent variable was satisfactory for the objective of this study. An amplification of this instrument, by addition of items referring to more specific situations and behaviors would allow greater in—depth study of attitudes toward women's roles. 111 Choice of independent variables. The measurement of personal variables was more extensive on demographic data and life experiences than on personality measures. The results of this study suggested that a stronger emphasis on personality measures would possibly account for some of the unexplained variance. Similarly, the family variables included social background characteristics, but no information on parents' attitudes was obtained. An effort should be made to include measurement of parental attitudes in future research. The control over some independent variables was limited by the distribution found in this sample. In some instances, such as age and employment experience, the homogeneity of the data did not permit elucidation of the relationships. Further investigations using samples with wider distributions on these variables would provide Opportunity for a more thorough examination of the relationships. Sex bias. The data collection period occurred during the Women's International Year. It is likely that women in general and women in this sample were more influenced by consciousness raising about women's status than men were and this might be reflected in the results of this study. It is also possible that since the topic was expressed as women's roles, men could not identify as 112 well as women with the content of the attitudinal items. A study encompassing both sex roles could avoid possible prejudice in favor of women. It would also give proper consideration to the likelihood that changes in women's roles bring changes in men's roles. CHAPTER VII SUMMARY AND IMPLICATIONS Summary This study was designed to explore sex—role attitudes among French-speaking college students of New Brunswick, Canada. Little was available in the literature on women's roles in this population. An attempt was made to estimate the relative degree of liberalization of attitudes toward women's roles and to explore the relationship between these attitudes and selected personal, family, and institutional variables. The sample consisted of 234 male and female college freshmen enrolled in a degree program in three French- speaking institutions. An attitudinal scale, developed for the present study was used to determine the participants' positions on a continuum of traditionalism-1iberalism with regard to their views of appropriate roles for women. A wide range of scores on this scale was obtained which indicated a variety of views concerning women's roles. The major variable, attitudes toward women's roles, was studied from the point of view of sex—role socialization. 113 114 Personal, family, and institutional factors were hypothesized to be associated with attitudes toward women's roles. The hypothesis testing was performed by chi-square tests, correlation, and multiple regression procedures. Substantial differences of attitudes between men and women were observed in this sample; women student's attitudes were significantly more liberal than men's. This finding was interpreted as indicative of a beginning stage of liberalization of attitudes, since a large proportion of men did not seem to have been influenced yet by the feminist movement. Different patterns of association were found between males and females. Younger students, in both subsamples, tended to hold more liberal attitudes than older students. Male and female respondents who scored high on a measure of self-actualization had more liberal attitudes than respon- dents who had low scores on this variable. Students who achieved higher academic performance were found to have more liberal attitudes than those with lower achievements. WOmen students with high scores on the scale of anomia were found to hold more liberal attitudes than women students with lower scores, while the reverse trend was observed among male students. The respondents who were residentially independent held more liberal attitudes than 115 those living with their parents or in the college dormitories; the association observed between these two variables was stronger among the men than among the women. Men students enrolled in the traditionally feminine fields of study were more liberal than those enrolled in the traditionally masculine fields. Women students who did not plan to be married within five years held sig- nificantly more liberal attitudes than women students who were already married or planned to be married within five years. The students attending a larger institution had more liberal attitudes than those attending a smaller institution; the association between size of institution and attitudes toward women's roles was stronger among the men than among the women. There was a tendency for the students coming from a more urbanized area and from a larger town to hold more liberal attitudes than those coming from a less urbanized area and from a smaller town. The social characteristics of the respondents' families were not found significant predictors of attitudes toward women's roles; only the father's level of education was significantly associated with the daughters' attitudes toward women's roles. The results of the regression analysis indicated that the best predictors of the men students' attitudes 116 toward women's roles were the size of institution attended, the choice of academic field, and residential independence. - The best predictors of the women students' attitudes were marriage plan, size of institution attended, and self- actualization scores. The results of this study suggested that a socialization model was appropriate for the study of sex- role attitudes and that a more complete operationalization of this model, emphasizing parental attitudes, might add a substantial contribution in explaining differences of attitudes toward women's roles. Our contribution of this study was to provide data about sex-role attitudes in the French-speaking population of New Brunswick. The findings corroborate those of earlier studies and indicate that on the whole, the sample studied responded with patterns similar to those observed in other populations. Implications for Future Research The results of this study illustrate that male and female college students in the French-speaking population of New Brunswick do hold different attitudes with regard to women's roles. More research is needed among the adult p0pulation and outside of the intellectual elite to deter- mine the extent to which changes are taking place in 117 attitudes and behaviors related to sex roles. A study of the effects of women's participation in feminist groups on their families by comparing them to families in which women do not participate in such groups, is needed to help the feminist movement evaluate its action. Regional differences observed should be further explored and extended to other groups: working and non- working women, husband and wives, school children of each region. Since New Brunswick is a bicultural area, and inter-ethnic marriages are frequent occurrences, a com- parative study of sex-role attitudes is suggested between members of both cultural groups. There is no objective basis to affirm the relative liberalism of this sample in regard to other groups of French-Canadians. Comparisons with other French-speaking groups in Canada should include subsamples where the French are also a minority group, as well as subsamples where the French represent the predom- inant cultural group, to explore cultural influences on attitude formation. The family factors should be further studied with an emphasis on parental attitudes, child-rearing practices, and actual division of family responsibilities between the sexes within the family. 118 Differences of attitudes observed between men and women have implications beyond research. These men and women are likely to be partners in professional circles as well as in their private lives. Their conflicting role expectations are potential sources of tension in these encounters. Differences in attitudes and expectations may increase if consciousness raising groups continue their action with only women as their target audience. Effort should be made to provide interactive contexts where men and women can share their personal expectations and their views of social expectations with respect to sex roles. As a result of these interactions, a reinterpretation of women's liberation ideology might bring a de—emphasis of sex stereotypes in all areas of life and primarily in the socialization of the next generation. A first step has been taken in providing data about sex—role attitudes in French New Brunswick. Although no claim of representativeness is made, the results of this study have suggestive value in indicating which variables might be predictors of attitudes toward women's roles. It is hoped that the information gained in this study will be helpful in planning future research, adult and youth education programs oriented toward a better understanding of sex-role relationships, and promotion of attitudes which treat both men and women as partners in society. APPENDICES APPENDIX A QUESTIONNAIRE, LETTER OF INTRODUCTION, AND LETTER OF RECALL SECTION I WOMEN'S ROLE SCALE On the pages that follow, you will find statements about women's roles. There are no right or wrong answers. Read carefully each statement and give your first reaction to the opinion expressed by encircling the number that best represents your opinion. Circle only one answer for each statement. KEY: l Strongly agree 2 Mildly agree 3 Mildly disagree 4 Strongly disagree 1. An important aspect of a woman's education is 1 2 3 4 preparation for her professional life. *2. The presence of the mother in the home is essential l 2 3 4 to the normal development of the child. 3. A woman's education is a means to help her husband 1 2 3 4 to get ahead in his profession. 4. In a family the girls should be the mother's respon— l 2 3 4 sibility and the boys should be the father's. 5. It is unjust that a woman should give up her 1 2 3 4 personal ambitions after her marriage. *6. There is nothing wrong if a married woman has 1 2 3 4 her vacation separately from her husband. *7. After obtaining her degree, a woman should find it 1 2 3 4 more important to have a job than to be married. *8. Professions in general are suitable to women as l 2 3 4 well as to men. *9. The greatest contribution a woman can bring to 1 2 3 4 society is to make a happy home for her family. *The items marked with an asterisk were used for the computation of the respondent's score on the women's role attitude scale. 119 *10. 11. *12. *13. *14. *15. *16. 17. 18. 19. *200 21. *22. *23. *24. 120 There is nothing wrong for a married woman to participate in recreational activities without her husband. The impression a woman gives depends in a great measure upon her appearance. In a couple, if the woman earns more money than the man, this should not create a problem to the m. It is more important for a man than for a woman to pursue education. A woman should be free to work outside of the home to be more independent. It would be desirable for a couple that the husband earn more than his wife. Even though women can play important roles in the world of work, their first place is in the home. It is important for a woman student to discuss theoretical and impersonal subjects outside the classroom. A woman should not let a male friend pay all her expenses when they go out together. It is acceptable that a woman delay her marriage to complete her education more easily. A woman should choose a career according to her interests and abilities without restraint regarding family responsibilities. After obtaining her degree, a woman should find it more important to have a job than to pursue her education. It is important for a woman to develop skills in homemaking and all that is related to the home. It is unfair that a woman lose her name when she she marries. It is important for a woman to be trained for a career, but it would be preferable that she does not exercise it after marriage. 121 25. In the case of divorce or separation, the mother 1 2 3 4 should be given the responsibility for the care of the children. *26. It is justifiable for a woman to separate 1 2 3 4 temporarily from her husband, if necessary, to pursue her education. *27. A marriage without children is incomplete. l 2 3 4 28. After obtaining her degree, it should be more 1 2 3 4 important for a woman to pursue her education than to be married. *29. In the case of divorce or separation, it would 1 2 3 4 be right that a court ordered a woman to pay alimony to her ex-husband, if he is disabled. 30. A woman should not have children if she does 1 2 3 4 not have an inclination for the maternal role. For items 31 to 35, give your responses in the same fashion you did for the preceding pages. 31. There is little use writing to public officials 1 2 3 4 because often they are not really interested in the problems of the average man. 32. Nowadays a person has to live pretty much for l 2 3 4 today and let tomorrow take care of itself. 33. In spite of what some people say, the lot of l 2 3 4 the average man is getting worse, not better. 34. It is hardly fair to bring children into the l 2 3 4 world with the way things look for the future. 35. These days a person does not really know whom l 2 3 4 he can count on. 122 SECTION II In the following section, give your answer by checking one item for each question. 36. Where do your parents live? 1. On a farm 2. In the country, but not on a farm 3. In a village with less than 2,500 people 4. In a town of more than 2,500 but less than 10,000 people 5. In a city with more than 10,000 people 37. How many brothers and sisters do you have? 1. l, 2, or 3 2. 4, 5, or 6 3. 7, 8, or 9 4. 10 and over 38. Is your family intaét? 1. No, my father is deceased 2. No, my mother is deceased 3. No, both parents are deceased 4. No, my parents are separated (or divorced) 5. Yes, my parents live together 39. In which category does your father's occupation belong? 1. Professional (physician, lawyer, professor, etc.) 2. Proprietor, manager of large enterprise 3. Proprietor, manager of small enterprise 4. Technician, clerical, sales personnel 5. Skilled worker (electrician, plumber) 6. Fisherman, logger, farmer 7. Unskilled worker 8. Other (specify) 40. How many years did your father attend school? 1. 7 years or less 2. Between 8 and 12 years 3. 13, 14, or 15 years 4. 16 years 5. More than 16 years 123 41. How many years did your mother attend school? 1. 7 years or less 2. Between 8 and 12 years 3. 13, 14, or 15 years 4. 16 years 5. More than 16 years 42. What was the total income of your family last year (before income tax deductions)? 1. Under $5,000 2. $5,000 to $7,999 3. $8,000 to $11,999 4. $12,000 to $15,999 5. $16,000 and over 43. Is your mother employed? 1. Not now, but some time in the past 2. No, she was never employed 3. Yes 44. Where do you live during the school year? 1. With my parents 2. In the student dormitory 3. In my own apartment 4. Other (specify) 45. Did you work during the summer vacation last year? 1. Yes 2. No 46. Do you have a part time job during the school year? 1. Yes 2. No 47. What is your age? 1. 17 years 2. 18 years 3. 19 years 4. 20 years 5. 21 years 6. 22 years 7. 23 years and over 48. Your sex? 1. Male 2. Female 124 49, In which Department or College are you registered? 1. 2. 3. 4. 5. 6. 7. Arts Administration Nursing Sciences Education Home Economics Social and Behavioral Sciences 50. What was your Grade Point Average for the last term? 1. 2. 3. 4. 5. 6. 7. Less than 1 1 to 1.4 1.5 to 1.9 2 to 2.4 2.5 to 2.9 3 to 3.4 3.5 and over 51. Is it your intention to be married? 1. 2. 3. 4. 5. I am already married Yes, in one or two years Yes, in three to five years No I don't know 52. Name of your school: 53. County of your home town: 125 SECTION I Instructions: S'il vous plait, lisez attentivement chacune des Code : 10. ll. affirmations suivantes et donnez votre opinion en encerclant le nombre qui correspond le mieux a votre opinion. 1 Tout a fait d'accord 2 Partiellement d'accord 3 Partiellement en désaccord 4 Tout a fait en désaccord Un aspect important de l'éducation d'une femme, l c'est la preparation 5 la vie professionnelle. La présence de la mere au foyer est essentielle 1 an dévelOppement normal de l'enfant. L'éducation est un moyen pour une femme d'aider 1 son mari a monter dans l'échelle sociale. Dans une famille, il est naturel que la mére l s'occupe davantage des filles et que le pére s'occupe davantage des garcons. Il est injuste qu'une femme en se mariant doive l abandonner ses ambitions personnelles. I1 n'y a rien de repréhensible a ce qu'une femme l prenne ses vacances séparement de son mari. Apres l'obtention de son baccalauréat, il devrait 1 gtre plus important pour une femme de trouver un emploi que de se marier. Les professions en général conviennent aussi bien l aux femmes qu'aux hommes. La plus grande contribution qu'une femme puisse l faire a la societe, c'est de fonder un foyer heureux. 11 n' y a rien de repréhensible 3 ce qu'une femme 1 ait des sorties récréatives séparement de son mari. L'impression qu'une femme donne dépend pour une l bonne part de son apparence. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 126 Dans un couple, si la femme recoit un salaire plus élevé que celui de son mari, cela ne devrait pas creer un probléme personnel au mari. Il est plus important pour un homme que pour une femme de faire des études. Une femme devrait avoir la liberté de travailler a l'extérieur du foyer aprés son mariage afin de demeurer plus indépendante. Il est souhaitable que dans un couple, 1e salaire du mari soit plus élevé que celui de la femme. Méme si les femmes peuvent jouer un r61e important dans le monde du travail, leur premiére place est au foyer. Il est important pour une étudiante de parler de sujets théoriques et impersonnels en dehors des heures de cours. Une femme ne devrait pas laisser son ami payer toutes les dépenses lorsqu'ils font des sorties ensemble. Il est acceptable qu'une femme retarde son mariage pour terminer ses études plus facilement. Une femme devrait choisir une carriére selon ses gofits et ses aptitudes sans égard aux respon- . ’ O . sabilltes familiales. Aprés l'obtention de son baccalauréat, i1 devrait gtre plus important pour une femme de trouver un emploi que de continuer ses études. Il est important pour une femme de développer des habiletés dans les arts ménagers et dans tout ce qui concerne la maison. Il est injuste qu'une femme perde son nom en se mariant. Il est important pour une femme d'avoir une profession, mais il est préférable qu'elle ne l'exerce pas aprés son mariage. S'il survient un divorce ou une séparation, il est tout naturel que la mere obtienne la garde des enfants. 26. 27. 28. 29. 30. Pour fait 31. 32. 33. 34. 35. P our 127 11 est justifiable qu'une femme s'éloigne de son 1 2 3 mari de facon prolongée pour aller poursuivre des études. Un mariage sans enfant est incomplet. l 2 3 Apres l'obtention de son baccalauréat, i1 devrait l 2 3 etre plus important pour une femme de continuer ses études que de se marier. Dans un cas de divorce, i1 serait juste qu'une cour l 2 3 K O I I ordonne a une femme de payer une pension alimentaire \ O O O a son mari, sl ce dernler est invalide. Une femme ne devrait pas avoir d'enfant si elle ne se 1 2 3 sent pas d'attrait pour le r81e maternel. ; s z A . les numeros 31 a 35, repondez de la meme maniére que vous l'avez pour les pages précédentes. Cela ne sert de rien d'écrire a une administration 1 2 3 O O I ’ parce que souvent les fonctionnaires ne s'interessent . K pas vraiment aux problemes de l'homme moyen. On doit vivre dans l'instant et laisser demain prendre l 2 3 soin de lui—meme. En dépit de ce qu'on raconte, la condition de l'homme 1 2 3 moyen ne s'ameliore pas, mais elle empire. ‘ I Cela ne vaut guere la peine de mettre des enfants au 1 2 3 z . ; monde, etant donne ce que l'avenlr nous reserve. N 00 On ne sait jamais aujourd'hui sur qui on peut compter. 1 SECTION II cette section du questionnaire, indiquez votre résponse par un crochet dans l'espace approprié. 36. Oh demeurez-vous durant l'année scolaire? 1. Avec mes parents 2. A la résidence étudiante 3. En appartement 4. Autre (specifiez) 128 37. OH vos parents demeurent—ils? 1. Sur une ferme 2. A la campagne, mais pas sur une ferme 3. Dans un village de moins de 2,500 habitants 4. Dans une ville de plus de 2,500, mais moins de 10,000 habitants 5. Dans une ville de plus de 10,000 habitants \ 38. Combien de freres et/ou de soeurs avez-vous? l. l, 2, on 3 2. 4, 5, on 6 3. 7, 8, on 9 4. 10 et plus 39. Est—ce que votre famille est intact (non brisée)? 1. Non, mon pEre est décédé 2. Non, ma mére est décédée 3. Non, mes deux parents sont décédés 4. Non, mes parents sont séparés 40. Dans quelle catégorie d'occupation le travail de votre pére se situe-t—il? l. Professionel (médecin, avocat, professeur, etc.) 2. PrOpriétaire, directeur d'une grosse entreprise 3. Propriétor, directeur d'une petite entreprise Commis de bureau, technicien, vendeur Ouvrier spécialisé Agriculteur, pécheur, bficheron Ouvrier non qalifié Autre (specifiez) CD\IO\U‘|A O 41. Combien d'annéesEll'école votre pore a—t-il complétées? 1. 7 années ou moins 2. Entre 8 et 12 annees 3. 13, 14, on 15 années 4. l6 années 5. Plus de 16 années 42. Combien d'années a l'ecole votre mEre a—t-elle complétées? . 7 années ou moins Entre 8 et 12 années l3, 14, cu 15 années l6 années Plus de 16 années U'lthNH O 43. 44. 45. 46. 47. 48. ‘49. 129 Quel a été 1e revenu total de votre famille l'année derniére (avant les déductions d'imp8ts)? l. Moins de $5,000 2. $5,000 a $7,999 3. $8,000 a $11,999 4. $12,000 a $15,999 5. $16,000 et plus Votre mEre a—t—elle un emploi? 1. Pas maintenant, mais elle en a déja en un 2. Non, et elle n'en a jamais eu 3. Oui . I . Avez-vous travaille durant les vacances l'année derniEre? 1. Oui 2. Non Avez-vous un emploi 5 temps partiel durant l'année scolaire? l. Oui 2. Non Quel est votre age? . 17 ans 18 ans 19 ans 20 ans 21 ans 22 ans 23 ans et plus \lO‘U‘l-bUJNl-J O Sexe: l. Masculin 2. Feminin Avez-vous l'intention de vous marier? 1. Je suis deja marié(e) 2. Oui, d'ici un an ou 2 3. Oui, d'ici 3 a 5 ans 4. Non 5. Je ne sais pas 130 50. Dans quelle faculté ou école étes—vous inscrit(e)? 1. Arts 2. Administrations 3. Nursing 4. Sciences 5. Education 6. Sciences domestiques 7. Sciences sociales et du comportement 51. Quelle a été votre moyenne cumulative pour le premier semestre? l. Moins de l 2. l a 1.4 3. 1.5 a 1.9 4. 2 a 2.4 5. 2.5 a 2.9 6. 3 a 3.4 7. 3.5 et plus 52. Indiquer le nom de votre institution: 53. Dans quel comté la résidence de vos parents est-elle située? APPENDIX B PERCENTAGE DISTRIBUTION OF RESPONSES ON ATTITUDINAL ITEMS 131 UNIVERSITE DE MONCTON Moncton, New Brunswick, Canada April 14, 1975 Dear Student: You have been chosen to participate in a study that is conducted among the French-speaking students of New Brunswick. We would like to ask you a few questions about yourself and your opinions concerning a problem that is much discussed: the roles of women in contemporary society. This problem is often debated among women only. We believe that men are also concerned in this respect and that the time has come for men to make their opinions known. This is why we solicit your c00peration in this study. For the first section of the questionnaire, there is no right or wrong answers; it is your personal opinion that counts. Give your responses by reference to the ideal woman for you, the one with whom you wish to share your life, or to the situation that would meet your expectations as partner in a couple. We want to assure you that all information will be treated with confidentiality. Your name will never appear on the questionnaire. WOuld you kindly return the completed questionnaire, using the return-envelope here included. Your cooperation is greatly appreciated. Sincerely, Monique Proulx Ecole des sciences domestiques 132 UNIVERSITE DE MONCTON Moncton, Nouveau-Brunswick, Canada le 14 avril 1975 Cher ami étudiant, Vous avez été choisi au hasard pour participer a une étude que nous menons parmi les étudiants francophones du Nouveau—Brunswick. Nous aimerions vous poser quelques questions sur vous-mama et sur vos opinions face a un probléme fort discuté, 1a place des femmes dans le monde d'aujourd'hui. Ce sujet est souvent débattu, a tort, entre femmes seulement. Nous croyons qu'il concerne aussi les hommes et qu'il serait temps pour eux de faire connaitre leurs opinions 3 cet égard. C'est pourquoi nous solicitons votre cooperation pour cette étude. Pour la premiére section du questionnaire, 11 n'y a ni bonne ni mauvaise réponse, c'est votre opinion personnelle qui compte. Répondez en vous référant 3 une femme idéale pour vous, celle avec qui vous souhaitez partager votre vie, ou encore a la situation qui comblerait vos attentes comme membre d'un couple. Soyez assuré que toute information demeurera strictement confidentielle. Votre nom n'apparaitra jamais sur 1e questionnaire rempli. S'il vous plait de retourner 1e tout le plus t6t possible en utilisant l'enveloppe-réponse ci-jointe. Votre coopération est grandement appréciée. Sincérement, Monique Proulx Ecole des sciences domestiques 133 UNIVERSITE DE MONCTON Moncton, New Brunswick, Canada April 14, 1975 Dear Student: You have been chosen at random to participate in a study that is conducted among the French-speaking students of New Brunswick. We would like to ask a few questions about yourself and your opinions concerning a problem that is much discussed: the roles of women in contemporary society. For the first section of the questionnaire, there is no right or wrong answers; it is your personal opinion that counts. Give your responses by reference to the woman you want to be, or to the situation that would meet your expectations as a woman. We want to assure you that all information will be treated with confidentiality. Your name will never appear on the questionnaire. Would you kindly return the completed questionnaire, using the return- envelope here included. Your cooperation is greatly appreciated. Sincerely, Monique Proulx Ecole des sciences domestiques 134 UNIVERSITE DE MONCTON Moncton, Nouveau—Brunswick, Canada 1e 14 avril 1975 Chére amie étudiante, Vous avez été choisie au hasard pour participer a une étude que nous menons parmi les étudiants francophones du Nouveau-Brunswick. Nous aimerions vous poser quelques questions sur vous-méme et sur vos opinions face a un porbléme fort diScuté, la place des femmes dans le monde d'aujourd'hui. Pour la premiére section du questionnaire, il n'y a ni bonne ni mauvaise réponse, c'est votre opinion personnelle qui compte. Répondez en pensant 3 la femme que vous avez 1e goflt d'étre, ou encore a la situation qui comblerait le mieux vos attentes comme femme. Soyez assurée que toute information demeurera strictement confidentielle. Votre nom n'apparaitra jamais sur le questionnaire rempli. S'il vous plait de retourner 1e tout 1e plus t6t possible an utilisant l'enveloppe-réponse ci-jointe. Votre c00pération est grandement appréciée. Sincérement - .7 Monique Proulx Ecole des sciences domestiques 135 UNIVERSITé DE MONCTON Moncton, New Brunswick, Canada May 12, 1975 Dear Student: You have recently received a questionnaire about the roles of women. If you have not returned your responses yet, you would make us a favor by doing so as soon as you can. Your Opinions are important and we would like to be able to include them in our study. Thank you for your cooperation. Sincerely, Monique Proulx Ecole des sciences domestiques 136 UNIvERSITé DE MONCTON Moncton, Nouveau-Brunswick, Canada le 12 mai 1975 Cher(e) étudiant(e), Vous avez recu récemment un questionnaire sur le rSle des femmes. Si vous n'avez pas encore retourné votre réponse, vous nous rendriez grand service en le faisant le plus t8t possible. Vos opinions sont importantes et nous aimerions pouvoir les inclure dans notre étude. Merci de votre coopération. .Sincérement, . ’3 Monique Proulx 137 Table 22 Percentage Distribution of Responses on Attitudinal Itemsa Response Categories Item No Number Sex 1 2 3 4 Response 1 M 31 53 13 3 F 45 48 6 2 2 M 70 23’ 2 5 F 59 34 7 O 3 M 21 47 16 16 F 19 43 22 16 4 M 4 18 22 56 F 1 9 25 65 5 M 55 28’ 6 10 1 F 67 20 6 7 6 M 18 25’ 34 23 F 13 32 34 20 l 7 M 27 46’ 19 8 F 23 37 28 10 2 8 M 45’ 35 17 3 F 64 26 9 l 9 M 44 23 15 8 F 23 30 33 14 l 10 M 48 34 ’15 3 F 58 28 12 2 l 11 M 21 5322’ I8;’ 8 F 32 45 17 6 12 M’ 65 21 10 4 F 68 24 6 2 13 M 27 31 12 30 F 10 15 17 56 1.6 14 M 12 20 39 29 F 20 40 20 20 l 15 M 32 31 21 16 F 16 35 25 24 16 M 45 228 14 12’ 1 F 26 34 25 14 l 17 M 226 40’ ’17 13 4 F 20 37 23 19 1.6 18 M ’49: 31 10 ’10 F 53 35 6 6 19 M 64 24 10 2 F 48 34 13 4 20 M 30 25 31 14 F 37 30 26 7 21 M 216’ 41 33 11 F S 21 46 27 2 22 M 49’ 35 ’2’12 4 F 37 37 19 7 23 M 16’ 25 21 37 1 F 13 21 30 34 2 24 ’fi' 4 22 41 33 F 2 4 33 60 1.6 25 ’M 18* ’37 Sufi 25* 1 F 14 41 30 15 l 26 M 9 22 48* 22 F 10 32 29 29 27 M 51’ 22 8’ 182’ F 39 26 15 20 28 M 14 33 41 10 2 F 9 35 41 11 3 29 M "28 41 18 ’11 2 F 29 37 24 9 l 30 MT 46’ 30 14 9 1 F 60 24 10 6 BIBLIOGRAPHY BIBLIOGRAPHY Babbie, E. R. Survey Research Methods. Belmont, Calif.: WadswortH PubliShing Co., 1973. Bandura, A., and R. H. Walters. Social Learning and Personality Development. New York: HOlt, Rinehart and Winston, 1963. Bayer, A. E. "Sexist Students in American Colleges: A Descriptive Note.“ Journal of Marriage and the Family 37 (1975): 391-397. Bernard, J. "Age, Sex, and Feminism.“ The Annals of the American Academy 415 (1974): 120-137. Biddle, B. J., and E. J. Thomas. "Basic Concepts for Classifying the Phenomena of Role." In Role and Role Theory: Concepts and Research. Edited by B. J. Biddle and E. J. Thomas. New York: John Wiley and Sons, 1966, pp. 23-42. Biller, H. B. "Father Absence and the Personality Develop- ment of the Male Child." In Personality and Socialization. Edited by D. R. Heise. ChiEago: Rand’McNally Co., 1972. Brim, O. G. "Adult Socialization." In Socialization and Society. Edited by J. A. Clausen. Boston: Little . Socialization After Childhood. New York: John Wiley afid’Sons,2l966. Brogan, D., and N. G. Kutner. "Measuring Sex—Role Orienta— tion: A Normative Approach." Journal of Marriage and the Family 38 (1976): 31—40. Brown, R. W. Social Psychology. New York: Free Press, 1965. (:arisse, C., and J. Dumazedier. Les femmes innovatrices. Paris: Editions du Seuil, 1975. ’ 138 139 Clarkson, F. E., S. R. Vogel et a1. "Family Size and Sex-Role Stereotypes." Science 167 (1970): 390-392. Cogswell, B. "Socialization in the Family: An Essay on Some Structural PrOperties of Roles." In Sourgebook in Marriage and Family 3rd ed. Edited by M. B. Sussman. New York: Houghton Mifflin, 1968, pp. 366-377. Dahlstrom, E. The Changing Roles of Men and WOmen. Boston: Beacon Press, 1971. De Beauvoir, S. Le deuxieme sexe. Paris: Gallimard, 1949. Deutsh, H. La psycholggie des femmes. Paris: Plon, 1949. Dunn, H. "Marriage Role Expectations of Adolescents." Journal of Marriage and the Family 22 (1960): D'Andrade, R. G. "Sex Differences and Cultural Institutions." In The Development of Sex Differences. Edited by E. Maccoby. Stanford,Calif.: StanfOrd University Press, 1966, pp. 174-204. Eicher, M., and L. Primrose. "A Bibliography of Materials on Canadian WOmen." In WOmen in Canada. Edited by M. Stephenson. Toronto: New Press, 1973, pp. 291- 327. Epstein, G. F., and S. L. Bronsfat. "Freshmen View Their Roles as WOmen." Journal of Marriage and the Family 34 (1972): 671-672. Feldman, K. A., and T. M. Newcomb. "The Impact of College: Epilogue." In Personality and Socialization. Edited by D. R. Heise. Chiéago: Rand McNally, 1972. Francis, B. R. "Effect of Maternal Employment on the Socialization of Sex Role Achievement in New Zealand." Unpublished doctoral dissertation, Cornell University, Ithaca, New York, 1975. Freeman, L. C. Elementary Applied Statistics. New York: John Wiley and Sons, 1965. Freud, S. An Outline of Psychoanalysis. New York: Morton, 1949. 140 Friedan, B. The Feminine Mystique. New York: W. W. Morton, I963} Gingsberg, E., et a1. Life Styles of Egucated WOmen. New York: Columbia University Press, 1966. Greer, G. The Female Eunuch. London: MacGibbon, 1970. Gregoire, M. Métiers de femmes. Paris: Plon, 1970. Haavio-Manilla, E. "Sex Differentiation in Role Expecta- tions and Performance." Journal of Marriage and the Family 29 (1967): 569-578. Harrocks, J. E., and D. W. Jackson. A Theory of Self- Process and Role Behavior. Boston: Houghton Mifflin, Inc., 1972. Hobart, C. "Egalitarianism After Marriage: An Attitude Survey of French— and English-Speaking Canadians." In Women in Canada. Edited by M. Stephenson. Toronto: New Press, 1973. Hartley, R. "Children's Concept of Male and Female Roles." Merrill-Palmer Quarterly 6 (1959-60): 83-91. . "Sex Roles from a Child's Point of View." In Sourcebook in Marriage and the Family. 3rd ed. Editedby M. B} Sussman. New York} Houghton Mifflin, 1968, pp. 266-274. Hjelle, L. A., and R. Butterfield. "Self-Actualization and WOmen's Attitudes Toward Their Roles in Contemporary Society." The Journal of Psychology 87 (1974): 223-230. Hoffman, L. W., and F. I. Nye. Working Mothers. San Francisco: Jossey-Bass, 1974. Hunt, R. G. "Role and Role Conflict." In Current Perspec— tives in Social Psychology. Edited by E. P. HolIander andIR. G. Hunt. London: Oxford University Press, 1967. .Jacobson, A. H. "Conflict of Attitudes Toward the Roles of Husband and Wife in Marriage." In Role and Role Theor : Conce ts and Research. Edited by B. J. BIddIe and E. J. Thomas. New York: John Wiley and Sons, 1966, pp. 297-310. 141 Jenkins, S. H. "Relationship of Teacher Characteristics to Rapport Between Teachers and Mexican-American Students." Unpublished doctoral dissertation, United States International University, 1970. Quoted by E. Shostrom in Manual: Personal Orientation Inventory. San Diego, Calif.: EducaEIOnal and Industrial Testing Service, 1972. Kagan, J. "Acquisition and Significance of Sex-Typing and Sex-Role Identity." In Review of Child Development Research. Vol. 1. Edited by M. Hoffman andfiil Hoffman. New York: Russell Sage Foundation, 1964, pp. 137-167. Kaley, M. "Attitudes Toward the Dual Role of the Married Professional Woman." American Psychologist 26 (1971): 301-306. Kammeyer, K. “The Feminine Role: An Analysis of Attitude Consistency." Journal of Marriage and the Family 26 (1964): 295-305. Katz, D., and E. Stotland. "A Preliminary Statement to a Theory of Attitude Structure and Change." In A Study of Science. Vol. 3. Edited by S. Koch. New York: McGraw-Hill Book Co., 1959. Kohlberg, L. "A Cognitive Developmental Analysis of Sex- Role Concepts and Attitudes." In The Development 9f Sex Differences. Edited by E. MacOOby. Stanford, Ca11f.: Stanford University Press, 1966. Komarovsky, M. "Cultural Contradictions and Sex Roles: The Masculine Case.” The American Journal of Sociology 78 (1973): 873F884. Lipman-Blumen, J. "How Ideologies Shape Women's Lives." Scientific American 226 (1972): 34-42. Lennard, H. L., and A. Bernstein. PatEgrns ig Human Interaction. San Francisco: Jossey—Bass, 1969. Macpherson, A. G. Studies in_Canadian Geography: The AplantiglProvinces. Toronto: University of Toronto Press,’1972. 142 Meier, H. C. "Mother-Centeredness and College Youth's Attitudes Toward Social Equality for WOmen: Some Empirical Findings." Journal of Marriage and the Family 34 (Ll972): 113-131. Millett, K. Sexual Politics. New York: Doubleday, 1970. Mischel, W. "A Social-Learning View of Sex-Differences in Behavior." In The Development of Sex Differences. Edited by E. Maccoby. StanfOrd, Calif.: Stanford University Press, 1966, pp. 56-81. Money, J., J. G. Hampson, and J. L. Hampson. "Imprinting and the Establishment of Gender Role." A.M.A. Arch. Neurol. Psych. 77 (1957): 333-336. Mowrer, O. H. Learning_Theory and Behavior. New York: John Wiley and Sons, 1960} Mussen, P. "Early Sex-Role Development." In Handbook pf_§ocializa§ion Theory and Research. Edited by D. Goslin. ChiOago: Rand McNally, 1969. Nie, N. H., D. H. Bent, and C. H. Hull. Statistical Package for the Social Sciences. New York: 2McGraw-Hill Book Co., 1970. Osmond, M. W., and P. N. Martin. "Sex and Sexism: A Comparison of Male and Female Sex-Role Attitudes." Journal of Marriage and the Family 37 (1975): 744-758. Pineo, P. C., and J. C. Porter. "Occupational Prestige in Canada.“ The Canadian Review of Sociology and AntrhOpology 4'(1967): 24-39. Prefontaine, M. "Role Orientation of College WOmen in Three Different Types of Institutions.“ Unpublished doctoral dissertation, Cornell University, 1969. Rapoport, R., and R. N. Rap0port. Dual Career Families. Baltimore: Penguin Books, 1971. Rocher, G. "Images of WOmen in Society." UNESCO, Inter— national Social Science Journal 14 (1962): 131-137. Sears, R. R., E. Maccoby, and H. Levin. Patterns of Child Rearing. Evanston, 111.: Row, Peterson, 1957. 143 Shostrom, E. Manual: Personal Orientation Inventory. San Diego, Calif.: Educational and Industrial Testing Service, 1972. Steinman, A. "A Study of the Concept of Feminine Role of Fifty-One Middle-Class Families." Genetic Psychological Monographs 67 (1967): 275-352. Srole, Leo, et a1. Mental Heglth in the_Metrgpolis. Vol. 1. New York: McGraw-Hill Book Co., 1966. Srole, Leo. "Social Integration and Certain Corollaries: Exploratory Study." American Journal of Sociology 62 (1956): 63-67. Von Le Fort, G. La femme éternelle. Paris: Editions du Cerf, 1948. VOgel, S. R. "Maternal Employment and Perception of Sex Role Among College Students." Developmental Psychology 3 (1970): 384-391. S E R A R m L Y .n s R E w N U E T A“ VIII-2 (“rung“)