P9? Bak‘ The and Ork; tend: Inorfi accei tO'pr quah are n 0r"n‘ HOn-a prESe ei 9 Q m 3‘? . ‘ ABSTRACT ; ‘ COLLEGE WOMEN'S SELF ESTEEM . AND ATTITUDES TOWARD WOMEN'S ROLES By Sharon Berliner Dunbar Current research has called attention to sex differences in personality and socialization processes, perhaps best articulated in Bakan's (1966) conceptualization of the agency/communion dichotomy. That is, the stereotypic American woman tends to be more dependent and non- achieving; the American man, independent and achievement- oriented. Explorations of antecedent socialization processes have tended to confirm arguments that women may generally be reinforced more by parents for communal attitudes and behaviors. Warm, accepting attitudes by the parents, particularly by the mother, appear to promote stereotypic femininity, while reinforcement for agentic qualities, and perhaps less unconditional acceptance by the parents are necessary for the development of agentic traits. While some previous research has suggested both that achieving or "non-feminine" women are maladjusted Elli that the feminine, non- achieving woman is neurotic, the current study attempted to present a somewhat more complex point of view in arguing that 313131: traditional or contemporary female attitudes and life-style may result from high 9; low self-esteem. Supportive of this point ha at: WC n I1 u! Sharon Berliner Dunbar of view was the Loevinger model of ego development, which suggests that the mature individual, male or female, must somehow integrate agentic and communal qualities, rather than remaining at a societally stereOtypic level. The current study attempted a delineation of four female "types" based on high and low levels of esteem and degrees of liberality in attitudes towards women, with a subsequent investigation of these women's perceptions of themselves and parents in terms of agentic/ communal qualities. Using objective measures, it was hypothesized that college women with high esteem would rate their parents as having been more communal, and women with contemporary attitudes toward women would rate their parents as having been more agentic. It was further hypothesized that women with high esteem and contemporary attitudes toward women would rate their mothers as less communal than their fathers, while the reverse would hold for high esteem. traditional women. Finally, it was hypothesized that high esteem women with contemporary attitudes would rate themselves as most equally agentic and communal, while low esteem, contemporary women would rate themselves as being most agentic; high esteem, traditional women, as most communal. A 4 X 4 analysis of variance design was used in which four levels of the esteem and attitudes toward women variables were used. §s were randomly discarded from the original 300 to form a group of 224 with 14 _S_s in each of the sixteen cells. F3. eq In M Sharon Berliner Danar Only the hypothesis that high esteem groups rated their parents as more communal than low esteem groups was tentatively supported. It was also found, however, that _S_S tended to rate themselves and mothers higher on 99313 agency and communion than fathers, that _S_s rated themselves and bOth parents as more communal than agentic, and, finally, that the middle esteem and medium-high attitudes toward women groups tended to rate themselves as having more equal amounts of agency and communion. The low esteem and traditional attitudes toward women groups rated themselves as having less balanced amounts (i. e. , were more communal). Approved Jeanne E. Gullahorn Committee Chairman Date COLLEGE WOMEN'S SELF ESTEEM AND ATTITUDES TOWARD WOMEN'S ROLES By Sharon Berliner Dunbar A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1975 To Phil ii CC Is D1 an gr 2’? his he tic [I‘E {hi ACKNOWLEDGEMENTS I would like to give a very special thanks to a very special person, Dr. Jeanne Gullahorn. Through her role as chairperson of my committee, she has given me a great deal of her time, as well as a personal support and encouragement for my research efforts which I shall always remember and appreciate. Through my years as a graduate student, and in this final project, Dr. Lucy Ferguson has given me both invaluable clinical training and at times a needed personal understanding for which I am very grateful. Dr. Norman Abeles has served on both my master' 3 and doctoral guidance and research committees and deserves a special thanks for his time in these, as well as for the clinical expertise and training he has provided me. I would also like to thank Dr. Terry Allen for serving as statis- tical consultant for my dissertation. Finally, Ms. Cynthia Haas deserves special credit for her tremendous statistical assistance with the computer. Without her, this project might never have been completed. iii CH CH CH. TABLE OF CONTENTS CHAPTER I Rationale for the Investigation The Problem Theory and Supportive Research Parental Behaviors and "Feminine" Personality Development Motivations for Agentic Behaviors "Adjustment" and Sex Typed Role Behaviors Testable Hypotheses CHAPTER II Design of the Study Sample Procedures Instruments Tennessee Self Concept Scale Attitudes Toward Women Scale The Child Rearing Practices Report Adjective Self Description Checklist CHAPTER III Results Additional Findings CHAPTER IV Discussion Suggestions for Future Research Summary iv 20 29 32 47 49 49 49 50 50 52 54 57 58 67 69 72 74 APPENDIX A Questionnaires and Keys 78 General Instructions 78 Background Information 78 Two Factor Index of Social Position 81 Tennessee Self Concept Scale 82 Attitudes Toward Women 88 Child Rearing Practice Report 92 Adjective Checklist 95 REFERENCES 103 LIST OF TABLES Table 1 Correlation Matrix Table 2 Significant Correlations Table 3 Cut -Off Scores for AWS-TSCS GrOUpS Table 4 Analysis of Variance Table 5 Two-Way Analysis of Variance Discrepancy (C -A) Scores APPENDIX B General Information For All Subjects APPENDIX C Means and Standard Deviations for AWS-TSCS Groups on Agency and Communion Scales I. Means (Per Item) and Standard Deviations on Agency -Communion Scores for All TSCS Groups 11. Means (Per Item) and Standard Deviations on Agency -Communion Scores for All AWS Groups III. Means (Per Item) and Standard Deviations on Agency -Communion Scores for All Subjects and For "Extreme" Groups IV. Mean (Per Item) Discrepancy Scores (C-A.) For All AWS-TSCS Groups A ' V. Means for all AWS-TSCS Groups for Self, Maternal and Paternal Communion and Agency Scores* vi 61 62 63 65 66 96 97 97 98 99 100 101 APPENDD< D Subjects with Housewife vs. Working Mothers 102 vii , . 12.339“ tit g( SI .E Q\ CHAPTER I Rationale for the Investigation The Problem Considerable attention has been directed recently to the study of sex differences in personality and socialization processes, and to questions of the relative degree of influence Of biological and cultural factors in male and female development. The core differences between the sexes to which both researchers and theoreticians have addressed themselves center chiefly on differences in perceptual style, orienta- tion to self and others, and in the relative salience of achievement goals. Various terms have been used to describe these contrasting styles: the principles of Logos (male) vs. Eros (female); allocentric (male) vs. autocentric (female) ego styles (Guttman, 1965); field independence (male) vs. field dependence (Witkin et al. , 1954); agency (male) vs. communion (Bakan, 1966). In the main all these distinctions seem to be saying that a woman's perceptual style and approach to life are more subjective and receptive, interpersonally sensitive, non- achieving and passive. She tends to see herself as being in interrelationship with the external, non-self world, as opposed to a man's active, analytic, individuated perception of self as being separate and independent. Thus women are viewed as being more socially responsive and dependent on others for feelings of self-esteem, while men are considered to be less socially sensitive, 2 more independent and, according to some, possessing greater ego strength. Witkin (1965, 1969) has demonstrated that in tests such as the Rod and Frame Test and the Embedded Figures Test, in which it is theorized that a clear body image (and hence, it is assumed, a differentiated inner self) is necessary for correct responses, women tend to be more "global" or "field dependent" (i. e. , they have more difficulty separating ground and figure and in using body cues effec- tively). These findings have been considered evidence that the "field dependent" person is more dependent on the external world and has less differentiated or separate sense of self. While Witkin notes that field dependence is more characteristically a female trait, and that it is relatively stable individually, though age specific (peeking in late adolescence) over time (Witkin, Goodenough and K arp, 1967), he also points out that it is related to scores on masculinity- femininity scales and hence does show some variability within one sex (i. e. , more "feminine" women are more field dependent). Other theorists (Sherman, 1967) have indeed questioned the validity of considering such tests as the BET and EFT to be measures of a pervasive cognitive style, suggesting that they are indicative only of spatial ability, an ability which does in fact seem to be less well developed in women (see, for example, Anastasi, 1958; Tyler, 1965). Guttman (1965) has argued that men have been described as having greater ego strength merely because of the way in which the 3 term has been defined. The basic definition of ego strength has been in terms of delay and derivatives of delay (e. g. , "secondary pro- cess"), a definition which appears to have originated in connection with Freud's assertion that a strong ego results from the decisive reconciliation of the instinctual drives to reality forces (secondary process dominating primary process in the Oedipal situation), a situation presumably less conflictual and hence less decisively resolved in the girl's development. Guttman asserts, however, that this definition of ego strength has more relevance for men than women. According to Guttman, male and female ego functions develop in and are coordinated to significantly different kinds of environments and hence to judge female ego strength and functioning to be regressive and/ or maladaptive on the basis of comparisons of their responses to those of males in a given test situation (e. g. , the TAT) is really unwarranted. In his previous study and analysis of TAT protocols from white urban middle class adults, women were in fact found to be less rational and more erratic, personalizing, and emotional in their responses. However, while men who gave evidence of using a disciplined "secondary process" (task-oriented, non-emo- tional) approach to the task also tended to score high on measures of morale and social effectiveness, the reverse trend was found for women. That is, women whose TAT protocols contained more "primary process" thought and content often scored very high on contentment, social effectiveness, etc. Guttman proceeds to point 5‘ '.‘Ill. :5.“ 9‘ WI oi w} wj fit 4 out that the "maturational milieux" of men and women are quite different and that in fact the average white middle-class woman's phenomenal world, her home, _i_s_ an extension of herself. It thus seems reasonable that if her primary experiences are in a world which is constant, bounded, and in which human relationships are quite personal and important, she will deveIOp the autocentric, field dependent, less differentiated self which seems characteristic of women generally. The white middle-class man's world, in contrast, is a more changing, unpredictable, open world in which objectivity and mastery are much more crucial, and in which interpersonal relationships are frequently less personal and more instrumental: hence the development of the typical male perceptual style. The idea of contrasting "maturational milieux" for men and women should be supplemented by Jane Loevinger's conceptualization of ego development (Loevinger and Wessler, 1970). The Loevinger model postulates a hierarchical number of stages of development which occur in an invariable order, each more complex than that which preceded it. According to Loevinger, however, personal adjustment is _r_1_gt_: to be equated with higher stages in the deveIOpmental sequence. Rather, each stage reflects the individual's attempt to cope with increasingly deeper problems of his or her life experience, integrating the resolution of past experiences with the issues of the present. Hence one may speak of good "adjustment" in any stage of ego development. In terms of the Loevinger model, however, it may C4 (1: 5 be seen that the "higher" stages of ego development represent an integration of stereotypically defined "masculine" and "feminine" modalities of experience, termed by Bakan (1966) agency (masculine) and communion (feminine). According to Bakan, these two "fundamental modalities" are characteristic of fig living forms, agency (being concerned with the individual) manifesting itself in self-protection and self-assertion, and communion (being concerned with the individual in interrelationship) manifesting itself in the sense of being at one with other organisms. The developmental task of the organism therefore becomes an effort to mitigate agency with communion. Loevinger's first stages, the impulsive and self-protective, in which the child, whether male or female, is concerned largely with self-centered self- assertion and self- expression, thus represent a virtually unmitigated "agency". The next level, conformity, in which the child becomes concerned with identification with significant others, and in acquiring appropriate sex roles and other conventional behaviors, however, is in fact experienced differently by boys and girls. Here boys are taught to control the expression of affect and feeling, while at the same time they are encouraged in more assertive or aggressive (agentive) behaviors. Girls, on the other hand, are encouraged to control aggression and assertion and to express affect and feelings (basically communal behaviors). The higher stages of development, the conscientious, the autonomous, and the integrated, in which the 6 examination and evaluation of self relative to one' s internalized values and cultural expectations lead to (in the integrated stage) a personalized identity which transcends cultural Stereotypes of masculinity- femini- nity, represent an integration of "agentic" and "communal" behaviors. The integration of these two fundamental modes of experience and behavior have most: recently been referred to as representing an ideal of androgynj (see, for example Bern, 1974; Rossi, 1964). Because of the predominantly differing emphases on agency and communion in men and women in our culture, however, the process of attainment of the higher levels of ego development in fact differs for men and women (Block, 1973). For men, the integration of agency and communion requires a tempering of self— assertion with awareness of interdependence and mutuality; for women concern for conformity and mutuality and the submergence of self must be integrated with self- assertion and expression. While this framework does in fact point to largely incomplete developmental tasks for both sexes in our society (and not merely the under-development or lack of agency in women), we shall concern ourselves here only with feminine development. The purpose of the present study was, thus, to explore the rela- tionship between various combinations of agentic and communal attitudes in college women, in conjunction both with level of self- esteem and with their perceptions of communal and agentic child- rearing attitudes and behavior in their parents. It is hoped that this 7 initial, largely exploratory research, rather than proposing causal relationships, will be a beginning attempt to gain a greater under- standing of the antecedent variables which may be involved in the growth of basic self-esteem and its relationships to the development of women's attitudes towards women. The relevant literature will now be reviewed. Theory and Supportive Research There appears to be considerable research evidence for sex differences in interpersonal style and orientation. The Douvan and Adelson ( 19 66) interview study of 3000 adolescents nOted that in sixth to twelfth graders, one of the clearest differences in the sexes was in the degree of coherence between fantasy and reality conceptions of future adulthood: the boys dreams and plans seemed appropriate to each other and centered around work goals and achievement; the girls, however, showed a marked discontinuity between their plans, which were concentrated on events before marriage, such as education and jobs, and their dreams, which related to marriage, personal attraction, etc. Throughout, the girls focused chiefly on interper- sonal roles, were much more compliant and dependent on parents, and less mature in the development of achievement and moral autonomy. Thus already for the girl a division has been formed between the "outer" objective world of work and the personal world of home, marriage, and human relationships. It is not clear to what extent the girls were consciously aware of the discriminations they were (l) ft 00 \R t} 8 expressing, but it seems plausible to assert that these distinctions represent to a large degree the culture's current double message to women that they should be both "traditionally" feminine and self- actualizing and productive. Clearly, then, a crucial focus of investigation of the antecedents of perceptual style and interpersonal orientation must be in the forma- tive years when parental attitude and behavior are most influential, and when socialization forces generally begin to give evidence of their impact. As a first step in the further exploration of female personality and experience, then, some brief consideration of self concept development and socialization processes seems appropriate. The basic sense of self probably develops fairly early, around the age of two, when the child is first coming to be able to sense his separateness from the mother, manifest overtly by his use of the pronoun, "I. " This core sense of identity includes body image and feelings, sexual identification (i. e. , knowing whether he is a boy or girl but not yet all the ramifications, role expectations, etc. , which later accompany sexual identification), and some beginning ability to discover techniques for enhancing and/ or protecting feelings of self worth. The crucial significance of the parents in both influencing the formation of the child's self concept and in conveying acceptance and positive valuation to the child has been discussed by countless theorists and researchers (e. g. , Freud, 1936; Horney, 1937; Angyal, 1951; Wylie, 1961). Murphy (1947) stated, "the tendency to value )arental approval. . . " )int of view in con- This is to certify that the - appralsals of the thesis entitled college Women' s Self-Esteem And force in the culture Attitudes Toward Women's Roles , ahaviors through which presente eptance. In learning Sharon Berl bar " essions, once formed has been accepted towar s fulfillment Other values’ Whatever 0' the '9’qu ments for onflict with established ‘_£ttt11._de in my _ ' l re1nforcement of the ' conceptualize them, 15” J L. , . // Major professor \. :arrying with them the a child's self, therefore, d characteristics given because the child's om, ‘ ' ife are a chief source 'or'secunty—merrpreseiyafionsoon becomes fioal in itself' (Lecky, 1954, p. 84). In this schema, then, the child's level of self-esteem may be conceptualized as being a function of how well he perceives himself to be performing in each of the roles in which he participates (more central or salient roles being more important), and the extent to which the range of roles and his performance in them uses his perceived capacity (Bardwick, 1971). 9 rather than disvalue the self is correlated with parental approval. . . " (p. 522). Sullivan (1947) expressed a similar point: of View in con- ceiving of the self as developing from "reflected appraisals" of the parents and significant others (p. 10). Thus parents are the primary socialization force in the culture who convey to the child the roles, values, and behaviors through which the child may acquire and maintain love and acceptance. In learning these values, Lecky (1954) nOtes that these impressions, once formed in the child, "set the conditions for rejection of other values, whatever their nature, which would tend to precipitate a conflict with established perceptions. " (p. 129). Parental affirmation and reinforcement of the child's role behaviors, however the parents may conceptualize them, thus become for the child a source of security, carrying with them the expectation of parental love and acceptance. The child's self, therefore, "tends very strongly to maintain the direction and characteristics given to it in childhood" (Sullivan, 1947, p. 131). And because the child's self concept and concomitant general scheme of life are a chief source of security" their preservation soon becomes a goal in itself" (Lecky, 1954, p. 84). In this schema, then, the child's level of self-esteem may be conceptualized as being a function of how well he perceives himself to be performing in each of the roles in which he participates (more central or salient roles being more important), and the extent to which the range of roles and his performance in them uses his perceived capacity (Bardwick, 1971). ll ‘ -‘| ML'va‘ F: g t f“? rt at e); an In it [O 2 'pc ind, 10 Accepting the importance of the parental role in socialization and recalling the previously cited differences in adult male and female perceptual style and orientation, what is the typical female socializa- tion experience, and how might this differ from that of the male? Indeed, at the pre-school level (age 3), when most children are able to verbalize not only to which sex they belong, but also which jobs are done by men and women (Dubin and Dubin, 1965), "appropriate" sex role typing appears to have little to do with self-esteem gratifica- tion. That is, for all children in this age group the mastery of cogni- tive, physical, verbal and social skills is of prime salience: age is a more important factor than sex in determining appropriate (i..e. , rewarded) behaviors (Emmerich and Smoller, 1964). However, even at this very early age there is evidence that parents have differential expectations for girls and boys (e. g. , permissiveness for aggression), and that these expectations are quite clearly conveyed to the children. In Goodenough's (1957) study of two to four year olds and their parents it was found that parents expected girls (as assessed through interviews) to be more interested in people, thus potentially promoting the more "personal" communal orientation of females to the environment. And, indeed, the girls in their drawings both drew more people and sponta- neously mentioned people more often than did the boys. One particular aspect of the socialization process, that of identification, appears to have special significance for the development of male-female persona- lity differences. mm 3 w 1e an the 11 In a very cogent discussion of female socialization, Donelson (1973) argues: The most telling and pervasively crucial difference between men and women in their personality and concomitant behavior occurs in large part because those factors which facilitate continual identification are more typical in the experiences of women than of men. Eros, autocentricity, field dependence - a sense of together- ness of self and the rest of the world - facilitate and in large part result from identification with Others. (p. 13) According to classical psychoanalytic theory, identification of the child with the parent originates as a result of the child's thwarted retaliation against the aggressor. However, many writers since have pointed out that hostility and fear are not the only motivations for identification: love for the parent also is a factor. In broader terms, identification may be conceived of as a social responsiveness process (Donelson, 1973) facilitated by the "learner" and "model" being in relationship to one another such that the learner perceives satisfaction of his needs by the model as contingent on his displaying appropriate (modeling) behavior and/ or attitudes. Bronfenbrenner (1960) has distinguished three classes of phenomena to which the term "identification" has been applied: behavior, motive, and process. While the evidence for similarity in behavior between model and learner is more straightforward, it is the underlying motives for and processes in identification to which many theorists have addressed themselves. _A In he in to 1311' 12 Kagan (1966) has argued that the chief motive underlying identifi- cation is a threat to the person's self-esteem. Thus if the child feels dependent on the parent for love and approval, his anxiety, feelings of helplessness and fear of the loss of parental support will lead him to identify with (i. e. , emulate appropriate behaviors of) the parent. Two principle processes of identification have been distinguished: "Identification with the aggressor" (A. Freud, 1946), sometimes termed "defensive" identification (Mowrer, 1950), is motivated by a fear of the loss of parental love, usually in a punishment situation. "Developmental" identification (Mowrer, 1950) is considered to be a result of the child's love of the parent and is an effort on his part to reproduce and incorporate aspects of the beloved parent within himself as a means of c0ping with his separation from the parent. The distinction between these two, while perhaps useful for discussion, may in reality be difficult to maintain. Because the parent does not M meet the child's needs, the child will likely have some ambiva- lent feelings about the parent (Ferguson, 1970): he will have positive feelings when. the parents meet his needs and hence tend to identify positively or "developmentally" to maintain these satisfactions; and he will have negative feelings when his needs are not met or are interfered with and hence may feel compelled to identify "defensively" to recover the satisfactions he feels he has lost. Also relevant to the identification process is the particular disci- plinary style which the parent uses to obtain desired child behaviors. 13 Bronfenbrenner (19 61) has argued that girls are exposed to more affection and less punishment in socialization: they receive more "love-oriented" discipline (1. e. , "psychological" techniques repre- senting a "communal, " personally oriented focus). The consequences of the use of this type of discipline, he states further, may have negative as well as positive aspects. While fostering the internaliza- tion of adult standards and the development of socialized behavior, it also appears to have the effect of undermining capacities for initiative and independence (1. e. , "agentic" behaviors), as well as creating more anxiety and conformity. Thus girls are more obedient and cooperative (Sears e_t 31: , 1957), but they are also more anxious, timid, dependent and sensitive to rejection (Bronfenbrenner, 1961). Given more affection, the girl experiences more anxiety with the threat of its removal, and hence a circular process is begun in which being given more affection leads to more dependency on Others and their approval, which leads to greater conformity and identification. These ideas are supported by Hoffman (1970), who argues that "withdrawal of love" disciplinary techniques intensify the need for adult approval, and Coopersmith (1967), who notes that "dependency behaviors represent an attempt to confirm or establish a definition of one's worth or capacities" (p. 221). While other research (see Ainsworth, 1967) has indicated that maternal affection alone cannot account for the strength of the child's dependency and, indeed, that a relatively functional attachment of the child to the mother ht (I (a in Da Ca bt 1'0r 14 should be discriminated from a more dysfunctional over- dependency of the child (see, for example, Ainsworth's "secure-attached" vs. "insecure-attached" child groupings), arguments asserting differences in parental attitudes and behaviors based on the sex of the child would still appear to be viable and hence to warrant further exploration. In addition to the hypothesized use of methods of discipline which tend to foster social dependency, conformity, etc. , the content of the roles for which the girl achieves reinforcement also relate to the development of the "autocentric" or "communal" personality pattern. The typical female role requires greater impulse control, conformity (Sears e_t 3.1.- , 1957), cooperation and passivity (Komarovsky, 1964). There is greater restriction on the girl's freedom, which in itself facilitates dependency on others and strengthens identification (Kagan, 1966). Restrictions on freedom in turn bOth insure that the individual will remain in the situation Laid. tend to increase his feelings of helplessness and dependence on the person who is in control. Freud (1933) suggested that the primary original form of identification (anaclitic) is a consequence of the undifferentiated perception of the infant in which the external object, the mother, is perceived as a part of the self. Secondary identification occurs only after the child can discriminate the object from his self (and is motivated chiefly by anxiety in the Oedipal situation). By inhibiting freedom and maintaining closer ties with the mother, the process of socialization for a girl thus seems to make much more difficult the development of i] \III II .ul l JI-Il‘lllil. \II . [II iii 15 these feelings of independence, autonomy, and a separate sense of self, for as Witkin (1969) has argued, physical and psychological movement away from the mother are vital factors in the development of body and self differentiation. In line with this theoretical reasoning Douvan and Adelson (1966) found that in the vast majority of cases, adolescent girls never went through a period of rebellion against their parents and continued to look to their parents for guidance, yielded to their demands, and tended to identify with parental authority. Summarizing the conse- quences of the typical female socialization process, Donelson (1973) states that it produces . . people who are socially sensitive and loving but also dependent and anxious; the dependency and anxiety maintain the social responsiveness. They lack the firm ego boundaries necessary for sense of self as separate from others and are thus dependent on others for self definition but also are able to "touch' others. Such characteristics favor continuing habits of pervasive and intense identification. In turn, identification supports dependency. (p. 29) Over time in the socialization process, role demands and expec- tancies change. That is, behaviors which might be acceptable in a young child of either sex (e. g. , activity, competitiveness, more "agentic" behaviors), become much less acceptable in the adolescent girl. Bardwick (1971) notes that pressures for traditional feminine identification do not really begin until adolescence, when dating and heterosexual activities become increasingly important. Heilbrun (1964) adds that, not only is the girl permitted more freedom as a U) 1‘. it tC DE SE Da 02} id{ C01 Drj 16 child to engage in cross-sex behavior, but also the traditional female role is changing, so that she is exposed to conflicting social rewards and punishments, no matter what role she assumes. This idea is in accord with Douvan and Adelson's discussion of the "integrative" task required of the adolescent girl (i. e. , integrating "masculine" activity and "feminine" role behaviors), as well as with Loevinger's conceptions of integrated ego development. In discussing the socialization and identification tasks which confront the young girl and adolescent, it should be kept in mind that identification and role modeling are not all-or- none processes. The strength of any identification can vary, and the child or adolescent can be identified at the same time with a variety of models (Kagan, 1966). In this regard, Slater (1961) distinguishes between "personal identification" (the child emulates the personal traits of the model) and "positional identification" (the child behaves in a way appropriate to the model's position, role, etc. ). Personal identification does not necessarily include sex- role appropriate behaviors, and in an earlier study (Slater, 1955), Slater showed that personal identification with parent was positively correlated with personal identification with the other. Lynn (1962) similarly postulated a difference between "parental identification" and "sex- role identification," and adds that a child could be identified with his same- sex parent but not with the "appro- priate" sex role, and vice versa. Lynn (1966) argues further that the n: M yak ha is» nea ckx aspt neet rnot ptin rest ship; ODIy TUOrE and I A gir DGrSO perVa: 17 basic {page of the identification process for males and females is different. Both males and females originally identify with the mother but the boy is forced to shift his identification subsequently (Sears pg a_l. , 1957). Thus the male tends to identify with a cultural stereotype of the masculine role (because his own father is usually not present nearly as much as is his mother and because the male role is much clearer than the female), whereas the girl tends to identify with aspects of her own mother's role specifically. Because she does not need to seek out an object of identification, but need only imitate her mother, Lynn suggests that the girl acquires a learning method that primarily involves a personal relationship and imitation rather than restructuring and abstracting. Thus learning in a personal relation- ship, her need for affiliation is stronger (Edwards, 1959), and needing only to imitate, rather than abstract or restructure, she becomes more dependent on the external context of the perceptual situation and more responsive to the standards of others (Witkin e_t 31: , 1954). A girl's relationship to her parents thus is crucial for her subsequent personality development, in part because of her presumed more pervasive identification with them. In a study using doll play with first graders and parental inter- views, Mussen and Rutherford (1963) found that the highly feminine girls (assessed by the IT scale) had mothers who were perceived as more nurturant and powerful and fathers who were more masculine (Wu-n... the her 866 wh i beh. Whjl late1 Corn 18 and who clearly encouraged more sex- appropriate play. Fish (1969) found in a study of female college students that degree of femininity was positively correlated with the girl's estimate of the amount of time the father spent at home and thus presumably was available to the child. Thus the affective quality of the girl's relationships with her parents, in addition to the parent's behaviors themselves, also seems crucial. Sears (1970) found degree of femininity in girls to be positively correlated with the father's, but not the mother's "warmth" (assessed by maternal interview data acquired seven years earlier). The intensive study of nursery school children by Sears, Rau and Alpert (19 65) however indicated that degree of femininity in girls was po_t related to either maternal warmth and nurturance, 21; to maternal power. Femininity was related negatively to parental permissiveness of sex play, and this finding seems consistent with the general observation that femininity is fostered by parental discouragement of aggression, promotion of submission to them, and a general nonpermissiveness and restriction more characteristic of traditional or conservative childrearing attitudes and behaviors, which also represent a reinforcement of communal rather than agentic behaviors (Sears e5 _a_1_. , 1957). Kagan and Moss' (1962) longitudinal study gave evidence that, while the mother's behavior appeared to be unrelated to her daughter's later degree of femininity, femininity of interests and activities were correlated with maternal "protectiveness" (i. e. , encouraging — W ’ ...., de; (19 he} by evic rel; 0f tl 01) vi 19 dependency) in early childhood. Similarly, Mussen and Rutherford (1963) found the girl's femininity to be related to her perception of her mother as being warm, nurturant, and affectionate (corroborated by interview ratings of the mother). Other studies have also indicated trends of the more feminine girls to have warm and nurturant relations with mothers (Slote, 1962; Preston, 1965, unpublished doctoral dissertations). In summary, then, b_otl_i_ parents should be considered involved in the girl's sex- role and personality development. There appears to be evidence that femininity in the girl is facilitated both by nurturant relations with the mother, and by the father's distinctive treatment of the two sexes. In specifying the behaviors that facilitate femininity, it becomes obvious further that individual differences in parental nurturance, expectancies, etc. will be crucial in facilitating or inhibiting femini- nity'in girls. Thus inspite of the "feminine" socialization process 9! described earlier, as Bronfenbrenner (1961) notes, . . socialization techniques do contribute to individual differences, within the same sex, precisely in the types of personality characteristics noted above [obedience, dependency, etc.]" (p. 10). In line with this fairly obvious reasoning, it is clear that many girls and women are asser- tive, active, and independent (1. e. , less "feminine" or more "agentic"), and many do "achieve" (in the traditional "masculine" sense). These facts raise some interesting questions as to how a non-traditional, star and ntaL have begj trad beha 20 independent female personality might develop: is her socialization more like the "typical" male's or is it different yet; is her mother less nurturant and the father less concerned with sex- typing, or are family relationships different; and, finally is the development of female independence, achievement, etc. , at the expense of some characteristics of femininity (i. e. , sensitivity, nurturance, etc. sacrificing the positive aspects of communion for agency -- or is an integration of the two attained7). In attempting a beginning under- standing of the factors involved in female autonomy, independence, and femininity, it will be necessary to investigate both studies of male-female differences and the relatively fewer studies which have explored contrasting female patterns of personality. We shall begin with an exploration of some of the parental antecedents of the traditionally "masculine" (agentic) and "feminine" (communal) behaviors. Parental Behaviors and "Feminine" Personality Development Bardwick (1971) has argued that it is go; direct training for independence or achievement that is crucial in motivating a child to develop an independent sense of self, but his realization that he will not be consistently rewarded. The child will then learn to make his own judgments of his behavior in order to make himself less vulnerable to parental rejection. He learns, therefore, that he may be rewarded (or punished) for a behavior, such as achievement, and not just for bei g. Bardwick adds that if the mother is predominantly supportive 21 and the father takes the "traditional" role with his daughter of being less instrumental and more nurturant, then neither parent is likely to push the daughter to develop an independent sense of self or to achieve, "divorced from affiliative cues" (p. 174). Veroff (1969) states directly that the development of achievement motives in girls requires a somewhat rejecting attitude by the mOther when the girl is young, an appropriate timing of stress and mastery when. She is in middle childhood, the mother's acceptance pf the appropriateness _o_f female achievement and a female role model who is nOt too strong and domineering. In addition, _33 should EE have experienced 1:29 strong ap emphasis pp interpersonal gratification during early child- hood. The two important points presented here thus seem to be first that achievement is viewed by the mother as being just as acceptable for a girl as for a boy, and second that interpersonal relationships have nOt been extremely rewarding for the child. In this situation, the girl might then feel that her needs for love and approval had not been met completely satisfactorily (in affiliative and parental rela- tionships or that she did not feel adequately esteemed there), and that she might gain this approval through achievement. This argument, thus raises the possibility that not all achievement is motivated by the individual's desire for personal competence, mastery, etc. ; it may also be motivated by dependency on others and/ or in the face of feelings of low self-esteem, a need to prove self-worth and accepta- bility to others. I [ 22 Crandall, Preston and Rabson (1960), in a study of achievement motives and mothers' behaviors with three to five year old girls and boys (using ratings of free play behavior in nursery school and at home) found that the mother's independence training and her rewarding of achievement were positively related. Mothers who encouraged achievement tended to be less nurturant, but the re was no clear-cut evidence that maternal affection was related to achieve- ment behavior. However, in a study of six to eight year olds (Crandall, Dewey, Katowsky and Preston, 1964), the mothers of high achieving girls (measured on standard academic achievement tests) page less affectionate and less nurturant than the mothers of less proficient girls (as assessed by parental interviews). There were many more significant relationships between parents' attitudes and behaviors and their daughters' as opposed to their sons' achievement. The more proficient girls also had fathers who more often praised and less often criticized the girls' intellectual or achievement efforts, and the mother's evaluations of and satisfactions with her child's general intellectual competence were positively related to the child's actual academic performance. The authors argue, similarly to Bardwick (1971), that it is not the mother's emphasis on independence and mastery (the "achievement" girls' parents did not express a greater valuing of intellectual expression), but her emOtional relation- ship with the child. Thus when the mother had a more "hostile" attitude | l tit cc re da KE (B; 01 Ve fa] Ch fr( be 19: ind not 23 toward the daughter in preschool, the daughter developed higher achievement motives as an adult. The ascription to maternal behaviors of the label "hostile, " however, should be viewed with caution. As noted by other researchers (see Donelson and Gullahorn, in press), it seems more likely that a moderate level of warmth (i. e. , less than a maximum or "overprotec- tiveness" and more than a minimum or total rejection) is the most conducive to the development of achievement motivation. Indeed, research has shown that highly protective mothers generally have daughters who are passive and conforming (i. e. , not achieving, see Kagan and Moss, 1962), but that coercive punishment (hostility) by the parent also often fosters low autonomy and low achievement (Baumrind, 1971). The use of the term "hostile" for moderate levels of warmth may have resulted in fact from a biased sampling (e. g. , very few parents willing to volunteer for longitudinal research on families could truly be considered hostile and rejecting toward their children and hence the full range of parents has not been sampled), or from the sex biases of the researchers, who View the same parental behavior directed toward a "fragile" little girl to be more hostile than that same behavior directed toward a "robust" little boy (see Hoffman, 197 2). Kagan and Freeman's (1963) report of a longitudinal study also indicates the complexity of the issue of the mother's emotional rela- tionship to the girl and the development of achievement motivation. 24 They studied children from two to seven and again at 13 to 18, in conjunction with parental observation and interviews. Daughters with higher IQ scores had mothers who were more "acceleratory" during ages four to seven, and these mothers had also been more critical during the ages two to four. Thus the pattern of initial criticality (perhaps perceived by the child as a means of gaining acceptance), which Veroff (1969) hypothesized, might be considered supported by this study. An additional finding showed, however, that severity of discipline and restrictiveness by the mother at ages four to seven in girls were associated with dependency and conformity in the girls at adolescence (assessed from interviews and self- rating scales). While no indications of any correlations between conformity and dependency and achievement/non- achievement were reported, it would be interesting to know just how the mothers of high achievement girls at four to seven differed, if at all, from the girls whose mothers were restrictive and severe in discipline. The implication seems to be that, while the "high achievement" mothers were not overly nur- turant and unconditional in their acceptance, they did not convey the degree of severity and potential rejection in their relationship with their daughters which the mothers of the conforming, dependent adolescent girls did. These ideas of "moderation" in parental warmth as facilitating achievement motivation also fit well with Baumrind' s (1971) distinction between the "authoritative" parent, who is sensitive to the child's 25 needs, clear in his communications, and firm but not harsh in discipline, and who, apparently, produces an achievement-oriented child, and the "authoritarian" parent, who uses coercive, punitive discipline, leading to dependent, non- achieving child behavior. Thus while independence and achievement are not necessarily correlated, this research seems to indicate that the two may often be found together. Kagan and Freeman further note that girls who had slightly higher IQ's (which may be indicative to some extent of higher achieve- ment motivation) also rejected traditional sex role behaviors, raising the issue of whether the deveIOpment of agentic motivations in women necessarily implies the inhibition of stereOtypic feminine (communal) behaviors or, concomitantly, whether stereOtypic "community" implies inhibition of agency. Douvan and Adelson (1966) discriminated two types of adolescent girls answering differently to the question, "With whom do you spend the most spare time?" Girls answering with the modal response ("my family") were seen to be less autonomous, to have less part in rule-making, to rely more on external authority, and to be punished more often both physically and psychologically. Those girls answering, "with friends" (28% of sample), showed the reverse of these characteris- tics and indicated that their parents expected them to be independent, though at times they wished to be closer to them. (The first group indicated greater closeness to their parents and that their parents expected them to get along with others. ) Douvan and Adelson discuss 26 some further characteristics of the more autonomous girl: In all families where the mother is more highly educated than the father, we detect a stress on child autonomy. Girls from these families report that their parents expect them to be independent and self- reliant more than other girls do; and they are less likely to say that their parents want them to be obedient and to respect authority. In that latter category we find a disproportionate number of girls whose fathers are clearly superior to their wives in educational background. Girls in the "educated mother" group are less reliant on adult authorities generally and they seem to be less tied to the family emotionally than other girls (p. 327). Thus it seems that the less traditionally feminine girl may become more independent and achievement-motivated in response to a high 5 achieving maternal model who is also less nurturant, combined with a parental encouragement of qualities of mastery and independence. Shaw and Dutton (1962) used parental interviews and ratings in corre- lation with adolescents' IQ scores and GPA's and found that mothers of female underachievers (those with an IQ of greater than or equal to 110 and a GPA of less than or equal to 2. 7) fostered more depen- dency and required suppression of aggression as opposed to mothers of female achievers (high IQ, GPA greater than or equal to 3. 0). Thus the traits being fostered by mothers of the "non- achieving" girls - e. g. dependency and non-aggressiveness - suggest that these girls will manifest a basically traditional female personality. In addition to consideration of mother-daughter, father-daughter relationships and their effect on achievement and independence, some studies have attempted a somewhat broader assessment and classifi- cation of the family as it affects child and adolescent personality and 27 adjustment. Murphy 33 31. (1963) studied "competent" adolescents (academically effective, good peer relations, etc. ) from their senior year in high school until the end of the first year in college. (Unfor- tunately, no information is provided on sexual composition of the sample. ) Child and parent interviews were employed and adolescents assigned to one of four groups, based on ratings of a research team concerning autonomy (ability to make separate, responsible choices) and relatedness (satisfaction of the adolescent in a predominantly positive relationship with parents). Parents classified as high on autonomy fl relatedness were more clear than others in their values and standards and tended to set down their own example. They provided situations for the child to develop autonomy, trusted his ability to take responsibility (note similarity with the "independent" girls of Douvan and Adelson's study) and were able to accept his mistakes and their own. Parents low on both autonomy and related- ness, on the other hand, tended to perceive their child as an exten- sion of themselves. There was a distinct lack of communication and recognition of individual needs and interests in other family members. Adolescents rated high on autonomy but low on relatedness appeared to have broken away from a family role, with the parents being unwilling to accept their new values. The authors conclude that the extent to which parents are able to satisfy their own needs, indepen- dently of the child rather than through him, affects the way they prepare the child for separation from family and the amount of 28 difficulty the adolescent and family will experience. In considering the "high autonomy - high relatedness" and "high autonomy - low relatedness" families, one may hypothesize that, while adolescents from both types of families appear autonomous, the motivations for their behavior in the two family types might differ radically. In the former situation, it may be considered a positive outgrowth of affec- tive need satisfaction, resulting in higher self- esteem and consequent motivation for and confidence in personal competence and autonomous functioning (i. e. , an integration of agency and communion). In the latter case it might be considered a reaction to an unsatisfying family life in which affective needs have not been satisfied, resulting in motivation to leave the situation (i. e. , agency inhibiting communion). Thus motivations for achievement, independence, or autonomy, may result from a variety of family life situations. When affective needs have not been met by the parents, a girl may hope to gain parental love and approval through achievement. When affective needs have been met, but the girl views the marital relationship or her mother's homemaking role as an unhappy or dissatisfying one, she may wish to escape a similarly unhappy situation and not identify with a "traditionally" feminine mother. In the ideal situation, a girl's affective needs have been met in a family with adequate marital satisfac- tions in which achievement and autonomy are welcomed and rewarded as a logical outgrowth and confirmation of the positive self concept and high self-esteem facilitated by parental love and acceptance .Qidwhfi. 29 (communion- agency integration). In attempting to understand the development of the "non- traditional" feminine personality, then, one must be aware that superficially similar attitudes and behavior may have vastly differing motivational antecedents. Motivations for Agentic Behaviors In this regard several researchers have noted that achievement orientations in women may differ in _bp_th content and motivational antecedents from those of men. 1 Tyler, Rafferty and Tyler (1962) found that girls in nursery school who attempted to obtain recognition for achievement 3139 made more attempts to get love and affection, suggesting to the authors that girls may sometimes use achievement as a primary means of securing love and approval from others. Similarly, Veroff (1969) argues that while the "achievement motive" is easily cued by internal standards of excellence in boys, external support for their efforts seems critically important for girls. It should be noted, however, that while "achievement" for boys is 1French and Lesser (1964) note that "achievement motivation" may be stimulated whenever the situational cues are relevant to the person's central roles. Thus women with a high intellectual role orientation will respond with heightened indications of achievement motivation when stimulated by intellectual cues, but women high in the traditional homemaker role orientation will also exhibit heightened achievement motivation when cues are relevant to that role. While the expansion of the definition of achievement motifit'i'on to incorporate "non-masculine" types of achievement seems both reasonable and important, because our discussion here deals chiefly with the issue of traditional vs. non- traditional female personality development we shall confine ourselves to the question of motivation for stereotypically "masculine" achievement (agency) in women, and, concomitantly, with stereotypically feminine non- achievement and affiliation (communion). 30 nearly always positively reinforced and hence easily becomes internalized, "achievement" by girls has frequently been negatively sanctioned. Thus a girl's heavier reliance on "external support" (i. e. , positive social reinforcement) in order to demonstrate achieve- ment behavior would seem in many cases to be a highly functional approach to take. These writers are thus suggesting affective and interpersonal dissatisfactions as pOtential achievement-motivating forces. Similarly, in discussing the meaning of the lack of feminine goals in some groups of adolescent girls, Douvan and Adelson (1966) hypothesize that it may mean . simply that the girl is psychologically immature and not yet ready to concern herself with any adult role. It may reflect a somewhat masculine concentration or individual achieve- ment and occupational plans . . . . Finally, [it] may represent a rejection of femininity and a serious problem in the girl's concept and self esteem (p. 261). These authors thus make the point that a non- feminine or "mascu- line" personality orientation may or may not be problematic and can be the result of either positive or negative psychological antecedents. In addition, lest one conclude that a clearly feminine orientation, in contrast, is necessarily "healthy, " Douvan and Adelson also present some interesting data on the early (age 11-14) female steady dater who is strongly oriented toward and preoccupied with adult feminine goals and fairly uninterested in personal achievement. These girls, while clearly feminine in orientation and behavior, were viewed as less well developed socially, having a more superficial understanding 31 of interpersonal relationships, and actually as being less introceptive, in contrast to the typical female perceptual style discussed initially. The dynamics of and motivations for this orientation are discussed fairly extensively and are presented here because of their relevance to the present discussion. Barely emerging from latency, with its absorbing concern for real skills and the physical world, she turns precipitously to a single, exclusive relationship with a boy. She has neither skills nor experience in mature object relations. Her only relations to this point have been either the original family ties of childhood, heavy with dependency and shaded by infantile emotionality or the fairly superficial friendships normal in latency. As she enters adolescence, the normal and appropriate period for skill in object ties, she avoids the essential lessons of the period and insulates herself from her peers by establishing an early and singular relationship with a boy (p. 225-226). . . . the function of these normal transitional ties to other girls . . . is to neutralize the ambivalence that marks the adoles- cent girl's relationship to her mother. If the girl for some reason misses her chance for working through her negative feelings toward her mother, she will . . . [have] only a competitive model of feminine relationships, a model inappropriately burdened with undiminished infantile hostility (p. 216). Indeed, Douvan and Adelson note that the young steady daters do £193 have close family ties, experience much family conflict, and do not choose their mothers as adult ideals. They thus hypothesize that the early choice of a heterosexual relationship may be due to hostility toward the mother and represent an avoidance of either a close or potentially rivalrous relationship with female peers, or even, a displacement from a close father-daughter tie. 32 These findings thus support the argument that pith—er a predomi- nantly "feminine" or "masculine" orientation in women can be moti- vated by or originate from either positive or negative familial affec- tive relationships. Further, the rather frequently presented idea that "achievement-oriented" girls or women are less "feminine" (e. g. , Sontag, Baker and Nelson, 1958) would not seem to reflect the complexity of the issues involved. That is, achievement strivings may result as a substitute for affiliative satisfactions or as a means of gaining love and acceptance, (which might result in less "femininity") but they may also develop as a further expression of personal competence in a securely feminine girl with high self-esteem. By the same token, a "feminine" girl may or may nOt develop achieve- ment motivation: if affiliative needs are satisfied and personal achievement is not a central or salient value in the girl's self concept, (she may in fact consider the acquisition of effective social skills as "achievement") then the development of high self-esteem within a basically "traditional" female personality seems likely. At this point, then, some exploration of the research on female roles and personal adjustment will be presented. "Adjustment" and Sex-Typed Role Behaviors Heilbrun (1964) studied "role consistency" in college men and women as a measure of adjustment, using Erikson's theory of ego identity. The adolescent's self-perceived role consistency (measured by the S's rank ordering of adjectives from most to least characteristic igcrlko . L4 33 of him for each of a series of eight interpersonal situations - e. g. , with his mother, with a close friend, etc. ) was correlated with his score on a standard test (checklist) of masculinity- femininity. For women, greater role consistency was related to e_i_th_e_r high femininity or high masculinity; lower role consistency was associated with intermediate scores of masculinity-femininity. Heilbrun theorized that the university provides two principal roles for the college woman, student or girlfriend, and that the coed can become immersed solidly in either one. If she vasCillates between the two, however, she feels at odds with herself and thus achieves a lower level of adjustment. While clearly self-perceived role consistency is not the pply definition of adjustment which might be employed (and role consis- tency is clearly p_O_t equivalent to high self-esteem) this study does suggest the importance of the girl's perceptions of her salient roles and models for adjustment, and that it may be easier to follow a clearly traditional model of femininity g a break away from this definition entirely than to attempt a personally satisfying integration of the two (i. e. , inhibit agency or communion rather than integrate the two). Unfortunately, no antecedents of the development of these differing personality styles were explored so that the findings in this study, and particularly the question as to how these differing salient roles developed, require more extensive exploration. The idea suggested earlier that "femininity" itself is not neces- sarily indicative of good adjustment is confirmed by several «56....» .. . .53.:5‘ 34 researchers. In a study of junior high students, Webb (1963) found that high femininity, assessed by the CPI femininity scale, was associated with high anxiety in girls, measured by the children's Manifest Anxiety Scale. Sopchack (1952) gave MMPI's to college women, which they filled out both for themselves as they thought their mothers and fathers would answer. She found that positive identification with the mother (defined as a high number of identical answers to MMPI questions for self and mother) was correlated with some types of abnormality, but that failure to identify with the father was even more closely associated with trends toward abnormality ("abnormality" not clarified). In addition, "masculine" women were found to identify _l_e_s_§ with their fathers than "feminine" women. Gray (1959), however, in a study of fifth to eighth graders, found no relation between the girl's degree of identification with her mother (as assessed by the child's similarity of rating of parent and self on 40 bipolar adjectives) and level of child adjustment (using the MAS, sociometric measures, tests of practical intelligence, leadership, etc. ). These discrepant results might easily be explained by the fact that no information on either maternal adjustment 2; actual degree of femininity in the mother was collected. The assumptions both that all mothers are "feminine" and that identification with the mother is in all cases a positive goal are of course untenable. A further consideration in level of adjustment, however, is the situational context 35 itself. (That is, does the girl have the skills, traits, etc. , necessary for her particular environment--skills which may vary from environment to environment?) Heilbrun and Fromme (1965) thus found that "adjusted" college girls (those not seeking psychological or vocational counseling) were identified with a less feminine mOther (assessed by the girls adjective checklist ratings of their mothers), while maladjusted girls (those seeking psychological counseling) were identified with a highly feminine mother. While the use of counseling- seeking as the sole measure of adjustment is highly questionable, the authors give one reasonable theoretical interpretation to their findings in saying that in the competitive college environment some degree of "masculine" (agentic) traits is necessary. Recalling Heilbrun's (1964) study of role consistency, however, this argument would seem to hold _o_nly if the academic aspect of college were important to the girl. If both this aspect of college and the "feminine, " interpersonal aspect were important, she might well develop the "fear of success" which Horner (1969) has described, and for which she might subsequently seek counseling. Other studies have focused on "low femininity" or "masculinity" in women and level of adjustment. In the study cited above, Webb (1963) also found that, while there was no relation between sex- role preference and social acceptance, the "low feminine" girls had a greater rate of absenteeism, which the author interpreted as possibly being a form of "acting-out" or a means of showing independence, 36 since interest in school, conformity, etc. , is more typically a female characteristic in earlier school years. These results, however, do not seem to indicate any serious adjustment difficulties in these girls. Almquist and Angrist (1970) conducted a longitudinal study of one class in a women's college of coed university (using yearly question- naires and two interviews) and similarly found that women with career salience choosing atypical occupations for women were neither less active socially, nor did they report more family conflict than other women. There was, however, a high correlation between career salience and the mother's employment during the girl's college years. These results suggest the continued importance of the mother as a primary role model, but the motivation for this assumed identifica- tion with the mother is nOt clear. In a study of 60 outstanding unmar- ried career women by Ellis (1952), it was suggested that "social mobility" (defined as attaining a higher occupational status than that of the father) is frequently inspired at least partially by emotional drives generated by unsatisfactory early primary group relations. The mobility itself has then theorized to lead to a further deteriora- tion of these relations, with accompanying neurOtic symptoms (Horney, 1937). From interview data, Ellis confirmed that a significantly greater proportion of the socially mobile career women had had a series of humiliating childhood experiences, experienced more rejection by the parents, rated their attachment to their parents as "less than average, " had fewer and shorter lasting friendships, 37 more conflict with parents in adulthood, and more psychosomatic symptoms. From these studies it is again clear that the relative salience of achievement and career- orientation may be associated with e_3_ith__pg positive or negative familial antecedents. A more recent study by Heilbrun (1968) supports these conclusions. Here college undergraduates, both clients and non-clients at the psychological clinic, participated in small groups of males and females. All members rated each other on two six-point scales: 1) relationship orientation (considered to be a femininity scale) and 2) goal orientation (masculine scale). Girls thus rated to be more masculine were negligibly less expressive than the girls rated feminine (i. e. , they had both desirable masculine 329 feminine traits), and on the whole the "masculine" girls were better adjusted than the feminine girls on an adjective checklist. The "clinic" girls were more feminine on the whole, but there were few personality differences between the "feminine non-clinic" and "feminine clinic" girls. The "masculine non-clinic" girls, however, were more need- achieving, dominant, enduring and less succorant and abasing than the "masculine-clinic" girls. Heilbrun speculates that the "masculinity" that appeared to be social independence in the case of the clinic girls was actually social alienation (agency inhibiting communion), again alluding to quite different motivational antecedents for superficially similar behaviors. $.39; 38 In facilitating many women's awareness of the possibility of achievement and alternative life styles for women, the rise of the women's liberation movement would seem to have increased both the number of women who will be able to achieve and realize their full potential, and the number who will turn to this newer model of woman as a solution to interpersonal dissatisfactions. Thus far there have been relatively few research projects which have made in-depth studies distinctly focused on the "liberated" and "traditional" woman, and the motivational and familial antecedents of these types. Two projects, however, may be reported which bear on these issues. In a survey of over 1000 married women, ages 18 to 54, Lipman- Blumen (197 2) divided subjects into "contemporary" and "traditional" groups on the basis of their responses to questions about beliefs regarding appropriate behavior for women. These women were all married to graduate students and themselves were college-educated. The actual breakdown of the sample therefore, (27% traditional, 73% contemporary) indicates generalizability may be limited, as does the fact that no levels of statistical significance were reported on the descriptive analysis. Nonetheless, the findings of this study are stimulating and merit consideration. Women with the contemporary sex- role ideology tended to come from families in which most of the time neither parent was dominant, or from families in which the mother was dominant. More of the traditional women, on the other hand, came from families in which 39 the father was dominant in the marriage. Contemporary women also viewed their mOthers as more dissatisfied and tended to regard M parents as unsuccessful though they admired their fathers more than their mothers. They reported more criticism from both parents or from the mOther (the traditional woman reported either more criticism from the father or from neither) and were most likely to have received encouragement from both parents or from the mother to go to graduate school. Lipman- Blumen summarizes: Women who emerge with the contemporary ideology tend in adolescence to achieve a certain psychological distance from their family, to evolve a sense of separateness as individuals (p. 39). The reluctance to please parents, coupled with a decreased admiration for parents and a tendency to see them as frustrating and critical, are closely associated with the contemporary sex role ideology (p. 40). Curiously, the author reports no differences in the two groups in level of self-esteem, and b_oph groups gave low ratings to all of the following traits: perseverance, ambition, ability to work under pressure, competitiveness, realism, self-confidence, enthusiasm, courage, physical attractiveness, and sexuality! These final findings are puzzling, but suggest the possible intrusion into the data of social desirability sets, as well as indicating that the median split of the sample into only two groups failed to capture differences in potentially important variables (e. g. , self-esteem) which might have been found in a finer division of groups (e. g. , high esteem and low esteem contemporary women). 40 The Douvan and Adelson (1966) interview study of sixth to twelfth grade adolescents provides both a more representative and extensive sampling of women, as well as a finer delineation of conceptual female "types". Using a standard measure of femininity, as well as female subjects' responses to a four-item measure of "non- femininity" (e. g. , choice of traditional masculine occupation, desire to be a boy, belief that a boy's life is more desirable, rejection of marriage), they were able to identify six different female personality patterns, two of which were termed "feminine" patterns and four of which were termed "non- feminine. " The "Unambivalent F eminine" group was seen as highly social, well-integrated (defined as having a high level of social activity, having an extended time perspective, having poise and social skill, more sources of self-esteem), less aggressive, narcissistic, intraceptive, and interpersonally mature. This group was the most thoroughly focused on the social and personal aspects of reality and gained self-esteem chiefly from helping others and being nurturant. There was little motivation for personal achievement, and educational goals were nOt high. These girls were also clearly identified with their mothers (they most often chose their mothers or a female relative as their adult ideal) and had close, compliant and dependent relationships with strong, traditional parents. The parents in turn, did not stress independence or resourcefulness in their daughters, nor did they allow them much participation in rule-making. The use 41 of psychological punishment was reported in this group more than in any other. The other feminine pattern, the "Ambivalent Feminine" group was similar to the first in their concern with marriage, motherhood and social development. In addition, however, these girls also had a lively interest in personal achievement in the present fig future and focused principally on individual development. Of _a_l_l_ groups, these girls most often had plans to go to college and most often dreamed about success in school and work. They more often chose masculine models as adult ideals, and these were also more often non-family members than was true of the Unambivalent Feminine girls. On the whole, this group gave evidence of a greater sense of competence and a greater readiness to rely on their own resources. These girls' parents, in contrast to the first group, encouraged independence and autonomy, and gave their daughters a share in rule-making. While the mothers of the first group were usually less well educated than the fathers, many mothers in the second group were better educated than the fathers, were quite ambitious, and worked outside the home. The four non-feminine patterns, while even less homogeneous than the first two groups, all gave evidence of problematic social development. The "Achievement-Oriented" group focused primarily on typically masculine occupational goals and gained self— esteem through job performance and personal achievement. They chose .35..er 42 non-family women as adult ideals, While they did nOt reject marriage, it was not a central concern, and in general these girls seemed less socially developed and less popular than the "feminine" girls. Their relationships with their parents were reported to be generally plea- sant but without strong identifications. Another group, the "Boyish Girls" were seen to be psychologically immature, even "child- like" in their emphasis on current activities, and "boyish" interests such as sports. While on the basis of the girls' reports, their parents did not seem unusually strict, they nonetheless viewed their parents as highly restricitve (perhaps in response to their "masculine" activity preference). The "Neutral Girls" stressed neither feminine nor masculine interests. Like the "Boyish Girls" they appeared socially immature, without being as rebellious as the former, and they were also more compliant and closer to their families than the average. Finally, the "Anti-feminine" girls, who did not want to marry, showed signs of severe pathology and emotional constriction, poor adjustment and self— rejection. Parents of this group were more punitive than any other group (over one-third used physical punishment), discouraged autonomy and emphasized obedience and respect. There were reports of much conflict in the homes, particularly with the mother. These girls were very inactive and immature socially though many had major responsibilities at home for maintaining the household. Of the six groups described, the "Unambivalent Feminine" and "Ambivalent Feminine" personalities would seem to be the most 43 functional and adaptive ones. These two groups are clearly different in their concepts of their central roles, and it would appear that while the parents of the two groups may have differentially reinforced autonomy and achievement efforts, the affective quality of the parent-child rela- tionship may _n_o_t have been highly dissimilar (i. e. , both may have been predominantly positive ones). It is unfortunate that so little specific information in regard to this point is available. This study gives strong evidence, however, that femininity and achievement- autonomy charac- teristics are _n_o_t_: necessarily mutually exclusive, and that it is only when girls use only certain aspects of their potential (either inter- personal pr; achievement), while motivated to gratify other needs as well, that lower self-esteem is likely to develop (Barkwick, 1971). Thus while both the "Ambivalent Feminine" and "Achievement" groups were concerned with personal competence and achievement, the generally lower level of social competence and exclusive concern with achievement of the latter group suggests the possibly defensive nature of this group's adaptation. The quality of this group's relationship with parents, how- ever, is not made clear and would require further investigation, as do the differences between the two "feminine" groups in this regard. Finally, it would seem that the most maladaptive personalities may result from a lack of nurturance or acceptance by the parents, while at the same time no alternative methods for gaining acceptance (e. g. , through achievement) are presented. Clearly, acceptance or approval for being, is preferable to no acceptance under any circumstances. milk»! 3. .rxv.£\ 44 The girl's perceptions of these alternatives, however, will be contingent both on their visibility in her environment (and the parental emphasis on them) and the perceived success with which these alternatives meet (both for her personally and fOr significant others - e. g. , whether her working mother is happy and receives approval from her father). In summary, then, in spite of the pervasive effects of the typically "feminine" socialization received by women in our culture, many women become bOth autonomous and achievement-motivated. While the familial interactional anteCedents of "typical" femininity and female autonomy and achievement are far from completely understood, at least some research suggests that "femininity" is promoted by very high levels of maternal warmth and the fostering of dependency, and by the father's active concern with sex- role typing. Concomitantly, greater independence and motivation for competence has frequently been associated with somewhat Leg parental, particularly maternal nurturance (i. e. , "moderate" nurturance but not hostility), as well as positive reinforcement for mastery and independence. While some have tended to view traditional femininity and achievement-motivation in women as mutually exclusive, or as one group or the other having higher esteem or better adjustment, the present presentation of research argues strongly that this is _n_o_t necessarily the case. Both level of self-esteem and adjustment appear related to the woman's perceptions of her performance in the roles which she perceives as 45 central to her self concept, as well as the motivational antecedents of perceiving these roles as central and crucial ones. In this regard, it may be noted that superficially similar behaviors and attitudes (particularly those assessed through interviews and objective questionnaires) may have vastly dissimilar motivational antecedents. In exploring the development of any conceptual person- ality "type," therefore, one must consider a number of factors in the parent-child interaction: marital satisfactions and needs and their consequent effects on the expectations, demands and love and rejection which are given to the child; the child's methods of obtaining love and approval (including his perception of "central" or crucial roles to be played in order to achieve or maintain love and self-esteem); the child's success in satisfying these needs, and the models which are available to him in striving to satisfy needs and maintain esteem. With particular regard to the development of female personality "types," for research purposes one might hypothesize that while traditional femininity will be fostered only by a predominantly positive, nu rturant relationship with the parents (i. e. , a positive communal orientation), concern with personal achievement requires the encouragement by the parents, particularly the mother, of independence and autonomy (additional agentic emphasis). These two may or may not occur together, but in either case, high self-esteem would be possible only in the situations in which love and acceptance of the child by the parents was bOth present and not entirely contingent on 46 behavioral (i. e. , achievement) performance. In line with this reasoning four differing conceptual "types" of female personality might be explored: 1) girls receiving parental acceptance and nurturance (positive communion), without emphasis on or much reinforcement for achievement, autonomy, etc. , resulting in a fairly "traditional" feminine personality with relatively high self- esteem; 2) girls receiving both nurturance and acceptance and addi- tionally encouragement for mastery, resulting in a "feminine" and achievement-motivated girl with relatively high self-esteem; 3) girls receiving little basic acceptance from either parent ("nega- tive" communion) but receiving some approval, encouragement, etc. , for achievement, resulting in lower self-esteem and an attempt to gain approval through achievement (as a more concrete sign of parental acceptance) with little interest in traditional "femininity;" 4) girls receiving little basic acceptance from either parent, while at the same time pp; finding achievement to be salient as a means of attaining approval or love, resulting in lower self-esteem with perhaps an emphasis on traditional femininity because no other models or methods of attaining love are available except the cultural stereotype. While clearly this is a highly simplified conceptualization of female personalities and their antecedents, it merits further exploration. Thus on the basis of this analysis of previous literature, hypotheses were investigated in the current study: ‘fl. ft. vii .1. .954; 47 Testable Hypotheses 1. College women scoring at the highest levels of objectively measured self-esteem will rate both parents as having more (posi- tively) communal childrearing attitudes and behaviors (on an objective measure); college women with the lowest levels of esteem will perceive their parents as being less (positively) communal. These differences will hold, regardless of the degree of liberality and agentic attitudes toward the roles, rights and responsibilities of women expressed on another questionnaire. 2. College women scoring at the highest liberal levels on an objec- tive measure of attitudes towards women (which encompasses agentic orientation) will perceive their parents as being more agentic in childrearing attitudes and behaviors, while college women with the most traditional attitudes towards women will perceive their parents as less agentic. These differences will hold, regardless of level of assessed self-esteem. 3. There will be an interactive effect within high esteem groups, between liberality of attitudes towards women and perceived agentic/ communal attitudes of the mother and father. a. For high esteem women with liberal attitudes, the father will be perceived as more communal than the mother, while the mother will be perceived as more agentic than the father. b. For high esteem women with traditional attitudes, the mother will be perceived as more communal than the father, while tint? 1. . . anarchy? 48 the father will be perceived as more agentic than the mother. 4. Based on the Loevinger conceptualization of ego development which considers maturity to be an irregration of communal and agentic qualities, the following hypotheses are proposed: a. College women with both high esteem and liberal attitudes will perceive themselves as having the most balanced (i. e. , least discrepant) amounts of agentic and communal traits (using an adjective checklist). b. College women with low esteem and liberal attitudes will perceive themselves as being most agentic. c. College women with high esteem and traditional attitudes will perceive themselves as most communal. CHAPTER 11 Design of the Study Taking subjects with varying combinations of scores on inven- tories of self-esteem and attitudes toward women, this study employed an analysis of variance design to assess differences between groups with various levels of these qualities in their perceptions of agentic and communal attitudes and behavior in themselves and in both parents. Sample A sample of 300 college women volunteers was obtained from the population of Michigan State University women enrolled in intro- ductory psychology courses. Ss received research credits for their participation. The great majority of these women were single, eighteen, and first or second year college students. Procedures Ss were told that their research participation would involve completing a number of written forms concerned with their attitudes towards a variety of things and situations and that there were no "right" or "wrong" answers to the questions, but only their own individual opinions. SS were also told that they could later obtain results from the study if they so desired. (See Appendix A for the complete written instructions given to Ss. ) Each S completed three objective instruments, in addition to completing a self-descriptive adjective checklist and answering 49 ‘7 ”Fifi-(iv - 50 demographic questions as well as questions concerning family back- ground and personal plans. All instruments and coding procedures are included in Appendix A. Testing lasted about one and a half hours, after which SS were told the general purpose of the investigation, asked not to discuss it with others until after the period of data gathering, and thanked for their participation. Instruments In addition to the background data questionnaire (see Appendix A), Ss completed the following instruments. Tennessee Self Concept Scale. The Tennessee Self Concept Scale (1965) is a lOO-item Likert scale consisting of self-descriptive state- ments which the subject uses to portray his own picture of himself. The items may be divided into a two-dimensional, 3 x 5 phenomeno- logical scheme consisting of statements concerning S's 1) identity, 2) self satisfaction, and 3) behavior in the areas of his a) physical self, b) moral-ethical self, c) personal self, (1) family self, and e) social self. This part of the scale contains 90 items, equally divided as to positive and negative statements. The remaining ten items comprise the Self Criticism Scale (SC), all taken from the L-Scale of the Minnesota Multiphasic Personality Inventory and employed in the TSCS in a similar fashion to their use in the MMPI. Two basic forms of the TSCS are available: the Counseling Form and the Clinical and Research Form. 51 The standardization group from which norms were developed was a broad sample of 626 people from various regions in the United States, of both sexes, Blacks and Whites, representing all social, economic, intellectual, and educational levels from 6th grade through the Ph. D. and ranging from age 12 to 68. On the basis of additional research (Sundby, 1962; Gividen, 1959; Hall, 1964), Fitts states that it has not been necessary to expand his normative group, since it appears that further variances in demographic data have not substan- tially changed any scores of the scales. While a large number of scores have been devised from the TSCS, the present research uses only one, the "Total Positive" score. This score reflects the subject's overall level of self-esteem: persons with high scores like themselves, feel they are valuable and worthy, have self-confidence and act accordingly. Those with low scores have many self-doubts, feel anxious, depressed and unhappy, and generally have little confidence in themselves. Fitts reports a mean for total P of 345. 6, with a standard deviation of 30. 7 and a reliability of . 92. One anticipated difficulty in using only total P scores was that S's scoring extremely high might in fact be expressing defensiveness rather than high self-esteem. While conjunctive use of the Self Criticism (SC) score might have been one way of dealing with this problem, there did not appear to be any clear way to take account of, for example a very low SC score combined with a very high P. Actually, in the present study only two S's did in fact exceed ,r M'Wrw ' 52 Fitt's reported upper limit of the normal range of P (i. e. , 421). An "eyeballing" of SC scores on high P S's, however, did not indicate consistently low SC scores. It might also be added that no significant differences on SC could be found by Fitts on a comparison of his normative group with 369 psychiatric patients, thus calling into some question the precise nature of the defensiveness that a high SC might indicate. Thus it did not seem worthwhile to combine SC and P scores, particularly since it was preferable to retain as many S's as possible in this exploratory study. Because of the long and widespread use of the TSCS and its clear acceptance as a viable instrument, it was not felt necessary to compute internal consistency for this particular sample. Such a computation would further appear to have limited usefulness, as the items do in fact seem to tap several different aspects of the self concept, aspects which may or may not be posi- tively associated in a given subject (e. g. , an S may have high "social" self-esteem but low "family" esteem). Attitudes Toward Women Scale. The Spence and Helmreich Attitudes Toward Women Scale (AWS, 1973) is a 55-item Likert scale assessing the liberality of S's attitudes toward the "prOper" roles of women in six general areas: 1) vocational, educational, and intellectual roles; 2) freedom and independence; 3) dating, courtship, and etiquette; 4) drinking, swearing, and dirty jokes; 5) sexual behavior; 6) marital relationships and obligations. Normative data have been collected on over 1000 male and female 53 college undergraduates at the University of Texas at Austin and on some of their parents (n=524), and parent-child correlations have been computed. In addition, the scale has been factor- analyzed for both college male and female responses. Three main factors were extracted for the males: attitudes relating to traditional notions about masculine superiority and the patriarchal family; attitudes toward equality of opportunity for women; and beliefs about the social- sexual relationships between men and women and what constitutes "lady-like" behavior. Two factors emerged for the females: attri- butes of the "conventional woman" in her relationship to men, and equal treatment of men and women in vocational and educational endeavors. Data from the California Personality Inventory Femininity Scale collected on 267 male and 343 female students who had also been given the AWS indicated no significant correlations between the two scales (1. e. , r = . 07 for the males; -. 05 for the females), and hence no relationship between masculinity-femininity and the AWS. Comparisons of the male-female student samples and the parental (mother-father) samples revealed mean scores to be higher for the women in both cases. The difference in the older sample (i. e. , mother- father), however, was less than that of the younger sample. Further, in comparing both responses of the student with his or her same- sexed parent, as well as the responses of the tOtal parent sample with those of the student samples, scores of the older group tended to be lower 54 (i. e. , more conservative). Correlations computed between genera- tional and marital groups indicated significant but relatively small r's: father-son, . 34 (p <. 01); mother-daughter, . 29 (p < . 01); and mOther-father, . 49 (p < . 001). For the Spence and Helmreich samples, the mean score for college women was 98. 21 (S. D. 23. 16) with a range of scores from 35 to 161. In a study of 105 Michigan State University women, Berliner (1973) obtained a mean score of 111. 7 (S. D. 17. 5) with a range of 52 to 152. This higher mean might be considered a com- bined function of regional differences, societal changes towards greater liberality over time, and perhaps also of the fact that the smaller sample was a more biased one (e. g. , only women were allowed to participate). While no direct reliability or validity information has made available by the authors, Spence and Helmreich suggest that the similarity of sample distributions from the fall, 1971 and spring, 1972 samples indicates indirectly that a reliable phenomena is being tapped by the scale. The Child Rearing Practices Report. The CRPR (Block, 1965) consists of 91 socialization- relevant statements, typically adminis- tered in a Q-sort fashion with a forced-choice, seven-step distribution. BOth first person and third person (mother/ father) forms have been used. To date, the CRPR has beeh administered to more than 6000 persons (in both the United States and foreign countries), with an age 55 range of 16 to 50, and varying socio-economic (unskilled workers to professionals) and educational levels (sixth grade to advanced degrees). Reliability has been assessed in two test- retest studies. The first study sample contained 90 undergraduates tested and retested on their own childrearing attitudes after an 8-month interval. The average correlation between the two tests was . 707. The second study involved 66 Peace Corps volunteers who described their parents' childrearing orientations. The follow-up testing was completed three years later. The range of cross-time correlations was . 04 to . 85 for maternal descriptions, . 13 to . 85 for paternal. The average correlations for males and females combined were . 64 (descriptive of mother) and . 65 (father). Construct validity has been measured by determining the degree of relationship between self- report as indexed by the CRPR and _ap_tpa_l maternal behaviors as observed and rated (by Q- sort) in three structured situations designed to assess achievement emphasis, modes and degree of control, and independence training (see Block, 1964). Eight types or clusters of mothers were identified on the basis of observed parent-child interactions. Four years later, the mothers were administered the CRPR and the results compared with t-tests. Many significant differentiating items were found for the eight groups, which would appear to validate the CRPR as reflecting at least par- tially actual maternal behaviors. However, the small number of 56 mothers in some of the clusters have attenuated these relationships at least to some extent. In four different sample groups of parents, Block was able to extract 23 items from the 91 which showed some significant sex differences (in at least two of the four samples) consistent with the agentic/communal differences discussed previously. (These samples included parents of nursery school children, of children with physical illnesses, of high school students, and of university students. ) That is, eleven items expressive of a "communal" orientation were significantly more emphasized by parents of girls (at the . 10 level or better), while twelve items expressive of an "agentic" orientation were significantly more emphasized by parents of boys. For the present study, only these 23 items,, presented in Likert scale fashion in both third-person mother and third-person father form, in addition to 15 other randomly chosen "filler" items from the CRPR, were administered to Ss. Thus each S responded to 61 items (11 maternal communion, ll paternal communion; 12 maternal agency, 12 paternal agency; 15 "filler"), presented in random order and received 4 scores: maternal communion, paternal communion, maternal agency; paternal agency. Due to the different number of agency and communion items, it was necessary to use "average item" means rather than total score means in comparing communion and agency scores. WIMP 57 Adjective Self-Description Checklist. SS were given a list of 20 descriptive adjectives and asked to rate themselves in Likert scale fashion on each adjective from 0 (not at all characteristic of me) to 4 (extremely characteristic of me). These 20 items were taken from Block (197 3). A group of adjectives was given to four psychologists who were instructed to identify those adjectives clearly expressing agency, those expressing communion, and those that were neutral or irrelevant to those concepts, according to the Bakan (1966) definitions given to the judges. These items were subsequently given to male and female samples in 6 different countries, including the United States, who were asked to choose the items closest to their "ideal self" descriptions. Those adjectives for which interjudge classification reliability was high and for which some statistically significant (at the . 10 level or better) sex differences were found in at least one of the samples were used for the present study. (See Appendix A for a list of these adjectives. ) Ss received three scores from the checklist: a "communion" score based on S's tOtal score from the ten "communion" adjectives; an "agency" score, based on S's total score from the ten "agency" adjectives, and a "discrepancy score, " (Communion score - Agency score) indicating the disparity between the amount of communal characteristics and agentic charac— teristics a given S perceived herself as having. CHAPTER III Results The average subject in this study was 18 to 19 years old, single, and a freshman. Her father (who had slightly more education than her mother) had some college training. Aproximately half of the Ss' mothers (n = 141) worked while S was growing up. The average sub— ject also intended to graduate from college and felt that both a career and marriage and family life were at least somewhat, if not very, important to her. She reported receiving at least some encourage- ment to go to college from both parents. While most Ss' mothers spent most of their time at home while they were growing up, SS tended to feel that their mothers were only somewhat satisfied in this role. Appendix B summarizes the means and standard deviations of the major variables, as well as ranges and internal consistency of scales where applicable. Codes for the interpretation of mean values may be found in Appendix A with the relevant questionnaires. The mean and standard deviation for the Tennessee Self Concept Scale (TSCS) are remarkably similar to those obtained by Fitts (i. e. , Fitts-i = 345. 6, S. D. 30. 7). For this sample, the mean and standard deviation in the Attitudes Toward Women Scale (AWS) are more discre- pant from previous research (i. e. , Spence and Helmreich x — 98. 21, S. D. 23. 16; Berliner i = 111. 7, S. D. 17. 5). The internal consis- tency is acceptably high, however. 58 " “Winn. 59 Thrning to the agency and communion scores, it appears clear that the average S tended to perceive herself and both parents as more communal than agentic. The fairly low degree of internal consistency for these scores indicated in Table 1 suggests some problems in the scales themselves which should be borne in mind in the further interpretation of the data to follow. Table 1 presents the correlation matrix for all twenty-four major variables, including the Ss' background and demographic data. For n = 300, .the critical r (p < . 05) is . 1143. The large n of the sample, size (n = 300) allows a very low r (i. e. , . 1143) to obtain statistical significance at the . 05 level. However, while a correlation of . 11 for two variables is statistically significant such a correlation would in fact account for only one percent of the variance (r2), which is clearly pg a substantively meaningful proportion. From this point of view, the correlational matrix presents very few "significant" correlations. Looking only at those correlations of . 40 or greater (accounting for sixteen percent of the variance or better), it may be seen that maternal and paternal communion scores, maternal and paternal agency scores, and reports of the extent of maternal and paternal encouragement to go to college are significantly correlated (. 40, . 48, and . 60, respectively). Other significant correlations (. 40 or greater) include age and year in school, . 72; father's occupational and educational status, . 68; mother's and father's educational status, . 51; mother's educational .Rlz....&a. . . PROP—fl... 60 and father's occupational status, . 42. (See Table 2. ) Since the correlation of the TSCS with the AWS was . 08 and hence was not statistically significant, these scores were used together in an analysis of variance to test hypotheses regarding self-esteem and liberalism in attitudes toward women, four levels of scores on each of the relevant questionnaires were used in a 4 X 4 analysis of variance. That is, Ss scoring high, medium-high, medium-low and low on both of the two variables were placed in a given cell. For computational convenience and to minimize effects of violations of assumptions (e. g. , unequal variances), Ss were randomly discarded (using a random numbers table) to form cells with equal _r_1. As a result of this process, 14 Ss filled each of the 16 cells, leaving an _n of 224 for the analysis of differences in self, maternal and paternal agency and communion scores. Cut-off scores for the AWS- TSCS groups are given in Table 3. Tables 4 and 5 present the ANOVAS. Means and standard deviations for the four "extreme" AWS- TSCS groups, for the marginal (AWS- TSCS) groups for agency and communion scores for all sixteen groups may be found in Appendix C. In terms of the :' (”331: . 61 Table 1 - Correlation Matrix omnamopma N N o s o o N o o Ho SH _N Ho as No Ho so Ho Ho No NH NN No N» amom N ooo .oo .No .No .oo .Ns .oN .oN .oN .oo .oN .oo .oN .oN woo woo .Hs .oo .No .oo .No .No .oo foo ozm N .oo ooo no» .Ns «No nos woo .oN nos .oo ”No .No .oo .oo .oo woo .oo .No .No «we woo woo THN .oo mo o .No no» woo .oo .os .oN .oN .No .oo «No now .oo .oo .oo .oN .oN .oN .os .os .oo .oN .oo noo eoo m» s .No .o» .oo ooo «No .oo .oo woo ._o «oo uoo now qoo woo noN .os .No .oo .No aoN .oo .oN .oo ooN mo o .oo no» .oo «No Noo «oN woo .No .No nos .os now .oo «oN .oo .oN .oo .oo «No .os woo qu woo woo go o .Ns .oo .oo .oo .oo ooo «_o .so .oo «No no» woo woo .oN uoN .oN .oo .oo .HN .No .No .No .No woo 2» N uoN «so .oN .oo woo "No ooo .oo .oo uoo .oN «NH .oo woo «oo .oo no» woo .oN woo .oN .No «oN foe on o ._N .oo .No "oo ._o .so .oo ooo now now .No "No «Ho .oN «No «No .oo .oo .oo .No .oo .oo .oo uoN o» o «oo nos .oo .Ho «No .oo .oo .oo ooo «oo «oo «oN .os .oo uoN «op "oN «oo .oo .oo .No .No .oo .oo >mm No «oo .oo «No Moo «op «oo «oo uoo woo woo .oo .NN uoo .oo .os uoo no» now now woo moo uoN ToN .oN 2m NH .oN .HH .oo «Ho .os no» .oN .Ho uoo .oo ooo .oo .oo "oo .oo .oo uoo ”No uoN .oN noN moo nos noN aw HN .oo .oo “oo doc now "so “SN "Ho noo .NN .oo ooo uoN .oo uoN uoN woo .oo uoN uoo «No .oo .oo noo e.oon No uoN ”NH .oo .oo .oo woo .oo "Ho .os uoo .oN qu ooo .No .oo .SN uoN .oo .os woo woo woo uoN nos x «on N» .oN .oN uoo noo uoN .oN "oo .oo woo .No woo .oo .No Poo .oo .oo noN .oo .oo «oo woo nNN «oo .oo woos Ho ”Ho uoo uoN uoN .oo uoN now "No uoN .os .No uoN .oo .oo ooo .o~ uoN qoo .oo uoo woo "No .oo nos zoos so «so .oo .oN .os qu .oN .oN "so «oo .oo .oo uoN .SN .oo .oo aoo uoo "om noN goo "so ”so .oo who H oo .o» .oo aoN .No “Ho .No nos .oo uoN nos woo .oo qu woo qu woo ooo ._s .No qu .NN .NN .oN .oo N No .oo .NN .os .oo "oo .oo "om .oo uoo qoo "so .oo .oo .No How Moo .os ooo .No use .oo .oo “NH .oN m o No .oo .No .os .No .No .NN .oo .oo .oo uoo uoN qu .os .oo .oo qu .No .No ooo qos to» .oN woo foo .u" s No .oo no» .oo qu .os .No woo .No .oo uoo .oN qoo uoN woo woo qoo uoN «so no» woo oo .oN .HN moo mxuo N_ .oo uoN .oN .oo now .No .No .No .HN ”so qu "Ho ”om uoN qoo "so .NN .os .os .oo .oo .oo .o» qu mwro NN .No “om .oo .oo uoN .No .No .oo .No aoN moo .oN ”Ho "NN «No "HN .HN .oo .oN .oN Too Foo .No qNN mu N No .oo ”HN Woo .oo uoN .oo uoN .No .oN .oN nos .oo noN moo .oo .oo .oN ”NH uoo .oN ._s .No Foo moo o N» .oo .oo ”Ho uoN Moo ”om .os qu .oN .oN uoN uoo nos .oN nos ”so .oo .oo .oo moo noN nHN woo woo 62 Table 2 Significant Correlations (r 9 . 40) Maternal Communion-Paternal Communion Maternal Agency— Paternal Agency Maternal Encouragement-Paternal Encouragement Age -Year in School Father's Occupation —F ather' 3 Education Mother's Education-Father's Education Mother's Education-Father' 3 Occupation .40 .48 .72 .68 .51 .42 63 Table 3 Cut-Off Scores for AWS—TSCS Groups AWS Low Med-Low Med-High High TSCS AWSé95 965AWSél 15 l loéAwsfl 36 AWSél 31 Low TSC 8531 3 TSC S531 3 TSC Sé3l 3 TSC S531 3 AWSé95 965A W551 15 l 165AWSél 36 AWSél 37 Med-Low 3145TSC $5343 3145TSCS5343 3145TSCS£343 3l4éTSCSé343 AWS595 965AWSél 15 1 l6£AWSél 36 AWSEI 37 Med- High ‘ 3445TSC Sé373 3445TSCSé373 3445TSC SE37 3 3445TSC S537 3 AWSé95 96£AWS51 15 l 16£AWS51 36 AWS-*1 37 High TSCSE374 TSC S5374 TSC Sé374 TSC $3374 64 previously formulated hypotheses, the results of the analysis indicate the following: 1. There is tentative support (p < . 10) that Ss scoring high on the TSCS also had high scores on maternal and paternal communion scales. 2. There is no support for the hypothesis that Ss scoring high on the AWS would also have high scores on maternal and paternal agency scales. 3. There is no support for the hypothesized interactive effect within the high TSCS groups between AWS scores and maternal and paternal agency and communion scores. 4. a. There is no support for the expected interactive effect between the TSCS and AWS scores in relation to discrepancy scores (i. e. , the prediction that Hi AWS - Hi TSCS, Ss would have the lowest discrepancy scores). b. There is no support for the hypothesized interactive effect between the TSCS and AWS scores and self- agency scores (1. e. , Hi AWS - Lo TSCS Ss would have the highest self- agency scores). c. There is no support for the expected interactive effect between the TSCS and AWS scores and self-communion scores (i. e. , the prediction that L0 AWS - Hi TSCS Ss would have the highest self- communion scores). tuner. .I $5355.. 65 Table 4 Analysis of Variance Sum of Source Squares df MS F P Eta2 Between TSCS 3. 2791 3 l. 0930 2. 4528 p <. 10 . 007 AWS 1. 3405 3 . 4468 l. 0028 n. s. TSCSxAWS 3. 6002 9 . 4000 . 8976 n. 3. error 92. 6857 208 . 4456 (. :l95)*** Within MFS* 6. 7901 2 3. 3951 17. 4201 p <. 001 . 036 TSCSxMFS 2. 4734 6 . 4122 2. 1150 p <. 10 . 005 AWSxMFS . 7375 6 . 1229 . 6301 n. s. TKZSXAWSXMFS 4. 5614 18 . 2534 l. 3003 n. 3 error 81. 0765 416 . 1949 - (. 1704)*** AC** 54. 9388 l 54. 9388 116. 58843 p<. 001 . 11548 TSCSxAC 2. 7345 3 . 9115 1. 93434 n. s. AWSxAC l. 3476 3 . 4492 . 95327 n. s TSCSxAWSxAC 3. 33420 9 . 37046 . 78167 n. 8 error 98. 0144 208 . 47122 - (. 2060)** MFSxAC 10. 62369 2 5. 31184 21. 92891 p<. 001 . 02233 TSCSxMFSxAC 2. 19561 6 . 36593 1. 51067 n. s. AWSxMFSxAC 1. 75541 6 . 29256 1. 20777 n. s. TSCSxAWSx MFSxAC 3. 45319 18 . 19184 . 79197 n. s _error 100. 7700 416 . 24223 - (. 21182)*** Total 475. 7118 *MFS - Mother vs. Father vs. Self **AC - Agency vs. Communion ***error 1‘... Eda! ) .Zfin‘fii‘w. 66 Table 5 Two-Way Analysis of Variance Discrepancy (C-A) Scores Source SS df MS F P Eta2 Between TSCS 3.8732 3 1. 2911 3. 2888 p . 10 . 0429 AWS 4. 5126 3 1. 5042 3. 8316 p .. 05 . 04995 TKISXAWS . 32923 9 . 0366 . 09318 n. s. . 00364 Within 81. 65429 208 . 39257 Total 90. 3693 223 67 Additional Findings A number of significant results were found for which no hypo- theses had been made: 1. Significant differences were found within SS between perceptions of mother, father, and self (p < . 001). That is, it appears that SS tended to rate mothers and themselves higher than their fathers on both agency and communion attitudes and behaviors (see Appendix C). 2. Significant differences were found within Ss between total agency and total communion scores (p < . 001). That is, Ss tended to rate themselves and parents as being more communal than agentic (see Appendix C). 3. An interactive effect was found between the agency/communion variable and the person being rated (i. e. , self, mother, or father, p < . 001). While for all three groups communal ratings were higher than agentic ratings, differences may be noted in comparing them (e. g. , maternal communion scores were higher than self communion, but self agency scores were higher than maternal agency as shown in Appendix C). 4. Significant differences were found within TSCS groups on discrepancy scores (p < . 10). That is, the middle two esteem groups appeared to have a lower discrepancy score than either the high or low groups (see Appendix C). 6'8 5. Significant differences were found between AWS groups on discrepancy scores (p < . 05). That is, the low AWS group tended to have a higher discrepancy score, while the middle-high group had a lower discrepancy score (see Appendix C). In summary, only the first hypothesis, that Ss with high esteem would perceive their parents as being more communal, was tentatively supported. No other hypotheses were supported. In addition, however, other significant differences were found between Ss' perceptions of their mothers, fathers, and themselves (they tended to rate mothers and themselves higher on both agency and communion variables). Though Ss rated both themselves and their parents as more communal than agentic, some differences were found in the relative positioning of such scores as maternal communion and self communion, in comparison to maternal agency and self agency. Finally, the middle esteem groups and medium-high AWS group tended to perceive themselves as having more equal amounts of agency and communion traits, while both the high and low esteem groups and the low AWS groups perceived themselves as having less equal amounts (i. e. , being more communal). gamer CHAPTER IV Discussion The results of the data analysis suggest that communal attitudes and behaviors may tend to foster the growth of self-esteem, though the low value of eta2 (. 007) indicates that only a very small part of the variance of scores can be attributed to the communion variables. The issue of the antecedents of the development of contemporary or traditional attitudes towards women cannot be further explicated by the results of this study. While it seems puzzling that in fact higher agency scores were not related to higher AWS scores, one interesting result which suggests the importance of maintaining an awareness of the potential difference between expressed attitudes and actual behaviors, may be seen in the Appendix concerning the differences between Ss with housewife mothers and Ss with working mothers. While an ex post facto t- test of differences between independent means failed to show significant esteem differences, a significant difference in AWS scores was found (Ss with working mothers scoring higher, p < . 05). While such an analysis "after the fact" should be viewed with caution, the results do suggest the possibility that a working role model may be important in the development of agentic attitudes and behaviors. These results are consistent with other studies regarding the impact of working mothers on their daughters' achievement orientations (Donelson and Gullahorn, in press). Another possible contributing 69 70 factor in the non- significant results might also have been the relative narrowness of the sample (1. e. , college women). The relatively high mean in comparison with previous studies indicates either a progressive societal change in women's attitudes, or that the particular sample was already more "agentic" than women in the general population. While the foregoing interpretations of the results seem important, a possibly more crucial factor may lie in the results concerning the "discrepancy" scores. It may be recalled that the middle TSCS groups and the middle-high AWS group all had significantly lgpv_e_r_ discrepancy scores than the more extreme groups. (The low eta2 values, however, require these interpretations to be viewed with caution. ) Since maturity requires an integration of agency and communion, it might be hypothesized that optimal adjustment may not be represented by extreme scores on the AWS, nor by extremely high TSCS scores (particularly since a lack of self criticism or the deliberate denial of negative aspects of self were not entirely controlled by use of the total P score). Rather, optimal "adjustment" may require a balancing or modulation of agentic/communal traits and behaviors, most successfully achieved by the more moderate AWS- TSCS groups. It appears important further to recognize that a self- report on Ss' own traits is a somewhat different task than a retrospective report of parental attitudes and behaviors. While the validity of 7'1 self— reports should be viewed with appropriate caution, the use of retrospective data, while frequently very useful in exploratory studies such as the present one, does in fact lend itself to certain problems. The fact that Ss rated both themselves and their mothers higher on both agency and communion than their fathers my indicate either the greater saliency of themselves and the mother in the childhood years, or a memory bias favoring the mother because of a greater amount of contact with her. The data also give some indication that "response sets" came into play: all SS rated themselves, mother and father higher on communion than agency, and significant correlations were found between maternal and paternal communion, maternal and paternal agency, and maternal and paternal encouragement to go to college. The response sets might involve a concern with social desirability. For example, it seems possible that some of the agentic adjectives "critical, " "dominating," and self-centered" might have been considered socially undesirable and hence were less frequently acknowledged. A final obvious difficulty which no doubt contributed to the large number of insignificant results was the generally low internal consis- tencies of the parental agency and communion scales. While these values would be increased by increasing the number of items in each scale, the present values suggest in part that the scales were 72 tapping more than one underlying variable. This possibility presents an interesting starting point for future research endeavors. Suggestions for Future Research The low internal consistency of the parental agency and communion scales suggests that future research might first require a cluster analysis of the scale items, which might then be used to form more unidimensional scales. (Because of their currently small length, efforts should also be made to expand the scales, perhaps by using initially all 91 items from the original CRPR in the cluster analysis. ) These new scales might then be studied in relation to the present TSCS-AWS groups. In addition, the previously mentioned problems with the TSCS (i. e. , "extreme" scorers) might be partially resolved by either throwing out Ss with very low Self Criticism scales, p_r_ by putting these Ss in separate groupings (e. g. , high esteem - low SC, high AWS) for a further analysis. That is, the S with a high TSCS score but with low SC might well be a very different woman from the S with an equally high TSCS score but a higher SC. Further, it might prove useful to break down TSCS scores into the various sub-scales mentioned earlier (e. g. , S's social self, family self, personal self, etc. ). The current research in part assumed self- esteem to be a unidimensional quality whereas clearly Ss may feel more positive about certain aspects of themselves than about others. Such sub-scale scores might thus prove more revealing 73 of how certain parental attitudes and behaviors promote or inhibit certain kin_d_s_ of self-esteem. In further research it would also be important to make efforts to control response sets and social desirability factors. The latter might be dealt with at least in part by asking male and female Ss first to rate the adjectives (on a Likert scale) as being more desirable either for a man or a woman. Items judged to be significantly more desirable for a man or woman by raters of both sexes might then be placed in the agentic or communal categories, respectively. Social desirability would then be controlled by a comparison of Ss' ratings on these agency and communion scales with their ratings of "neutral" items (1. e. , items judged no more desirable for men than for women) which are also considered to be positive qualities (see also Bem, 1974). Many of the other response sets referred to earlier suggest the desirability of not using retro- spective data because of the many memory distortions which tend to occur. While it may be argued that the memory distortions themselves, whether factually incorrect or not, are psychodynamically significant, it seems desriable to attempt some corroborative informa- tion gathering. This might be done by attempting to obtain ratings on the CPRP from parents (an admittedly difficult task!), throwing out those SS for which wide discrepancies between parent and child ratings exist (or, better, by analyzing these separately), or perhaps by using a younger age sample in which the children would still be in 74- the home. Here one might also be able to obtain actual behavioral observations, in addition to ratings on attitudinal questionnaires. In attempting to explicate the correlates and antecedents of self-esteem and women's attitudes toward women, this exploratory study has used a fairly narrow sampling range (i. e. , college women), which clearly should be expanded in future research. While the partially retrospective method employed here is one way of beginning to generate hypotheses, which then must be investigated with younger samples, another approach is the longitudinal one, in which young children and their parents would be followed from pre-school to adulthood. This approach, while a more costly and time-consuming one might prove especially valuable in future research endeavors. Summagy Current research has called attention to sex differences in personality and socialization processes, perhaps most succinctly labeled in terms of the agency/communion dichotomy. That is, the stereOtypic American woman tends to be more passive, receptive, interpersonally dependent, sensitive, and non- achieving, while the stereOtypic American man tends to have the opposite agentic traits (e. g. , activity, independence). The investigation of these differences has led to an exploration of childrearing practices and socialization processes, which appear to confirm arguments that women may generally be reinforced more for communal attitudes and behaviors by both parents, though the research evidence is not conclusive as 75 to precise maternal and paternal roles. Nonetheless, it appears that a warm, accepting attitude by the parents, perhaps particularly by the mother, tends to promote stereOtypic femininity (while these behaviors also appear to promote high self-esteem), and some research has even suggested that some degree of "non-acceptance" or, perhaps more, reinforcement for agentic behaviors has more typically been stressed, it is argued, in the socialization of boys, but may apply also to those women who do in fact achieve in our society. While some previous research has suggested both that achieving or "non-feminine" women are maladjusted 329. that the feminine, non-achieving woman is neurotic, the current study attempted to present a somewhat more complex point of view in arguing that either traditional or contemporary female attitudes and life-style may result from high p_r_' low self-esteem. Supportive of this point of view was the Loevinger model of ego development which suggests that the mature individual, whether male or female, must somehow integrate agentic and communal qualities in whatever life-style he chooses, rather than remaining at a societally stereOtypic level. For the woman in American society, this appears to mean integrating agentic traits with (typically more well-developed) communal ones. The current study thus attempted a delineation of female "types" (women with high esteem and "traditional" attitudes towards women; women with high esteem and "contemporary" attitudes; women with 76 low esteem and traditional attitudes; and women with low esteem and contemporary attitudes) and a concurrent investigation of these women's perceptions of themselves in terms of communal/agentic qualities and of their parents' childrearing attitudes and behaviors. It was hypothesized that all women with high esteem (based on scores on the Tennessee Self Concept Scale) would perceive their parents as demonstrating communal qualities (based on scores on communal items on the Child- Rearing Practices Report), while women with contemporary attitudes towards women (as assessed by the Attitudes Towards Women Scale) would perceive their parents as demonstrating agentic attitudes and behaviors towards them. Further, because of the earlier research suggestion that agency might be promoted by less "acceptance, " it was hypothesized that within the high esteem groups, women with contemporary attitudes would perceive their mothers as being less communal than their fathers, while the reverse would hold for the high-esteem traditional college women. Finally, it was hypothesized that high-esteem women with contemporary attitudes would perceive themselves as the most equally agentic and communal (based on a Likert scale adjective checklist of ten agentic and ten communal items), while the low esteem-contemporary group would perceive themselves as most agentic and the high esteem-traditional group would perceive themselves as the most communal. I first" 77, After ascertaining that the AWS and TSCS were not significantly correlated, a 4 X 4 analysis of variance design was used in which four levels of both of the two variables were used. To ease compu- tation and insure at least approximate fulfillment of ANOVA assumptions (e. g. , equal variance between cells), Ss were randomly discarded to leave 224 (out of 300) Ss, with 14 Ss in each cell group. Only the first hypothesis, that high esteem groups perceived their parents as more communal than low esteem groups, was tentatively supported (p < . 10). In addition, it was found that Ss tended to rate themselves and their mothers higher on p_o_th_ agency and communion traits than their fathers. Ss also rated themselves and both parents as more communal and, finally, the middle esteem groups and medium-high AWS groups tended to perceive themselves as having more equal amounts of agency and communion traits, while the low esteem and low AWS groups perceived themselves as having less balanced amounts (i. e. , were more communal). APPENDIX A Questionnaires and Keys 78 APPENDIX A General Instructions On the following pages you will be asked to complete a variety of objective questionnaire tasks. In most cases, the type of response required will be clear to you. Additional instructions are given when necessary. _Plggsp be sure to answer a_l_l_ items for 311 questionnaires. If you are unsure about an answer, respond with the first choice that comes to your mind. If you have any further questions regarding the completion of these tasks, please ask the experimenter. The experimenter greatly appreciates your willingness to participate in this study and will be glad to share results of the data analysis. Should you be interested in receiving such feedback, please leave your name and mailing address at the bottom of this front sheet, and detach it and give it ot the experimenter at the end of the session. 79 Background Information Age (1 ‘—‘ 17; 2 = 18-19; 3 = 20-21; 4 = 21-22; 5 5 22) Marital Status: (Single, Married, Engaged, Divorced, Widowed) l = single; 2 = married Year in School (freshman, etc. ) l = freshman; 4 = senior Maj or Father's occupation see Agust B. Hollingshead Index of Social Position 9' Mother's occupation H Father' 5 education Mother's education Please record your answers to the following questions on the computer answer sheet directly behind this page. 1. Do you plan to graduate? a) Definitely yes b) Probably c) Don't know d) Probably not e) Definitely not 1 2. What is the highest degree you hope ultimately to earn? a) High school b) Undergraduate (Bachelor's) c) Master's d) Doctoral 5 (Ph. D. MD) 3. How important is a professional career to you? a) Not at all important b) somewhat unimportant c) Don't know (1) Somewhat important e) Very important 5 4. How important is marriage and family life to you? a) Not at all important b) Somewhat unimportant c) Don't know d) Somewhat important e) Very important 5 80 5. How much encouragement to go to college did you receive from your mother? a) Very much b) Some c) Very little d) None 1 6. How much encouragement to go to college did you receive from your father? a) Very much b) Some 0) Very little d) None 1 7. How satisfied do you feel your mother was/ is with her homemaking role? a) Very dissatisfied b) Somewhat dissatisfied c) Don't know (1) Somewhat satisfied e) Very satisfied 1 8. While you were growing up, how much of her time did your mother spend performing the homemaking role? a) Full-time b) Part —t:ime 1 81 Two Factor Index of Social Position August B. Hollingshead A. Educational Codes Code \IOUTI-BODNI-d B. Occupational Codes Code 1 \IO‘Cnt-P- Years of School Completed Professional (MA, MS, ME, MD, PhD) Four year college graduate l to 3 years college High school graduate 10 to 11 years of school 7 to 9 years of school under 7 years of school Higher executives, prOprietors of large concern or major professionals Business managers in large concerns, proprietors of medium-sized businesses, lesser professionals Administrative personnel, owners of small independent businesses, minor professionals and farmers Clerical and sales workers, technicians, owners of little businesses, and farmers Skilled manual employees and farmers Machine operators, semi-skilled employees and farmers Unskilled employees and farmers 82 Tennessee Self Concept Scale William H. Fitts Responses: Partly false Completely Mostly and Mostly false false partly true true 1 2 3 4 19. 21. 23. 37. 39. 41. 55. 57. 59. 73. 75. 77. 91. 93. I have a healthy body I am an attractive person I consider myself a sloppy person I am a decent sort of person I am an honest person I am a bad person I am a cheerful person I am a calm and easy going person I am a nobody Completely true 5 I have a family that would always help me in any kind of trouble I am a member of a happy family My friends have no confidence in me I am a friendly person I am popular with men I am not interested in what other people do I do not always tell the truth I get angry sometimes 5:; VI -nnkfifu.‘ 20. 22. 24. 38 40. 42. 56. 58. 74. 76. 78. 92. 94. 83 I like to look nice and neat all the time I am full of aches and pains I am a sick person I am a religious person I am a moral failure I am a morally weak person I have a lot of self control I am a hateful person I am losing my mind I am an important person to my friends and family I am not loved by my family I feel that my family doesn't trust me I am popular with women I am mad at the whole world I am hard to be friendly with Once in a while I think of things too bad to talk about Sometimes, when I am not feeling well, I am cross 11. 25. 27. 29. 43. 45. 47. 61. 63. 65. 79. 81. 83 95. 97. 84 I am neither too fat nor too thin I like my looks just the way they are I would like to change some parts of my body I am satisfied with my moral behavior I am satisfied with my relationship to God I ought to go to church more I am satisfied to be just what I am I am just as nice as I should be I despise myself I am satisfied with my family relationships I understand my family as well as I should I should trust my family more I am as sociable as I want to be I try to please others, but I don't overdo it I am no good at all from a social standpoint I do not like everyone I know Once in a while, I laugh at a dirty joke 10. 12. 26. 28. 44. 46. 48. 62. 64. 66. 80. 82. 84. 96. 98. 85 I am neither too tall nor too short I don't feel as well as I should I should have more sex appeal I am as religious as I want to be I wish I could be more trustworthy I shouldn't tell so many lies I am as smart as I want to be I am not the person I would like to be I wish I didn't give up as easily as I do I treat my parents as well as I should (Use past tense if parents are not living) I am too sensitive to things my family say I should love my family more I am satisfied with the way I treat other people I should be more polite to others I ought to get along better with Other people I gossip a little at times At times I feel like swearing l 3. 15. 17. 31. 33. 35. 49. 51. 53. 67. 69. 71. 85. 87. 89. 99. 86 I take good care of myself physically I try to be careful about my appearance I often act like I am "all thumbs" I am true to my religion in my everyday life I try to change when I know I'm doing things that are wrong I sometimes do very bad things I can always take care of myself in any situation I take the blame for things without getting mad I do things without thinking about them first I try to play fair with my friends and family I take a real interest in my family I give in to my parents. (Use past tense if parents are not living) I try to understand the other fellow's point of View I get along well with other people I do not forgive Others easily I would rather win than lose in a game fl‘gfi 14. l6. 18. 32. 34. 36. 50. 52. 54. 68. 70. 72. 86. 88. 90. 100. 87 I feel good most of the time I do poorly in sports and games I am a poor sleeper I do what is right most of the time I sometimes use unfair means to get ahead I have trouble doing the things that are right I solve my problems quite easily I change my mind a lot I try to run away from my problems I do my share of work at home I quarrel with my family I do not act like my family thinks I should I see good points in all the people I meet I do not feel at ease with other people I find it hard to talk with strangers Once in a while I put off until tomorrow what I ought to do today AS AS AS DS AS DS DS DS DS 88 Attitudes Toward Women The statements listed below describe attitudes toward the role of women in society which different people have. There are no right or wrong answers, only opinions. You are asked to express your feelings about each statement by indicating whether you (A) Agree strongly, (B) Agree mildly, (C) Disagree mildly or (D) Disagree strongly. Please indicate your opinion by marking the column on the answer sheet which corresponds to the alternative which best describes your personal attitude. Please respond to every item. (A) Agree strongly (B) Agree mildly (C) Disagree mildly (D) Disagree strongly Response keyed Q 1. Women have an obligation to be faithful to their husbands. 2. Swearing and obscenity is more repulsive in the speech of a woman than a man. ‘ 3. The satisfaction of her husband's sexual desires is a fundamental obligation of every wife. 4. Divorced men should help support their children but should not be required to pay alimony if their wives are capable of working. 5. Under ordinary circumstances, men should be expected to pay all the expenses while they're out on a date. 6. Women should take increasing responsibility for leadership in solving the intellectual and social problems of the day. 7. It is all right for wives to have an occasional, casual extramarital affair. 8. Special attentions like standing up for a woman who comes into a room or giving her a seat on a crowded bus are outmoded and should be discontinued. 9. Vocational and professional Schools should admit the best qualified students, independent of sex. DS 10. Both husband and wife should be allowed the same grounds for divorce. AS DS AS DS AS AS DS DS DS DS DS AS DS DS AS AS 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 89 Telling dirty jokes should be mostly a masculine prerogative. Husbands and wives should be equal partners in planning the family budget. Men should continue to show courtesies to women such as holding open the door or helping them on with their coats. Women should claim alimony not as persons incapable of self support but only when there are children to provide for or when the burden of starting life anew after the divorce is obviously heavier for the wife. Intoxication among women is worse than intoxication among men. The initiative in dating should come from the man. Under modern economic conditions with women being active outside the home, men should share in household tasks such as washing dishes and doing the laundry. It is insulting to women to have the "obey" clause remain in the marriage service. There should be a strict merit system in job appointment and promotion without regard to sex. A woman should be as free as a man to propose marriage. Parental authority and responsibility for discipline of the children should be equally divided between husband and wife. Women should worry less about their rights and more about becoming good wives and mothers. Women earning as much as their dates should bear equally the expense when they go out together. Women should assume their rightful place in business and all the professions along with men. A woman should not expect to go to exactly the same places or to have quite the same freedom of action as a man. Sons in a family should be given more encouragement to go to college than daughters. AS AS DS AS AS AS DS DS DS AS AS AS AS DS DS AS 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 90 It is ridiculous for a woman to run a locomotive and for a man to darn socks. It is childish for a woman to assert herself by retaining her maiden name after marriage. Soceity should regard the services rendered by the women workers as valuable as those of men. It is only fair that male workers should receive more pay than women even for identical work. In general, the father should have greater authority than the mother in the bringing up of children. Women should be encouraged not to become sexually intimate with anyone before marriage, even their fiances. Women should demand money for household and personal expenses as a right rather than as a gift. The husband should nOt be favored by law over the wife in the disposal of family property or income. Wifely submission is an outworn virtue. There are some professions and types of businesses that are more suitable for men than women. Women should be concerned with their duties of childrearing and housetending, rather than with desires for professional and business careers. The intellectual leadership of a community should be largely in the hands of men. A wife should make every effort to minimize irritation and inconvenience to the male head of the family. There should be no greater barrier to an unmarried woman having sex with a casual acquaintance than having dinner with him. Economic and social freedom is worth far more to women than acceptance of the ideal of femininity which has been set by men. Women should take the passive role in courtship. AS DS AS AS AS DS DS AS DS AS DS AS 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 91 On the average, women should be regarded as less capable of contribution to economic produCtion than are men. The intellectual equality of woman with man is perfectly obvious. Women should have full control of their persons and give or withhold sex intimacy as they choose. The husband has in general no obligation to inform his wife of his financial plans. There are many jobs in which men should be given preference over women in being hired or promoted. Women with children should not work outside the home if they don't have to financially. Women should be given equal opportunity with men for appren- ticeship in the various trades. The relative amounts of time and energy to be devoted to household duties on the one hand and to a career on the other should be determined by personal desires and interests rather than by sex. As head of the household, the husband should have more responsibility for the family's financial plans than his wife. If bOth husband and wife agree that sexual fidelity isn't important, there's no reason why bOth shouldn't have extramarital affairs if they want to. The husband should be regarded as the legal representative of the family group in all matters of law. The modern girl is entitled to the same freedom from regulations and control that is given to the modern boy. Most women need and want the kind of protection and support that men have traditionally given them. 92 Child Rearing Practice Report Please record your answers to the following questions on the computer answer sheet directly behind this questionnaire. Indicate whether you MC MC MA PA MC MC PA MA PA MA MA MA 4 a) agree strongly MC: maternal communion b) agree mildly MA: maternal agency c) don't know or are unsure PC= paternal communion d) disagree mildly PA= paternal agency e) disagree strongly My mother felt I should be given comfort when upset. My mother made sure she knew where I was. My mother made sure I knew she appreciated my efforts. 4. My father encouraged me always to do my best. 9‘ 10. 11. 12. 13. 14. 15. 16. 17. My mother would not go out if she had to leave me with a stranger. My father felt I should be given comfort when upset. My mOther and I had warm intimate times together. My father felt I should have time to think, daydream and to loaf sometimes. . My father taught me to control my feelings at all times. My father expressed affection by holding, kissing, and hugging me. My mOther felt it was good for me to play competitive games. My father would not go out if he had to leave me with a stranger. My father did not allow me to question his decisions. My mother sometimes teased and made fun of me. My father took my preferences into account in making family plans. My mother did not allow me to question her decisions. My mother taught me to control my feelings at all times. ,‘n 022.7. . MA PA MC MC PA PA MC MA MC PA PA MC MA PA 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 93 My father believed praise gOt better results than punishment. My mOther encouraged me always to do my best. My mother did not allow me to say bad things about my teacher. My father felt physical punishment was the best method of discipline. My mother used to punish me by putting me off somewhere by myself for a while. My mother took my preferences into account in making family plans. My mother expressed affection by holding, kissing, and hugging me. My father believed scolding and criticism helped me improve. My father had strict rules for me. My mother encouraged me to wonder and think about life. My mother expected me to be grateful and appreciate my advantages. My mother was easygoing and relaxed with me. My father wanted me to make a good impression on others. My father felt it was good for me to play competitive games. My father sometimes teased and made fun of me. My father encouraged me to talk about my troubles. My mother placed my father's wishes ahead of those of her children. My father encouraged me to wonder and think about life. My mOther encouraged me to talk about my troubles. My mother felt physical punishment was the best method of discipline. My father threatened punishment more often than he actually gave it. My father and I had warm, intimate times together. My father expected me to be grateful and appreciate my advantages. PA MA MC MC PA MA MA PA 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 94- My mother believed in starting toilet training as early as possible. My father let me make many decisions myself. My father found it difficult to punish me. My father made sure I knew he appreciated my efforts. My mother had strict rules for me. My mother worried about the bad and sad things that could happen to me as I grew up. My father often felt angry with me. My mother found it difficult to punish me. My mother encouraged me to be independent of her. My mother believed praise got better results than punishment. My father used to tell me how ashamed and disappointed he felt when I misbehaved. My mother was too wrapped up in her children. My father was easygoing and relaxed with me. My father thought I should be encouraged to do better than others. My father was reluctant to see me grow up. My mother believed scolding and criticism helped me improve. My father made sure he knew where I was. My mother did not believe children should have secrets from their parents. My mother thought I should be encouraged to do better than others. My father respected my opinions and encouraged me to express them. My father did not allow me to say bad things about my teacher. >OO>O>>>O>O>OO>>OOO> 95 Adjective Checklist Still using the same answer sheet, please rate how characteristic you feel each of the following qualities is of you, according to the following scale: 0 a) b) C) d) 4 e) 9. 10. l 1. 12. 1 3. 14. 15. 16. 17. 18 . 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. not at all characteristic of me C = communion item characteristic only to a minimum degree A = agency item characteristic to a moderate degree characteristic to a relatively large degree extremely characteristic of me Assertive Altruistic Helpful Considerate Practical, Shrewd Ambitious Sensitive Loving, Affectionate Rational, Reasonable Sympathetic Critical Responsive Independent Dominating Self Centered Idealistic Competitive Artistic Generous Adventurous APPENDIX B General Information for all SubjeCts (n = 300) 96 APPENDIX B General Information for All Subjects (n = 300) Variable X S. D. Range Internal Cons. Age 2. 25 . 66 17 to 27 (n=1) Marital Status 1. 01 . 13 Year in School 1. 47 . 82 Fresh. to Sr. Father' s Occ. 2. 91 l. 55 Mother's Occ. 3. 37* l. 27 F ather' 3 Ed. 2. 79 l. 30 MOther' 3 Ed. 3. 21 l. 05 Objective 1 3. 71 . 63 Questions2 l. 80 . 77 3 3. 48 . 78 4 3. 34 . 94 5 2. 53 . 74 6 2. 45 . 93 7 2. 78 1. 32 8 . 28 . 45 TSCS 343. 49 29. 61 217 to 431 AWS 1215.63 20. 50 62 to 160 . 92 Communion-Self 27. 06(2. 71)** 4. 19 13 to 38 . 64 Agency—Self 23. 21(2. 32)** 4. 75 5 to 35 . 65 Discrepancy 3. 84( . 19)** 6. 15 +28 to -17 . 59 Communion-Maternal 31. 08(2. 82)** 7. 17 0 to 44 . 79 Agency-Maternal 26. 79(2. 23)** 5. 34 12 to 43 . 46 Communion-Paternal 27. 18(2. 47)** 8. 50 1 to 44 . 83 Agency-Paternal 27. 25(2. 27 )** 5. 79 12 to 43 . 53 *Scores are based on _S_s with working mothers (n = 141) only ** Mean score per item APPENDIX C Means and Standard Deviations for AWS-TSCS Groups on Agency and Communion Scales Nauru In... Ivar. 97, APPENDD( C L Means (Per Item) and Standard Deviations on Agency-Communion Scores for All TSCS Groups SC SA MC MA PC PA D LO TKES f 313 2. 63 2. 16 2. 68 2. 23 2. 38 2. 35 . 46 (n = 56) (. 37)* (. 57) (. 68) (. 49) (. 75) (. 44) (. 77) MED-LO T338 314-343 2. 66 2. 33 2. 81 2. 27 2. 54 2. 31 . 32 (n = 56) (. 45) (. 48) (. 61) (. 44) (. 75) (. 48) (. 67) MED-HI TSCS 344-373 2. 66 2. 36 2. 91 2. 20 2. 41 2. 27 . 29 (it: 56) (. 38) (. 44) (. 51) (. 45) (. 79) (. 49) (. 54) HI TSCS 9- 374 2. 89 2. 44 3. 01 2. 18 2. 57 2. 19 . 45 (n = 56) (. 41) (. 36) (. 60) (. 48) (. 79) (. 50) (. 49) *Standard Deviation PC =Paternal Communion PA =Paternal Agency MC =Maternal C ommunion MA=Mate rnal Agency D=Discrepancy (SC - SA) SC =Self C ommunion SA=Self Agency I‘M-ZIP A? _"J! ”JR 98 11. Means (Per Item) and Standard Deviations on Agency—C ommunion Scores for All AWS Groups SC SA MC MA PC PA D LOW AWS ‘—‘ 95 2. 77 2. 24 2. 92 2. 31 2. 58 2. 21 . 53 (n = 56) (. 36) (. 45) (. 51) (. 39) (. 79) (. 47) (. 58) MED-LO AWS 95-115 2. 71 2. 35 2. 85 2. 29 2. 49 2. 33 . 36 (n = 56) (. 44) (. 44) (. 63) (. 47) (. 67) (. 44) (. 59) MED-HI AWS 116-136 2. 64 2. 36 2. 83 2. 11 2. 33 2. 31 . 28 (n = 56) (. 39) ( 49) (. 60) (. 51) (. 76) (. 50) (. 61) HI AWS 9 137 2. 71 2. 34 2. 83 2. 17 2. 51 2. 26 . 37 (n = 56) (. 46) (. 52) (. 71) (. 47) (. 85) (. 50) (. 70) \r’ 99' III Means (Per Item) and Standard Deviations on Agency-C ommunion Scores for All Subjects and For "Extreme" Groups SC SA MC MA PC PA D ALL SS 2. 71 2. 32 2. 85 2. 22 2. 48 2. 28 . 38 (n = 224) (. 41)* (. 48) (. 61) (. 47) (. 77) (. 48) (. 63) All Com All Agency Self Maternal Paternal 8. 04 6. 82 5. 03 5. 08 4. 75 (1. 31) (1. 03) (. 64) (. 73) (. 87) HI TKJS - HI AWS 2. 92 2. 54 2.95 2. 24 2. 62 2. 12 . 39 (n = 14) (. 37) (. 42) (. 78) (. 46) (. 87) (. 50) (. 48) HI TSCS - LO AWS 2. 84 2. 30 3. 14 2. 34 2. 97 2. 24 . 54 (n = 14) (. 40) (. 32) (. 38) (. 47) (. 50) (. 49) (. 43) L0 TSCS - HT AWS 2. 67 2. 00 2. 74 2. 10 2. 43 2. 30 . 67 (n = 14) (. 38) (. 63) (. 63) (. 68) (. 85) (. 55) (. 79) L0 TSCS - LO AWS 2. 74 2. 11 2. 69 2. 38 2. 54 2. 30 . 63 (n = 14) (. 37) (. 57) (. 72) (. 35) (. 69) (. 33) (. 74) *Standard Deviation 100 N. Mean (Per Item) Discrepancy Scores (C -A) For All AWS-TESS Groups Ali/.8 LO MED-LO MED-HI H1 H1 .54 .56 . 31 . 39 (. 43)* (. 57) (. 46) (. 48) MED-HI .45 .41 .08 .24 g (. 69) (. 49) (.43) (. 47) MED-LO .49 .26 .36 .19 (. 48) (. 64) (. 60) (.92) L0 .63 . 18 . 36 .67 (.74) (. 66) (. 88) (.79) *Standard Deviation than; .lvd. 101 V. Means for all AWS-TSCS Groups for Self, Maternal and Paternal Communion and Agency Scores* AWS Lo Med-Lo Med-Hi Hi Hi 2. 84 2. 30 2. 91 2. 35 2. 89 2. 58 2. 92 2. 54 3. 14 2. 34 2. 87 2. 20 3. 08 1. 93 2. 95 2. 24 2. 97 2. 34 2. 36 2. 23 2. 32 2. 15 2. 62 2. 12 MED-HI 2. 78 2. 33 2. 79 2. 38 2. 44 2. 36 2. 61 2. 37 2. 86 2. 15 3. 05 2. 26 2. 84 2. l4 2. 88 2. 25 § 2. 06 2. 14 2. 75 2. l9 2. 29 2. 30 2. 51 2. 43 ‘" MED-LO 2. 74 2. 24 2. 61 2. 35 2. 64 2. 27 2. 64 2. 44 2. 97 2. 37 2. 79 2. 29 2. 76 2. 32 2. 74 2. 11 2. 73 2. 18 2. 46 2. 40 2. 51 2. 47 2. 48 2. 17 LOW 2. 74 2. 11 2. 51 2. 34 2. 58 2. 21 2. 67 2. 00 2. 69 2. 38 2. 67 2. 40 2. 63 2. 06 2. 74 2. 09 2. 54 2. 30 2. 38 2. 51 2. 19 2. 20 2. 43 2. 30 *Order of Presentation: Self Communion Self Agency Maternal Communion Paternal Communion Maternal Agency Paternal Agency APPENDIX D Subjects with Housewife vs. Working Mothers 102 APPENDIX D Subjects with Housewife vs. Working Mothers Working Mothers Housewife Mothers Total Sample (n = 141) (n = 159) (n = 300) x S. D. x S. D. 2 S. p._ 1. TSCS 346. l 30. 7 341. 1 28. 5 343. 5 29. 6 2. AWS 118. 2 l9. 7 113. 3 20. 9 115. 6 20. 5 3. SC 26.9 4.2 27.2 4.1 27.1 4.2 4. SA 23. 1 5. l 23. 3 4. 4 23. 2 4. 7 5. SD 43. 9 6. 1 43. 8 6. 1 43. 8 6. 1 6. MC 31. 4 7. 0 30. 8 7. 3 31. 1 7. 2 7. MA 26. 5 5. 6 27. 0 5. 1 26. 8 5. 3 8. PC 27. 7 8. 3 26. 7 8. 6 27. 2 8. 5 9. PA 26. 8 5. 6 27. 7 6. 0 27. 2 5. 8 17. age 2.3 .7 2.2 .6 2.2 .6 18. M8 1.0 .1 1.0 .1 1.0 .1 19. Yr in School 1. 5 . 8 4. 1 8 1. 5 . 8 20. FsOc 3. 0 1. 6 2. 8 l 4 2. 9 1. 5 21. MsOc 3. 4 1. 3 - - - - 22. FsEd 2. 8 1. 2 2. 8 1. 3 2. 8 1. 3 23. MsEd 3. 0 1. 1 3. 4 . 9 3. 2 1. 0 F- test for Independent Means Esteem: 346. 121 - 341. 151 = 4. 970 S* 215:2 = 3. 419 t =§_Z_‘91_}g_= 1. 45 n. s. AWS: 118. 206 - 113. 352 = 4. 854 S £1-22 = 2. 3546 t =§L§§§5= 2. 0615 p < .05 *Standard error of difference between means \.. Ala. MU-lr. y. 1E ,. r If? R EFER ENCES .. 185%.»? R EFER ENCES Ainsworth, M. D. Infancy in Uganda: infant care and the growth of love. Baltimore, Marfland: Johns Hopkins University Press, 1967. Almquist, E. M. and Angrist, S. S. Career Salience and Atypicality of Occupational Choice Among College Women. Journal of Marriage and the Family, 1970, SS, 242-249. Anastasi, A. Differential psychology: individual and group differences in behavior (310 editiofi). New York: MacMillan, 1958. Angyal, A. A theoretical model of personality studies. Journal of Persmality, 1951, EU), 131-142. Bakan, D. The duality of human experience. Chicago: Rand McNally, 1966. Bardwick, J. M. Psychology of Women: A study_’ of biocultural conflicts. New York: Harper and Row, 1971. Baumrind, D. "Current patterns of parental authority. Develop- mental Psychology Monographs, 1971, 1, (Pt. 2) Bern, Sandra L. The measurement of psychological androgyny. Journal of Consulting and Clinical Psychology, 1974, 42, 2, 155-162. Berliner, Sharon. Attitudes toward women, perceptions of parents, future goals and achievement/ interpersonal needs in college women. Unpublished study, 1973. Block, Jack, Vonderlippe, Anna, and Block, Jeanne H. Sex-role and socialization patterns: Some personality concomitants and environmental antecedents. Journal of Consulting and Clinical Psychology, 1973, ii (3), 32I-IEI. 103 104 Block, Jeanne H. Conceptions of sex-role: some cross-cultural and longitudinal perspectives. American Psychologist, June, 1973, 512-526. Block, J. H. The child-rearing practices report. Berkeley: Institute of Human Development, University of California, Berkeley, 1965. (Mimeo) Block, Jeanne H. The child-rearing practices report (CRPR) A set of Q items for the Description of Parental Socialization AttitudEs and Values. Berkeley: Institute of Human Develop- ment, Berkeley, 1964. (Mimeo) Bronfenbrenner, V. The changing American child: A speculative analysis. Journal of Social Issues, 1961, 17 (1), 6-18. Bronfenbrenner, V. Freudian theories of identification and their derivatives. Child Development, 1960, S1, 15-40. Coopersmith, S. The antecedents_of self-esteem. San Francisco: W. H. Freeman and Co. , 1967. Crandall, V. , Dewey, R. , Katowsky, W. and Preston, A. Parents' attitudes and behaviors and grade-school children's academic achievements. Journal of Genetic Psychology, 1964, 194, 53-66. Crandall, V. , Preston, A. , and Rabson, A. Maternal reactions and the development of independence and achievement behavior in young children. Child Development, 1960, S1, 243-251. Donelson, E. "Sense of Separateness and Social Responsiveness" in Donelson, E. and Gullahorn, J. (Eds.), Women: A Psycho- 1% ical Perspective, to be published. Donelson, E. and Gullahorn, J. Women: A Psychological Perspec- Eiy_e. John Wiley, in press. Douvan, E. and Adelson, J. The adolescent experience. New York: Wiley, 1966. Dubin, R. and Dubin, R. Children's social perceptions. A review of research. Child Development, 1965, 36(3), 809-838. Edwards, A. L. Edwards, Personal Preference Schedule, Psycho- logical Corporation, 1959. .r. .533»? 105 Ellis, E. Social Psychological Correlates of Upward Social Mobility of Unmarried Career Women. American Sociological Review, 1952, l_7, 558-63. Emmerich, W. and Smoller, F. "The role paterning of parental norms." Sociometry, 1964, 2_7, 382-390. Fitts, William H. Tennessee Self Concept Scale (Manual). Nash- ville, Tennessee: CounseBr Recordings and Tests, 1965. Ferguson, L. R. Personality Development. Belmont, California: Brooks /Cole, 1970. Fish, K. D. Paternal availability, family role structure, maternal employment and personality development in late adolescent females. Unpublished doctoral dissertation, University of Massachusetts, 1969. Fisher, S. The Female Orgasm: Psychology, physiology , fantasy. New York: Basic BOOTS, Inc. , I973. French, E. G. and Lesser, G. S. Some characteristics of the achievement motive in women. Journal of Abnormal and Social Psychology, 1964, 119-128. Freud, A. The ego and the mechanisms of defense. New York: International Universities Press, F946. Freud, S. New Introductory Lectures in Psychoanalysis. New York: Norton, 1933. Freud, S. The problem of anxiety. New York: W. W. Norton and Company, 1936. Gividen, G. M. Stress in airborne training as related to the self concept, motivation and biographical factors. Unpublished Master's Thesis. Vanderbilt University, 1959. Goodenough, E. W. "Interest in persons as an aspect of sex difference in the early years. " Genetic Psychology Monographs, 1957, _5_5_, 287-323. Gray, S. W. Perceived similarity to parents and adjustment. Child Development, 1959, 3_0_, 91-107. 106 Guttman, D. L. "Women and the conception of ego strength. " Merrill-Palmer Quarterly, 1965, E (3), 229-240. Hall, J. D. An investigation of acquiescence response set, extra- version, and locus of control as related to neuroticism and maladjustment. Unpublished doctoral dissertation, George Peabody College, 1964. Heilbrun, A. B. Conformity to masculine-feminine stereotypes and ego identity in adolescents. Psychological Reports, 1964, 14, 351-357. Heilbrun, A. B. Sex role instrumental-expressive behavior, and psychopathology in females. Journal of Abnormal Psychology, 1968, 7_3_(2), 131-136. Heilbrun, A. B. and Fromme, D. K. Parental identification of late adolescents and level of adjustment: The importance of parent- model attributes, ordinal position, and sex of the child. Journal of Genetic Psychology, 1965, 101, 49-59. Hoffman, L. W. "Early Childhood Experiences and Women's Achievement Motives," Journal of Social Issues, 28, 1972, 129-155. Hoffman, M. L. Moral development. In P. H. Mussen (Ed.) Carmichael's manual of child psycholo y (Vol. II). New York: J—ohn Wiley and Sons, Inc. , T970, 261500. Horner, M. S. Fail: Bright women. Psychology Today, 1969, 3 (6), 36. _ Horney, K. The neurotic personality of our time. New York: .—‘ W. W. Norton and Company, Inc. , 1937. Kagan, J. "Acquisition and significance of sex typing and sex role identity" in M. L. Hoffman and L. W. Hoffman (Eds.) Review of Child Development Research, 1966, Vol. 1. New YorE: Russell Sage Foundation, 137-167. Kagan, J. and Freeman, M. Relation of childhood intelligence, maternal behavior and social class to behavior during adolescence. Child Development, 1963, _3_4_, 899-911. Kagan, J. and Moss, H. A. Birth to Maturity: A study in psycho- logical development. New York: Wiley and Sons, T962. 107 Komarovsky, M. Learning the feminine role. In E. M. Schur (Ed.) The family and the sexual revolution. Bloom ington: Indiana University press, 1964, 213-223. Lecky, P. Self-consistency: A theory of personality. New York: Island Press, 1945. Lipman-Blumen, J. How Ideology Shapes Women's Lives. Scientific American, 1972, 2_26_, 34-42. Loevinger, Jane and Wessler, Ruth. Measuring ego development (Vol. 1). San Francisco: Jossey-Bass, Inc. , 1970. Lynn, D. B. The process of learning parental and sex-role identifi- cation. Journal of Marriage and the Family, 1966, _2_8 (4), 466 -470. Lynn, D. B. Sex-role and parental identification. Child Development, 1962, _3_3, 555-564. Mowrer, O. H. Learning theory and personality dynamic. New York: Ronald, 1950. Murphy, E. B. , Silber, E. , Coelho, G. V. , Hamburg, D. A. , and Greenberg, 1. Development of autonomy and parent-child inter- action in late adolescence. American Journal of Orthopsychiatry, 1963, 33, 643-652. Mussen, P. and Rutherford, B. Parent-child relations and parental personality in relation to young children's sex-role preferences. Child Development, 1963, S4, 589-607. Preston, G. A. Parental role perceptions and identification in adolescent girls. Unpublished doctoral dissertation, University of Michigan, 1965. Rossi, A. S. "Equality between the sexes: An immodest proposal." Daedalus, 1964, 93, 607-652. Sears, R. R. "Relation of early socialization experiences to self concepts and gender role in middle childhood," Child Develop- ment, 1970, ill, 267-289. Sears, R. R. , Maccoby, E. E. , and Levin, H. Patterns of child rearing. New York: Row Peterson and Co. , I957. 108 Sears, R. R. , Rau, L. , and Alpert, R. Identification and child rearing. Palo Alto, California: Stanford University Press, Shaw, M. C. and Dutton, B. E. The use of the parent attitude research inventory with parents of bright academic underachievers. Journal of Educational Psychology, 1962, 53, 203-208. Sherman, Julia A. "Problem of Sex Differences in Space Perception and Aspects of Intellectual Functioning," Psychological Review, 1967, Vol. 74, No. 4, 290-299. Slater, P. E. Psychological factors in role specialization. Harvard University. Ph. D. Dissertation, 1955, unpublished. Slater, P. E. Toward a dualistic theory of identification. Merrill- Palmer Quarterly, 1961, 7 (2), 113-126. Slote, G. M. Feminine character and patterns of interpersonal perception. Unpublished doctoral dissertation, New York University, 1962. Sontag, L. W. , Baker, C. T. and Nelson, V. A. Mental growth and personality development: A longitudinal study. Mon ra hs of the Society for Research in Child Development, 1958, 23(65). Sopchack, A. Parental 'identification' and tendency toward disorder as measured by the MMPI. Journal of Abnormal and Social Psychology, 1952, 41, 159-165. Spence, Janet T. and Helmreich, Robert. "The Attitudes Toward Women Scale: An Objective Instrument to Measure Attitudes Toward the Rights and Roles of Women in Comtemporary Society" Journal Supplement Abstract Service, Selected Documents in Psychology, American Psychological Association, 1973. Sullivan, H. S. Conceptions of modern psychiatry. Washington: William Alanson White Psychiatric Foundation, 1947. Sundby, E. S. "A Study of Personality and Social Variables related to Conformity Behavior. Unpublished doctoral dissertation, Vanderbilt University , 1962. Tyler, F. B. , Rafferty, J. , and Tyler, B. Relationships among motivations of parents and their children. Journal of Genetic Psychology, 1962, El, 69-81. 109 Tyler, L. E. The psychology of human differences. New York: Appleton -Century -Croft , 1965 . Veroff, J. Social comparison and the development of achievement motivation. In Smith, C. (Ed.) Achievement related motives in children. New York: Russell Sage Foundation, 1969, 46-101. Webb. A. P. Sex-role preferences and adjustment in early adoles- cence. Child Development, 1963, S1, 609-618. Witkin, H. A. Psychological differentiation and forms of pathology. Journal of Abnormal Psycholog_y, 1965, 79 (5), 317-336. Witkin, H. A. Social Influences in the development of cognitive style. In D. A. Goslin (Ed.) Handbook of Socialization Theory and Research. Skokie, Ill. : Rand McNally and Co. , 1969. Witkin, H. A. Goodenough, D. R. , and Karp, S. A. Stability of cognitive style from childhood to young adulthood. Eurnal of Personality and Social Psychology, 1967, _7 (3), 291-300. Witkin, H. A. , Lewis, H. B. , Hertzman, M. , Meissner, P. , Machover, K. and Wapner, S. Persgnality through perception. New York: Harper and Brothers, 1954. Wylie, R. The self-concept. Lincoln, Nebraska: University of Nebraska Press, 1961. "IIIIIIIII‘IIIIIIIIIT