IDENTITY PROCESSING STYLE AND SEX-ROLE IDENTITY PREDICT ATTACHMENT STYLE AND PSYCHOSOCIAL BALANCE by Michael M. Pratt B.Sc., University of Northern British Columbia, 1998 THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in PSYCHOLOGY THE UNIVERISTY OF NORTHERN BRITISH COLUMBIA APRIL 2004 © Michael M. Pratt, 2004 1^1 Library and Archives Canada Bibliothèque et Archives Canada Published Heritage Branch Direction du Patrimoine de l'édition 395 W ellington Street Ottawa ON K 1A 0N 4 Canada 395, rue W ellington Ottawa ON K 1A 0N 4 Canada Your file Votre référence ISBN: 0-494-04667-8 Our file Notre référence ISBN: 0-494-04667-8 NOTICE: The author has granted a non­ exclusive license allowing Library and Archives Canada to reproduce, publish, archive, preserve, conserve, communicate to the public by telecommunication or on the Internet, loan, distribute and sell theses worldwide, for commercial or non­ commercial purposes, in microform, paper, electronic and/or any other formats. AVIS: L'auteur a accordé une licence non exclusive permettant à la Bibliothèque et Archives Canada de reproduire, publier, archiver, sauvegarder, conserver, transmettre au public par télécommunication ou par l'Internet, prêter, distribuer et vendre des thèses partout dans le monde, à des fins commerciales ou autres, sur support microforme, papier, électronique et/ou autres formats. The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. L'auteur conserve la propriété du droit d'auteur et des droits moraux qui protège cette thèse. Ni la thèse ni des extraits substantiels de celle-ci ne doivent être imprimés ou autrement reproduits sans son autorisation. In compliance with the Canadian Privacy Act some supporting forms may have been removed from this thesis. Conformément à la loi canadienne sur la protection de la vie privée, quelques formulaires secondaires ont été enlevés de cette thèse. While these forms may be included in the document page count, their removal does not represent any loss of content from the thesis. Bien que ces formulaires aient inclus dans la pagination, il n'y aura aucun contenu manquant. Canada ABSTRACT Differences in relationship style scores (secure, fearful, preoccupied, dismissing) and psychosocial balance scale scores (basic trust, autonomy, identity, intimacy, generativity) were examined as a function of identity style (informational, normative, diffuse/avoidant), sex-role identity (feminine, masculine, androgynous, undifferentiated), age (young adult, middle adult, late adult), and gender. Three hundred and eighty-eight participants (166 men and 222 women; M age = 41.38, SD = 17.45) completed self-report questionnaires designed to measure the relevant constructs. Results indicated that identity processing style did not prove to be significantly related to relationship style or psychosocial balance scores. However, differences in sex-role identity were predictive of different patterns of scores on the relationship style measure and the psychosocial balance measure. In addition, the secure relationship style was positively predictive of all five psychosocial balance scale scores. These findings suggest that identity processing style may play a lesser role than sex-role identity classification in determining patterns of differences in relationship styles and psychosocial adjustment. TABLE OF CONTENTS Abstract ii Table of Contents v List of Tables vil List of figures ix Acknowledgement x INTRODUCTION 1 Erikson’s Theory of Psychosocial Development 2 Marcia’s Ego-Identity Status Paradigm Diffuse Identity Status Foreclosure Identity Status Moratorium Identity Status Achievement Identity Status Identity Status and Intimacy 3 4 5 5 6 6 Identity Style Paradigm Identity Style and Intimacy 7 11 Goals of the Present Research 12 METHOD 16 Participants 16 Procedure 20 Measures 21 Identity Style Inventory: Sixth Grade Reading Level Bem Sex-Role Inventory Relationship Style Questionnaire Inventory of Psychosocial Balance 21 23 25 25 RESULTS 27 Overview of the Analyses 27 IV Identity Processing Style and Relationship Style 29 Identity Processing Style and Psychosocial Balance 31 Sex-Role Identity and Relationship Styles 32 Sex-Role Identity and Psychosocial Balance 36 Sex-Role Socialization and Identity Style Subscale of Commitment 38 Prediction of Eriksonian Stage as a Function of Relationship Style 38 Standard Regression on Basic Trust Standard Regression on Autonomy Standard Regression on Identity Standard Regression on Intimacy Standard Regression on Generativity 39 39 39 40 40 DISCUSSION 64 Hypothesis 1 Hypothesis 2 Hypothesis 3 Hypothesis 4 Hypothesis 5 Hypothesis 6 64 66 67 71 73 75 Limitations of the Current Study 79 Recommendations for Future Research 81 Summary and Conclusion 82 REFERENCES 85 APPENDICES A. Demographics Form B. Consent Form C. Bem Sex-Role Inventory D. Relationship Style Questionnaire E. Identity Style Inventory - Grade Six Reading Level E. Inventory of Psychosocial Balance 91 91 92 93 95 97 99 List of Tables 1. Demographic Information for Participants. 18 2. Means and standard deviations for an age main effect on relationship style. 41 Means and Standard Deviations for the Secure Relationship Style as a function of Age, Gender, and Identity Style. 42 Means and Standard Deviations for the Fearful Relationship Style as a function of Age, Gender, and Identity Style. 43 Means and Standard Deviations for the Preoccupied Relationship Style as a function of Age, Gender, and Identity Style. 44 Means and Standard Deviations for the Dismissing Relationship Style as a function of Age, Gender, and Identity Style. 45 Means and Standard Deviations for the Trust Psychosocial Balance Scale as a function of Age, Gender, and Identity Style. 46 Means and Standard Deviations for the Autonomy Psychosocial Balance Scale as a function of Age, Gender, and Identity Style. 47 Means and Standard Deviations for the Identity Psychosocial Balance Scale as a function of Age, Gender, and Identity Style. 48 Means and Standard Deviations for the Intimacy Psychosocial Balance Seale as a function of Age, Gender, and Identity Style. 49 3. 4. 5. 6. 7. 8. 9. 10. 11. Means and Standard Deviations for the Generativity Psychosocial Balance Scale as a function of Age, Gender, and Identity Style. 12. 13. 14. 15. 50 Means and Standard Deviations for the Secure Relationship Style Scale as a function of Bem Class and Gender (For Two Age Groups Only). 51 Means and Standard Deviations for the Fearful Relationship Style Scale as a function of Bem Class and Gender (For Two Age Groups Only). 52 Means and Standard Deviations for the Preoccupied Relationship Style Seale as a function of Bem Class and Gender (For Two Age Groups Only). 53 Means and Standard Deviations for the Dismissing Relationship Style Scale as a function of Bem Class and Gender (For Two Age Groups Only). 54 VI 16. Means and Standard Deviations for the Trust Psychosocial Balance Scale as a function of Gender, and Sex-Role (Bem) Class. 55 17. Means and Standard Deviations for the Autonomy Psychosocial Balance Scale as a function of Gender, and Sex-Role (Bem) Class. 56 18. Means and Standard Deviations for the Identity Psychosocial Balance Scale as a function of Gender, and Sex-Role (Bem) Class. 57 19. Means and Standard Deviations for the Intimacy Psychosocial Balance Scale as a function of Gender, and Sex-Role (Bem) Class. 58 20. Means and Standard Deviations for the Generativity Psychosocial Balance Scale as a function of Gender, and Sex-Role (Bem) Class. 59 21. 60 ANOVA Results for IPB Scales as a function of Bem Class. 22. Standard Multiple Regression of Relationship Styles on IPB Subscale of Basic Trust. 61 23. Standard Multiple Regression of Relationship Styles on IPB Subscale of Autonomy. 61 24. Standard Multiple Regression of Relationship Styles on IPB Subscale of Identity. 62 25. Standard Multiple Regression of Relationship Styles on IPB Subscale of Intimacy. 62 26. Standard Multiple Regression of Relationship Styles on IPB Subscale of Generativity. 63 VII List of Figures 1. B atholomew and Horowitz ’s ( 1991 ) Model of Adult Attachment 13 2. Mean preoccupied relationship style scores as a function of gender 34 3. Mean dismissing relationship style scores as a function of gender 35 V III Acknowledgements Appreciation is extended to all of the people who took part in this research project, including Dr. Peter McMillan for his assistance with all matters statistical, Loreen Obst, Psychology secretary, for her help and support. Dr. Robert Tate for his compassion and understanding, Julie Orlando and the Institute for Social Research and Evaluation for assistance with data collection, Jessica Madrid for her thoroughness in data entry. Dr. Michele Worth for her excellent suggestions, Dr. Deborah Poff for the financial assistance that made much of the data collection possible, to my committee members. Dr. Ken Prchachin and Dr. Alex Michalos, for their diligent editing and constructive criticisms, and to those in the media who took my plea for participants to the community. All of these people have made contributions that paved the way for the success of this project. My greatest thanks go to my wife, Gail, who put in so many hours preparing mail-out packages and supporting me in a multitude of ways. Words seem insufficient to express how important she is to me. Finally, to Dr. Sherry Beaumont, I would like to convey my gratitude in being able to work with such an enthusiastic and capable supervisor, one who was always there when I needed a hand, but also had the wisdom to step back when I needed to learn on my own. IX Identity Processing Style and Sex-Role Identity Predict Attachment Style and Psychosocial Balance INTRODUCTION Developmental psychology is the study of growing up. That is, it examines the pathways of development from birth (and in some circumstances, before birth) through to adulthood with the goals of describing and explaining the various stages or transitions along the way. For example, for many psychologists, the manner in which we are attracted to, meet, and become intimate with romantic partners forms a major and compelling area of interest in the study of lifespan development. From this perspective, development may be seen as the progression from infancy to adulthood with the goal of adult social encounters and commitments that have the purposes of affiliation and, ultimately, progeny (Erikson, 1959). This developmental interest in relationships raises the question of how we come to choose those with whom we would become romantically involved and form bonds of sufficient magnitude that we would remain bonded with one person. Theories abound on how relationships develop. For example, Freud and Bowlby (Shaver & Hazen, 1993) assumed that early mother - child relations determined later romantic relationships. Maslow (1954) considered the experience of affiliation an aspect of a hierarchy of needs; once basic physiological and safety/security needs were attained, an individual sought belongingness and love before striving for esteem and self-actualization. Alternatively, Fromm (1968) proposed that the individual who had attained ego maturity sought to more fully express his or her humanity through the mutual tenderness and vulnerability traditionally found in romantic relationships. Identity Processing Style and Sex-Role Identity Theory may tell us, in the most general terms, why an individual may seek romantic attachments, but not the style of relationship an individual may prefer. In other words, we can assume that most people will pursue a relationship, but it is a more challenging task to predict the nature or quality of that relationship. The question of which aspect(s) of our development influence relationship duration, level of intimacy and commitment, and our sense of self within relationships remain grist for developmental research. The goal of the present research was to shed some light on this question from the perspective of lifespan development within an Eriksonian framework. Erikson’s Theory o f Psychosocial Development The theoretical underpinnings for this research will be based on Erikson’s (1959) eight stages of psychosocial development in which it is suggested that individuals encounter eight lifespan crises, or turning points. These begin with basic trust versus mistrust in infancy and culminate with integrity versus despair in old age. Erikson proposed that individuals reach “decisive encounters” when they are ready to process information that is supplied by their environment regarding particular ideas and concepts that impact the individual’s sense of self. The successful resolution of these decisive encounters, or crises, contributes to the individual’s ego identity, personality, and social interaction style through changes in interpersonal perspective that come about in the emergence of different capacities at various opportune moments while growing up (Erikson, 1959). Of particular importance to this research, Erikson considered that intimacy and affiliation during one’s adult years were a product of an ego identity established in the teen years. “The youth who is not sure of his identity shies away from interpersonal intimacy; but 2 Identity Processing Style and Sex-Role Identity the surer he becomes of himself, the more he seeks it” (Erikson, 1959, p. 101). Accordingly, individuals who have not developed ego identity strength find themselves in social isolation or serial attempts at intimacy out of a sense of tradition-bound, stereotyped relationship behavior. In the latter case, relationships take on an aura of formality where there is a lack of spontaneity, warmth, and trust (Erikson, 1959). Successful resolution of the identity-role diffusion crisis is predicted to lead to appropriate levels of intimacy in young adulthood and unsuccessful resolution is expected to lead to shallower, less intimate relationships. A more fine-grained analysis would seem required, given that the ego-identity crisis resolution is unlikely to be an all-or-none phenomenon. In order to explore the levels of the ego-identity/role diffusion crisis outcome, Marcia operationalized Erikson's concept of identity formation (Read & Adams, 1984). Marcia’s Ego-Identity Status Paradigm Marcia (1966) developed the notion of identity statuses in order to add nuance to Erikson’s theory. In his concept, Marcia suggested four levels of identity outcome: diffusion, foreclosure, moratorium, and identity achievement (Orlofsky, Marcia, & Lesser, 1973). From diffusion through to identity achievement, each identity status is said to represent a higher level, or more successful resolution, of Erikson’s concept of ego-identity attainment (Marcia, 1966). Marcia’s system contained two basic manners of differentiating ego-identity statuses. First, one could contrast the identity statuses in which there is commitment (foreclosure and identity achievement) with those lacking commitment (diffusion and moratorium). Second, based on experience with crisis, one may compare identity statuses 3 Identity Processing Style and Sex-Role Identity with the presence (moratorium and achievement) versus the absence (foreclosure and diffusion) of exploration. Whereas identity statuses can be linked to distinct personality characteristics, socialcognitive variables, and certain gender differences, progression from lower to higher identity statuses (i.e., from diffusion through identity achievement) is generally paralleled by an increasing complexity of personality attributes (Read et al., 1984). Therefore, similarities between individuals of a higher level of identity status may be subtler in nature than for lower statuses. For example, it has been suggested that greater gender differences exist for lower status versus higher status individuals (Waterman, 1982) and Erikson argued that identity formation was “a process of increasing differentiation” (1958, p. 23). Although some authors (e.g., Adams et al., 1985; Read, Adams, & Dobson, 1984) have stated that few, if any, significant gender differences exist between statuses, others have suggested that some gender differences do exist. For example, Fitch and Adams (1983) found that occupational identity for men and religious identity for women tended to contribute to higher identity status. It may be that gender differences emerge as broad trends rather than specific distinctions between identity statuses. Finally, for the purpose of the present research, an additional aspect of importance is the potential link between an individual’s identity status and style of intimacy in relationships (Fitch & Adams, 1983). Dijfuse identity status. Individuals with a diffuse identity status may or may not have experienced a crisis period, but cannot identify specific, meaningful commitments to a personal identity (Orlofsky, Marcia, & Lesser, 1973). This type of identity status may lead to an active pursuit of noncommitment and avoidance of demanding circumstances. Alternatively, it may be characterized by aimless drifting or by personal malleability and 4 Identity Processing Style and Sex-Role Identity suggestibility (Orlofsky, Marcia, & Lesser, 1973). On measures of social behavior, individuals with a diffuse identity status scored the lowest of the identity statuses for conformity behavior to gain from achievement and highest for being influenced by peer pressure (Adams, Ryan, Hoffman, Dobson, & Nielson, 1985). That is, these individuals engaged in conformity behavior, not in order to achieve their goals, but in response to peer pressure to behave in certain ways. Diffuse individuals may be seen as less developed in such personality dimensions as self-esteem, locus of control, and moral reasoning (Adams & Shea, 1979). Foreclosure identity status. Foreclosure status refers to those who have made a commitment, but without going through a crisis or exploratory period (Marcia, 1966). They have accepted the identity presented to them by parents or other authority figures. Of the four statuses, they tend to be the most authoritarian and rigid in their views (Orlofsky et al., 1973). They appear to operate rather well within the bounds of familiar circumstances, but may quickly find themselves at a loss if faced with an unfamiliar challenge (Orlofsky et al., 1973). For example, foreclosure women tend to express fear of the extra-familial environment and reduced tolerance for ambiguous situations (Read et al., 1984). Marcia (1980; cited in Read et al., 1984) found that foreclosure women scored lower than moratorium or identity achievement women on measures of ego strength and that higher status men demonstrated greater psychological maturity compared to lower status men. Moratorium identity status. The moratorium status involves a period of crisis in which the person is currently engaged in exploration for the purpose of identity formation (Marcia, 1966). Men with a moratorium identity status may appear very similar to identity achievement males on measures of social cognition and behavior (Orlofsky et al., 1973). 5 Identity Processing Style and Sex-Role Identity Foreclosure individuals exhibit contradictory desires for independence and direction from others as seen in their ambivalence towards authority (Macia, 1966). They can be engaging, active, and creative or mired by an inner struggle against indecisiveness. In general, they appear to be the most verbal and variable of the statuses (Orlofsky et al., 1973). Achievement identity status. As the most developed and differentiated, the identity achievement status involves commitment after a period of crisis (Marcia, 1966). People at this stage arrive at their identity after extensive exploration and form strong commitments. They appear stable, capable of dealing with shifting environmental demands, and can establish and pursue realistic goals (Orlofsky et al., 1973). Adams et al. (1985) noted that achievement identity men (compared to lower status men) were more relaxed, less prone to worry, experienced less social anxiety, demonstrated greater maturity in interpersonal functioning, and did not tend to exhibit extremes of either extraversion or introversion. Achievement identity women appeared more adept (compared to lower status women) at social-cognitive functions governing encoding, decoding, and analyzing interpersonal social information (Adams et al., 1985). Identity status and intimacy. The link between identity status and intimacy can generally be described as one where the higher the individual’s status, the greater their capacity for interpersonal relationships (Fitch & Adams, 1983). Orlofsky et al. (1973) described five levels of intimacy: intimate, preinitimate, stereotyped, pseudointimate, and isolate (ranging from greatest to least capacity for intimacy, respectively). Research findings indicate that more advanced identity statuses (i.e., moratorium and identity achievement) are associated with more advanced intimacy statuses, such as intimate and preintimate (Orlofsky et al., 1973). Indeed, the mutual distinction between moratorium and identity achievement 6 Identity Processing Style and Sex-Role Identity status as well as between preinitimate and intimate status is the element of commitment, which is present in the identity achieved and intimates statuses and absent in the moratorium and preintimate statuses (Orlofsky et al., 1973). Erikson (1959) stated that commitment to important social structures was an integral component enabling intimate relations. To this end, Orlofsky et al. (1973) found that whereas men appeared to require an occupational commitment in order to achieve higher intimacy status, women needed a religious commitment to do the same. The authors attributed this difference to assumed distinctions between men and women, with men believed to be more instrumental and achievement oriented and women more concerned with interpersonal issues of affiliation and personal improvement (Fitch & Adams, 1983). Fitch and Adams (1983) observed that the clearest relationship between identity and intimacy status occurred when data for each was measured cross-sectionally. The most notable difference was that moratorium and identity achievement individuals possessed a deeper capacity for intimacy when compared to lower identity status individuals, regardless of gender. Identity Style Paradigm Erikson (1959) emphasized the process aspect of ego-identity development and how this processing style then affected the manner in which individuals responded to interpersonal information. Extensive research based on Erikson’s theory of psychosocial development has illustrated the social-cognitive nature of the identity status paradigm, with the purpose of predicting behavioral and cognitive outcomes predicated on one’s identity status (Streitmatter, 1993). Yet, a less-than-perfect connection between identity status and social and cognitive behavior exists, due to the somewhat static or steady state nature of the 7 Identity Processing Style and Sex-Role Identity identity status construct (Berzonsky, 1989). Because of the stress Erikson placed on the process aspect of identity construction and its outcome for social interaction (e.g., occupation, relationships, etc.), a social-cognitive process paradigm appeared to better represent Erikson’s original concept (Berzonsky & Neimeyer, 1994). A second criticism of the status paradigm is that it does not adequately capture all individuals assessed by the construct in accordance with the theory (Eitch & Adams, 1983). Again, this is thought to be an issue of process versus state, since Fitch and Adams (1983) noted that over a span of one year, individuals assessed in a given identity status at the beginning of the research project were observed to have shifted to another status by the conclusion of the project. For example, foreclosure individuals were as likely to have changed as they were to have remained stable during the period of observation. This suggests that these individuals experiencing transition from one status to another would fail to be captured by any of the four Marcian statuses. To this end, Berzonsky (1989) developed the identity style theory in order to account for the link between identity status and cognitive style through an Eriksonian developmental process. Berzonsky (1989) developed a model of identity styles based on his theory that the existing structures outlined by Marcia (1966) described different social-cognitive processing orientations. The identity styles consisted of three styles (informational, normative, and avoidant) rather than the four statuses developed by Marcia, due to the notion that once commitment was held constant, moratorium and achievement statuses could not be meaningfully differentiated on the basis of social-cognitive functioning (Berzonsky, 1989). At the heart of social-cognitive functioning lies the thought that existing Marcian identity structures are made up of self-relevant schemata that supply the basis for interpreting social 8 Identity Processing Style and Sex-Role Identity information and help direct decision making and problem solving (Berzonsky, 1989). These adaptive efforts frequently imply the necessity of accommodation of disparate aspects of the identity structure with the individual’s environment. According to Berzonsky and Neimeyer, an individual’s “optimal identity development would involve an ongoing dialectical interchange between assimilative processes driven by the identity structure and contextdriven accommodation processes aimed at revising that structure’’ (1994, p. 426-427). Berzonsky and Neimeyer (1994) suggested that moratorium and identity achievement individuals depend on an informational style of decision making and problem solving with regard to self-relevant social input. These individuals search for options and actively process relevant information that impinges upon identity issues and engage in an “assimilativeaccommodative cycle’’ (Berzonsky & Neimeyer, 1994, p. 427). They express reluctance to automatically accept their self-constructions and seek to test and revise their identity when confronted by dissonant feedback (Berzonsky & Kuk, 2000). These individuals engage in self-reflection, are cognitively complexity, demonstrate a high need for cognition, and exhibit higher levels of the big five personality factors of conscientiousness, agreeableness, and openness (Berzonsky & Kuk, 2000). Foreclosure individuals, in contrast, rely on a “preemptive’’, normative orientation that is over-reliant on assimilation and that reflects the expectations and proscriptions of influential others. Research suggests that these individuals are conscientious and agreeable in the big five personality sense (Clancy Dollinger, 1995), but may experience a marked intolerance for ambiguity and a clear need for structure and cognitive closure (Berzonsky & Kuk, 2000). 9 Identity Processing Style and Sex-Role Identity Identity diffuse individuals are said to use an avoidant style and to accommodate readily to situational demands of their social environment. Such an avoidant style is thought to represent specific and transient behavioral acts or verbal acquiescence rather than concrete changes in identity style. They may be reticent to confront personal problems, preferring to delay decision making until situational demands dictate a response (Berzonsky & Kuk, 2000). The avoidant identity style has been linked to such maladaptive characteristics as self-handicapping, avoidant coping, and an external locus of control (Berzonsky & Kuk, 2000 ). Several studies have examined the validity of the connection between identity status, identity style, and Eriksonian identity formation processes. For example, Streitmatter stated that her results supported Berzonsky’s “theoretical speculation concerning the process of identity formation” (1993, p. 214) and that measures comparing identity statuses and identity styles yielded positive correlations in the predicted manner. Thus, more advanced identity styles correlated positively with the more advanced identity statuses. Schwartz, Mullis, Waterman, and Dunham (2000) also concluded that Berzonsky’s identity style construct effectively captured that essence of Marcia’s identity status paradigm while embracing the dynamic process element essential to Erikson’s theory. Research findings have been mixed with respect to gender differences for identity styles. For example, Schwartz et al. (2000) did not report any significant sex differences in their findings. Berzonsky (1993) also failed to find sex differences, noting that in a university population sample, late adolescent males and females were equally likely to employ each of the three identity styles. Conversely, Berzonsky (1992) found that more male than female subjects preferred a diffuse/avoidant style, whereas more female than male 10 Identity Processing Style and Sex-Role Identity subjects were categorized as informational. Berzonsky (1992) found that an informational style loaded positively on factors related to introspection, self-reflection, and openness to personal feelings and fantasy, suggesting that subjects with a feminine sex role engaged in self-focused attention to a greater extent than did masculine subjects. It has been suggested that future research efforts in this area attempt to determine whether sex differences that have been observed are a product of gender socialization versus biological gender (Berzonsky, 1992; 1993). Identity style and intimacy. Although no studies to date have focused on identity style and intimacy in a romantic relationship sense specifically, it is logical to consider that the research on identity status and romantic attachment style would guide assumptions concerning identity style and attachment. Berzonsky and Neimeyer (1994) have commented that structure and process cannot be considered independent. Just as the commitments that might arise through an identity structure (e.g., marriage or occupation) are influenced by the decision and problem-solving strategies associated with a given identity style, those processes are then governed by these very same commitments. Thus, it seems reasonable to suggest that the relationship between identity style and romantic attachment style would be similar in nature to identity status and romantic attachment. For example, a higher identity style (information) would likely be associated with a greater capacity for intimacy, based on the research data suggesting that individuals with an informational identity style have a greater capacity for openness, agreeableness, problem-solving, and self-reflection (Berzonsky & Kuk, 2000). In addition, it has been noted that the informational identity style bears the hallmarks of the achievement identity status (Streitmatter, 1993) and that 11 Identity Processing Style and Sex-Role Identity individuals with an achievement identity status demonstrate a deeper level of intimacy than less advanced identity statuses (Fitch & Adams, 1983). Goals o f the Present Research The present research was exploratory in nature and had several goals predicated on the extant literature concerning identity style, Erikson’s model of psychosocial development, attachment style, and the above noted question of gender differences in relation to identity style. The primary goal was to determine what, if any, associations emerge between identity style as measured by the Identity Style Inventory - Grade six reading level (1S1-6G; White et al., 1998) and adult romantic attachment style for men and women across three levels of adult age range: young adults (18-40 years), middle-aged adults (40-60 years), and the elderly (61 years and up). Attachment style was assessed through the Relationship Styles Questionnaire (RSQ; Griffen & Bartholomew, 1994), which examines romantic attachment style on two dimensions. The first is the individual’s model of self and can be conceived of as dependence, whereas the second is an individuals’ model of others and can be thought of as avoidance (Bartholomew & Horowitz, 1994). Both dimensions capture a low (or positive) level and a high (or negative) level, as is illustrated in Figure 1 (from Bartholomew & Horowitz, 1991, p. 227). 12 Identity Processing Style and Sex-Role Identity Figure 1. Batholomew and Horowitz’s (1991) model o f adult attachment. A Two-Dimensional Model of Adult Romantic Attachment Model of Self (Dependency) Model of Other Positive (low) (Avoidance) Negative (high) Positive (low) Negative (high) Cell I Secure; Comfortable with intimacy and autonomy Cell III Dismissing: Dismissing of intimacy; counter-dependent Cell II Preoccupied: Preoccupied with relationships CelllV Fearful: Fearful of intimacy; socially avoidant 13 Identity Processing Style and Sex-Role Identity According to this model, individuals are thought to demonstrate distinct relationship behavior patterns depending on their respective attachment style. For the secure individual, evaluation of self and others is positive, intimacy is valued, and a well-developed level of personal autonomy allows the secure individual to remain comfortable in the absence of the attachment object. For the preoccupied individual, evaluation of others is positive, whereas evaluation of the self is negative. These individuals are not comfortable with autonomy and are enmeshed in their relationships, seeking validation through attachment objects. Individuals with afearful attachment style are thought to place a negative value on the self and others, leading to a sense of unworthiness as well as a fear of others as untrustworthy and rejecting. They are uncomfortable with either autonomy or intimacy. Finally, those with a dismissing attachment style are said to evaluate themselves positively, but others negatively, leading to a tendency to reject intimacy yet experience a sense of self as love-worthy. They protect themselves from disappointment by eschewing close relations with potential intimates and vigilantly maintaining their autonomy (Bartholomew & Horowitz, 1991). Given the above-mentioned information regarding links between identity status and intimacy (e.g., Fitch & Adams, 1983), it is reasonable to predict that similar associations might emerge between measures of attachment style and identity style. However, it should be noted that Fitch and Adams (1983) used a five-category measure of intimacy developed by Orlofsky, Marcia, and Lesser (1973), not the four-category RSQ developed by Griffin and Bartholomew (1994) that will be employed in the present research. A second goal for the present research will be to assess for stage of psychosocial development in an Eriksonian framework. For this purpose, the Inventory of Psychosocial Balance (IPB; Domino & Affonso, 1990) will be employed. The IPB assesses all eight of 14 Identity Processing Style and Sex-Role Identity Erikson’s stages, but for the present research, only the stages considered directly relevant will he used. These are: basic trust versus mistrust, autonomy versus shame and doubt, identity versus role confusion, intimacy versus isolation, and generativity versus stagnation (Domino & Affonso, 1990). The objective of assessing for psychosocial stage is twofold: first, to explore the relationship between stage of development and attachment style, and second, to use as an outcome measure in relation to identity style. The importance of the latter objective is to confirm the link between Erikson’s psychosocial stage construct and the dynamic model of identity style as developed by Berzonsky. In other words, if identity style is a better reflection of Erikson’s concept, this should be reflected in the data by positive results. With respect to the first objective, it might be predicted that individuals who have successfully resolved the stages of trust, autonomy, identity, and intimacy will demonstrate a higher level of attachment style (i.e., secure) than individuals who have not successfully resolved these crises (Erikson, 1959). Conversely, an individual who may never have resolved basic trust issues, might be observed with a dismissing attachment style and evince a quality of mistrust of others where intimacy is concerned (Bartholomew & Horowitz, 1991). The second objective is to use psychosocial stage as an outcome variable of identity style. That is, individuals with an informational identity style might be found to have successfully resolved Erikson’s identity stage; conversely, diffuse/avoidant individuals would be expected to have unsuccessfully resolved that crisis (Erikson, 1959). 15 Identity Processing Style and Sex-Role Identity METHOD Participants The participants for this research project were men and women from the Prince George region from the age of 18 years and up (range = 18 to 85). Participants were grouped in age categories of 18 - 40 (young adults), 4 1 - 6 0 (middle-aged adults), and 61 years and older (seniors). The goal of the project was to recruit approximately 90 men and 90 women in each age category in order to acquire approximately 30 participants in each age range in each Identity Style (Information, Normative, and Diffuse; see Measures section for more detail). Recruitment took place in a number of ways. Participants were drawn from: 1) the UNBC Psychology student subject pool; 2) advertisements in local media, such as radio, television, and newspapers; 3) circulation to various community agencies dealing with seniors, such as intermediate- and long-term care facilities, local health nurses, and others that go out to seniors’ homes in the community; 4) businesses and service agencies that have large staffs, such as RCMP, PGRH, Telus, CNC, etc.; 5) local community service associations (e.g.. Elks Club, Kiwanis, Rotarians, etc.); and, 6) a mail out of 1,000 surveys sent to names that were randomly selected from an existing database. Precautions were taken to ensure the young adult age group was not entirely made up of participants from the UNBC student subject pool. Recruitment efforts yielded a research sample of 388 individuals, although the above mentioned target of equal representation in each age category and identity style was not met. For example, participants from the senior age category were much more difficult to recruit than it was presumed that they would be. As well, the young adult participants were also 16 Identity Processing Style and Sex-Role Identity more difficult to recruit, with over half the young adults coming from the UNBC student subject pool. Finally, the mail out yielded a return rate of 22% and given the relatively young age of the population of Prince George (modal age group is 25 - 44; Statistics Canada, 2001), fewer senior adults returned surveys than was hoped. Table 1 summarizes the demographic information for the research sample based on a Demographic Questionnaire completed by each participant (see Appendix A). 17 Identity Processing Style and Sex-Role Identity Table 1. Demographic information fo r participants. Variable Frequency Percent Age (overall) M SD 41.38 17.45 Age Groups Young Adult 185 47.7 26.10 6.712 Middle Adult 134 34.5 4&60 6.684 Senior Adult 69 17.8 68.33 7.043 Male 166 428 Female 222 57.2 English 354 91.2 Other 27 7 Aboriginal 8 2.1 Asian 4 1.0 Caucasian 336 86.6 Black (other than African) 2 0.5 African-Canadian 2 0.5 Other 22 5.7 186 47.9 Gender Mother Tongue* Ethnicity* Marital Status* Married/common-law Table continues 18 Identity Processing Style and Sex-Role Identity Table 1 cont. Single 31 8.1 Divorced 129 33.9 Separated 16 4.2 Widowed 19 5.0 Elementary school 10 2.6 Secondary school 35 9.0 High school diploma 72 18.6 Trade/Technical 33 8.5 Some college 41 10.6 College diploma 49 12.6 Some university 85 21.9 University degree 49 12.6 Other 13 3.4 Full-time 200 51.5 Part-time 79 20.4 Retired 43 11.1 Unemployed 49 12.6 Level of Education* Level of Work* Note. N = 388 * denotes that percentage does not equal 100 due to missing data 19 Identity Processing Style and Sex-Role Identity Procedure Surveys were disseminated in a variety of ways. First, a mail-out of 1,000 questionnaire packages with prepaid return envelopes was conducted with names supplied from a list of randomly drawn local addresses. The return rate was 18.3%. Second, a local newspaper and radio station provided coverage of the research project through a print article in the first case and a series of public notices in the latter. The newspaper article and radio announcements asked for volunteers and informed the public as to how they might participate. Third, students at the University of Northern British Columbia enrolled in an introductory Psychology course were advised of an opportunity to enlist in the project and participate in order to receive extra course credit. Fourth, a number of agencies (e.g., senior centers, intermediate care facilities, and adult daycare agencies) in Prince George were approached in an effort to acquire as many senior adults as possible to volunteer for the project. Fifth, colleagues and other individuals with access to large numbers of potential participants were given survey packages to disseminate in an informal manner. Respondents to newspaper advertisements and radio and TV segments (as well as those contacted by word-of-mouth) were given the option of: 1) going to the UNBC lab to complete the questionnaires; 2) having them brought to their residence; or, 3) picking up questionnaire packages from the university (sent back by pre-paid envelope). University students enlisted from the Psychology subject pool were surveyed in a group setting. Attention was paid to research ethics and participants were briefed on informed consent (see informed consent form in Appendix B) as well as debriefing, which consisted of a brief description of the overview of the project’s main research objectives with an emphasis on the participant’s contribution to an analysis of broad trends versus individual traits or characteristics. 20 Identity Processing Style and Sex-Role Identity Measures Participants completed the following questionnaires which were bound together as a package with the order counterbalanced across participants. Identity Style Inventory: Sixth Grade Reading Level (ISI-6G). The IST6G (White et al., 1998; see Appendix C) consists of 40 sentences regarding the individual’s identity style. The statements reflect the individual’s cognitive processes marshaled in response to the demands at various life stages or developmental crises (White et al., 1998). Eleven items center around the information style, nine for the normative style, 10 for the diffuse style, and 10 for the commitment scale (a secondary analysis sub-scale used to determine style when two identity styles receive the same mean score). Respondents are asked to indicate agreement/disagreement with item statements by rating them on a five-point Likert scale ranging from “strongly agree” to “strongly disagree”. The ISI-6G categorizes participants into one of three discrete identity styles: Information, Normative, or Diffuse/Avoidant. Scoring involved the computation of a mean rating (raw score) for each scale, conversion into Z-scores (using the present research sample population mean), and assignment to a particular identity style based on the highest Z-score. Internal reliability (Cronbach’s alpha) has been reported as good by varying sources. For example, Berzonsky (1992) reported Cronbaeh al.phas ranging from .62 for Information to .79 for Diffuse/Avoidant (Berzonsky, 1994). White et al. (1998) reported internal reliability of .59 for Information style, .64 for Normative and .78 for Diffuse/Avoidant. Test-retest reliability over a two-month span has been reported at .71 for Diffuse/Avoidant and .75 for Information style. Evidence for convergent validity between the ISI-6G and the original ISI is excellent (Information, r = .81; Normative, r = .85; Diffuse/Avoidant, r = .85; all 21 Identity Processing Style and Sex-Role Identity correlations significant to p = .001). The 1S1-6G is considered valid for the purpose of assessing identity style (Berzonsky, 1993; Streitmatter, 1993). The Identity Style Inventory has three identity style subscales and a fourth subscale that measures what has been termed identity commitment (White et al., 1998). The commitment subscale items are intermingled with the three identity style items, but are scored separately with their own mean and z-score. The commitment subscale is used for secondary analyses after, or in addition to, the primary analyses to determine an individual’s level of commitment to his or her particular identity style. Examples of commitment sub­ scale items are, “’’People need to be committed to a set of values to live a full life’’ and “1 have a strong set of beliefs that 1 use when 1 make decisions’’ (White et al., 1998). Although White et al. (1998) envisioned diffuse/avoidant individuals as being low on the measure of commitment and normative individuals as measuring high, the aspect of commitment was seen as separate from the primary measure of identity style (see Figure 1, p. 226, White et al., 1998). For example, individuals assessed as possessing an information identity style could be either high or low on a measure of commitment. Indeed, Berzonsky (1989) and Streitmatter (1993) stated that when comparing the concepts of ego identity status and ego identity style, a significant, positive correlation between the information identity style and the moratorium identity status only appeared once commitment was statistically controlled for. Thus, commitment might be best thought of in the context of identity style as the dedication to one’s particular social-cognitive information processing style. In this sense, it should be considered an aspect of identity structure as opposed to process since it appears to act as a mechanism by which one adheres to a given identity style process. Broadly speaking, the 1S1-6G commitment scale reflects an individual’s ability to make a commitment to the 22 Identity Processing Style and Sex-Role Identity contributing source(s) of their identity model. Whereas diffuse/avoidant individuals exhibit a lack of identity orientated commitments, normative individuals form commitments to authority figures and powerful others and information individuals to aspects of their own process of exploration (Schwartz et al., 2000). Bern Sex Role Inventory (BSRI). The BSRI was designed to assess the degree of masculinity and femininity exhibited by an individual by their endorsement of adjectives reflecting aspects of gender socialization (Bem, 1981; see Appendix D). The short form consists of 30 items, 10 of which measure characteristics that are stereotypically masculine, and 10 items that are stereotypically feminine (the remaining 10 items are neutral “fillers”). Examples of stereotypic feminine sex role adjectives include “shy”, “loyal”, “sympathetic” and “warm”. Examples of stereotypical masculine sex role adjectives include “dominant”, “aggressive”, “ambitious”, and “analytical”. Respondents indicate how accurately an item describes themselves on a Likert scale ranging from 1 (Never or almost never true of me) to 7 (Always or almost always true of me). Raw scores computed by adding the 10 masculine statements and the 10 feminine statements are compared to a median split score calculated from the research sample (Lenney, 1991). Individuals with a higher raw score for feminine items, but a lower raw score for masculine items would be categorized as Feminine. Those with raw scores higher than the median split on the masculine items, but lower than the split on the feminine items would be classed as Masculine. Those with both masculine and feminine raw scores higher than the median split would be classed as Androgynous, whereas if both raw scores were lower than the split, the individual would be classed as Undifferentiated. The BSRI regards masculinity and femininity as two distinct 23 Identity Processing Style and Sex-Role Identity characteristics rather than points on a single continuum, and thus, an individual can be scored as high or low on both aspects. Bem (1974) computed a coefficient alpha from a sample of 444 males and 279 females; (Masculinity = .86; Femininity =.80). Wilson and Cook (1984; cited in Lenney, 1991) found internal reliability for the BSRI of .88 for Masculinity and .78 for Femininity. Test-retest reliability over a 4-week interval was very good at .90 for both Masculinity and Femininity (Bem, 1974). Determining validity, particularly convergent validity, is a complicated matter with the BSRI as other instruments that attempt to assess the expression of gender tend to define sex roles differently, making direct comparison difficult if not impossible (Bem, 1974). However, Bem’s results from her Stanford norming study suggest that masculinity and femininity are independent constructs. In the Stanford study, a low correlation was seen between men’s total sex role inventory scores and the femininity scale and women’s total sex role inventory scores and the masculinity scale (r = .11 and r = -.14, respectively; Bem, 1974). Moreover, the BSRI appears to be free from response set and low and nonsignificant correlations were found between masculinity, femininity, and androgyny T-scores and social desirability scales (Bem, 1974). The BSRI is believed to have good discriminant validity (Lenney, 1991). Although items were not selected to differentiate between respondents on the basis of biological sex, the genders do differ in their scale scores, with men scoring significantly higher on the masculinity scale and women scoring significantly higher on the femininity scale (Bem, 1974). Finally, several studies support the BSRI on the matter of concurrent and predictive validity (Bem, 1974). Relationship Scale Questionnaire (RSQ). The RSQ (Griffin & Bartholomew, 1994; see Appendix E) consists of 30 brief statements culled from Hazen and Shaver’s (1987) 24 Identity Processing Style and Sex-Role Identity attachment measures, Bartholomew and Horowitz’s (1991) Relationship Questionnaire, and Collins and Read’s (1990) Adult Attachment Scale. Respondents rate the descriptiveness of each statement for their relationship style in romantic relationships on a 5-point Likert scale. Relationship styles consist of Secure, Preoccupied, Dismissing, and Fearful. Styles are determined by calculating the mean scores for those items representative of each style. The RSQ is not a categorical measure of relationship style, but yields four separate measures as continuous variables. Test-retest reliability appears adequate, with stability over periods of two weeks to four years in the .60 range (Brennan & Bosson, 1998). Internal consistency also appears to be adequate, with consistent results emerging from various studies employing the RSQ (e.g., Brennan & Bosson, 1998). The validity of the RSQ has been established by several studies (see Griffen & Bartholomew, 1994, for a review). The Inventory o f Psychosocial Balance (IPB). Developed by Domino and Affonso (1990; see Appendix F), the IPB is intended to measure all eight of Erikson’s stages of psychosocial development. The IPB consists of 120 items using a five-point Likert-type scale assessing the eight stages (trust, autonomy, initiative, industry, identity, intimacy, generativity, and ego integrity) with each sub-scale based on 15 items. For the purposes of this research project, only five sub-scales were relevant; Trust, Autonomy, Identity, Intimacy, and Generativity. Ego Integrity, the last stage, was also part of the questionnaire, but was not used in the analysis because of the relatively small sample size for the late adults and because this sub-scale would not be relevant for the other early and middle adults. Each sub-scale consists of 15 items, which are added up to form a raw score, providing a continuous variable. 25 Identity Processing Style and Sex-Role Identity The IPB is considered appropriate for adolescents and adults (Domino & Affonso, 1990). Coefficient alpha ratings resulting from testing with the first college student norming population for the sub-scales ranged from .74 for Industry to .48 for Autonomy with between-samples correlations ranging from .90 for Initiative to .78 for Ego Integrity (Domino & Affonso, 1990). It is important to note that similar results emerged from further testing with different populations (i.e., community adults and a sample of elderly people). Domino and Affonso (1990) cite various sources for support of the IPB validity and conclude that the IPB is a valid instrument for the purpose of assessing lifespan development from an Eriksonian perspective. 26 Identity Processing Style and Sex-Role Identity RESULTS Overview o f the Analyses All data from the test instruments were entered item-by-item. The data were then screened for possible problems with normality, homogeneity of variance, outliers, and missing data. No significant problems were identified, and the data were found to fall within acceptable parameters according to the standards described by Tabachnick and Fidell (2001). Although the distributions for some variables showed skewness (e.g., IPB scales for Basic trust and Generativity), a decision was made to analyze the raw data rather than transformed data because data transformations provided only marginal improvement in skewness (or similar skewness in the opposite direction). Departure from normality was also checked with the Kolmogorov - Smirnov statistic (with Lilliefors Significance Correction), and no significant statistics were found. There were a small number of cases in which the missing data exceeded Tabachnick and Fidell’s (2001) five percent rule (i.e., a recommendation to rectify situations in which more than five percent of data were missing), but there was no indication that the data were missing in a nonrandom fashion. In no case did missing data constitute more than 6.9% of the total data set (e.g., IPB subscale for Autonomy). Additionally, even with missing data, the sample size was deemed sufficient for the present analyses. Tabachnick and Fidell (2001) suggest that some outliers can be expected in a large sample and are a part of the real-world situation the research sample attempts to emulate. To this end, only data for one participant were eliminated from the study as an outlier due to extreme scores on a large percentage of variables as well as a suspicion that this participant’s responses might not have been genuine (reflected a patterned response set). Finally, scatter 27 Identity Processing Style and Sex-Role Identity plots were examined for troubles in homogeneity of variance. No curvilinear relationships were observed and where a relationship between variables existed, it appeared linear. Several types of analysis were conducted to adequately address the proposed hypotheses as follows. In all significant multivariate analyses of variance (MANOVA), which were followed by univariate analyses of variance (ANOYA), significant effects were followed by examinations of differences between pairs of means using Tukey’s HSD tests. An alpha level of .05 was applied for all statistical tests. Means and standard deviations for all variables considered are displayed in appropriate tables that can be found at the end of the Results section. 1. Identity Processing Style and Relationship Style: Analyses of variance (multivariate and univariate) were conducted to address the question of whether there were differences in relationship styles as a function of age, gender and identity style category. 2. Identity Processing Style and Psychosocial Balance: Analyses of variance (multivariate and univariate) were conducted to address the question of whether there were differences in psychosocial balance as a function of age, gender and identity style category. 3. Sex-Role Identity and Relationship Style: Analyses of variance (multivariate and univariate) were conducted to address the question of whether there were differences in relationship styles as a function of age, gender and sex-role identity (Bem class). 4. Sex-Role Identity and Psychosocial Balance: Analyses of variance (multivariate and univariate) were conducted to address the question of whether there were 28 Identity Processing Style and Sex-Role Identity differences in psychosocial balance as a function of age, gender and sex-role identity (Bem class). Sex-Role Identity and Identity Processing: A one-way analysis of variance was 5. conducted to determine whether there were differences in participants’ identity style commitment scores as a function of Bem class. Relationship Style and Psychosocial Balance: A series of regressions were 6. conducted to determine if scores on the various relationship style scales (secure, preoccupied, fearful, avoidant) predicted scores on the psychosocial balance scales. Identity Processing Style and Relationship Style Participants’ scores on the relationship style scales were analyzed by a 3 (age) by 2 (gender) by 3 (identity style) between-subjects multivariate analysis of variance (MANOVA) with the four relationship styles (secure, preoccupied, fearful, dismissing) as the dependent variables. This analysis returned a significant multivariate interaction between relationship style and age group, F (6, 672) = 3.19, p = .004, eta-squared = .028. There were no significant multivariate main effects. The means and standard deviations for the multivariate interaction of relationship styles and age group are displayed in Table 2. Although there were no significant differences in relationship style scores as a function of identity style, for interest, the means and standard deviations for the full analysis are presented in Tables 3 through 6. The significant multivariate interaction of age group and relationship style was examined at the univariate level for each of the four relationship style scales (secure. 29 Identity Processing Style and Sex-Role Identity preoccupied, fearful, and dismissing). The results of one-way analyses of variance (ANOVA) revealed significant main effects of age for the secure and preoccupied relationship styles, F (2, 365) = 8.25, p < .001, for secure; F (2, 378) = 13.01, p < .001, for preoccupied. Examination of the relevant means for the age main effect for the secure relationship style indicated that young adults reported significantly higher secure relationship style scores than did the oldest adults {M = 3.22, SD = .53, and M = 1.9\, SD = .47, respectively). In addition, the middle adults reported significantly more security than did the older adults {M =3.11, SD = .54, and M = 2.91, SD = .47, respectively). With respect to the preoccupied relationship style, the young adults reported significantly higher preoccupied relationship scores compared to the middle adults (M = 2.77, SD = .62 versus M = 2.48, SD = .63). The young adults also reported greater relationship preoccupation compared to the senior adults (M = 2.77, SD = .62 versus M = 2.40, SD = .54). Secondary analyses that excluded the oldest age group were conducted because the assignment of elderly participants to the identity style categories resulted in too few participants in some categories (e.g., normative females, n = 6). As a result, the young adults’ and middle adults’ relationship style scores (secure, preoccupied, fearful, and dismissing) were analyzed by a multivariate 2 (age group) by 2 (gender) by 3 (identity style) between-subjects design. Analyses revealed significant multivariate interactions between relationship style and age group and between relationship style and gender: for age group, F (3, 285) = 3.68, p = .013, partial eta-squared = .037; for gender, F (3, 285) = 4.01, p = .008, partial eta-squared = .040. The significant multivariate interactions between relationship style and age group and between relationship style and gender were examined at the univariate level by conducting 30 Identity Processing Style and Sex-Role Identity separate one-way ANOVAs for each of the relationship style dependent variables. These analyses revealed a significant main effect of age group for the preoccupied relationship style, F (1, 311) = 16.12, p < .001. A comparison of the relevant means for the age group main effect for the preoccupied relationship style indicated that young adults reported significantly higher preoccupied relationship style scores than did the middle adults (M = 2.77, SD = .62 versus M = 2.49, SD = .63, respectively). The univariate analyses also revealed significant gender effects for two of the relationship style scales: F ( I , 311) = 10.60,p = .001, for preoccupied, and F (1, 311) = 6.73, p = .01, for dismissing. Examination of the relevant means for the significant effects of gender indicated that females reported significantly greater scores than did males on the preoccupied relationship style scale (M = 2.75, SD = .66 versus M = 2.52, SD = .58, respectively). Additionally, there was a significant main effect of gender with respect to the dismissing relationship style. Females scored significantly higher compared to males on the dismissing relationship style scale (M = 3.44, SD = .53 versus M = 3.27, SD = .52). Identity Processing Style and Psychosocial Balance Another question of interest, given that the present research was seated in an Eriksonian framework, was whether differences in psychosocial balance (as measured by the IPB) would emerge as a function of differences in identity style as well as age group and gender. Again, since certain cells contained too few individuals (e.g., females, senior adults, normative identity style, n = 3), these analyses were conducted employing the first two age groups only (i.e., senior adults were dropped). Thus, the young adults’ and middle adults’ scores for IPB Eriksonian stage of psychosocial development were analyzed by a 31 Identity Processing Style and Sex-Role Identity multivariate 2 (age group) by 2 (gender) by 3 (identity style) between-subjects design. Analyses revealed two two-way multivariate interactions. The first interaction was IPB stage by age group, F (4, 274) = 7.67, p < .001, partial eta-squared = .101, and the second interaction was IPB stage by gender, F (4, 274) = 4.08, p = .003, partial eta-square = .056. The means and standard deviations for this multivariate analysis are displayed in Table 7 through 11. The significant multivariate interactions between IPB stage and age group and between IPB stage and gender were examined at the univariate level by conducting separate one-way ANOVAs for each of the IPB stage dependent variables. Univariate analyses revealed main effects of age for the basic trust sub-scale, F (1, 302) = 3.99, p = .047, and for the generativity sub-scale, F (1, 311) = 9.98, p = .002, and a main effect of gender for the intimacy sub-scale, F (1, 305) = 8.85, p = .003. Examination of the relevant means for the age main effects indicated that the middle adult group (M = 57.54, SD = 6.94) demonstrated higher basic trust scores than the young adults (M = 56.00, SD = 6.52). As well, the middle age group (M = 60.82, SD = 5.66) reported higher generativity scores than did the young adult group (M = 5S.51, SD = 6.62). For gender, females had higher scores for the intimacy scale (M = 58.45, SD = 7.54) than did males (M = 55.82, SD = 7.88). Sex-Role Identity and Relationship Styles The second major question of interest was whether differences in relationship style are, in part, a function of gender versus gender socialization. To explore this matter, a multivariate 4 (Bem sex-role socialization class) by 2 (gender) between-subjects design was 32 Identity Processing Style and Sex-Role Identity conducted with the four relationship styles (secure, fearful, preoccupied, and dismissing) as the dependent variables. Again, due to the small sample size for some age groups, this time not restricted to the senior group, age was not used as an independent variable. Indeed, for this set of analyses, the senior group was not included because it would have resulted in certain cells with only one individual (e.g., senior adult female with Bem Masculine sex-role class). The analyses employing the young adults and middle age adults returned a significant multivariate interaction of relationship style and Bem sex-role class, F (9, 876) = 5.03, p < .001, partial eta-squared = .049, and a marginally significant interaction of relationship style, Bem sex-role class and gender, F (9, 876) = 1.76, p = .072, partial eta-squared = .018. The means and standard deviations for this analysis are presented in Tables 12 through 15. The significant multivariate interaction between relationship style and sex-role socialization was examined at the univariate level by conducting separate one-way ANOVAs for each of the relationship style variables. With respect to the univariate analyses, the results demonstrated two significant interactions between Bem class and gender. For the preoccupied relationship style, F (3, 305) = 3.15, p = .025, females were found to have significantly higher scores for the preoccupied relationship style when in the Bem classes of feminine and androgynous, but not masculine or undifferentiated, in comparison to males (see Figure 2). Similarly, females also scored significantly higher for the dismissing relationship style as a function of Bem class, F (3, 305) = 2.78, p = .004. That is, females had higher mean dismissing relationship style scores in the Bem classes of masculine and undifferentiated, but not in the feminine or androgynous (see Figure 3). 33 Identity Processing Style and Sex-Role Identity H females ■ males masculine feminine androgynous undifferentiated Bem Sex-Role Identity Style Figure 2, Mean preoccupied relationship style scores as a function of gender. 34 Identity Processing Style and Sex-Role Identity a females ■ males : masculine feminine androgynous undifferentiated Bem Sex-Role Identity Style Figure 3. Mean dismissing relationship style scores as a function of gender. 35 Identity Processing Style and Sex-Role Identity In addition to the interactions, there were two main effects for Bem sex-role class for the remaining two relationship styles, secure and fearful. For the secure style, F (3, 301) = 5.40, p = .001, individuals with a sex-role class of androgynous (M = 3.29, SD = .55) scored significantly higher on the secure relationship style scale than did individuals whose sex-role class was undifferentiated (M = 3.03, SD = .54), but did not score significantly higher than those individuals with a feminine or masculine sex-role class. For the fearful relationship style, F (3, 307) = 7.01, p < .001, individuals with the sex-role class of undifferentiated (M = 2.94, SD = .80) scored significantly higher on the fearful relationship style scale than did those with sex-role class of androgyny (M = 2.53, SD = .80), but not score significantly higher than those with a masculine or feminine sex-role class. Sex-Role Identity and Psychosocial Balance Another aspect of interest to the current project is the possible relationship between Bem classification and psychosocial balance. Specifically, the goal was to assess psychosocial balance (i.e., Eriksonian stage) as a function of Bem class versus gender. To test this, a multivariate 4 (Bem class) by 2 (gender) between-subjects design was run with IPB stages as the dependent variables. Results indicated a significant multivariate interaction between psychosocial balance and Bem class as well as between psychosocial balance and gender; for IPB scales and Bem class, F (12, 840) = 5.02, p < .001, for IPB scales and gender, F (4, 278) = 2.93, p = .021. The means and standard deviations for this analysis are displayed in Table 16 through 20. 36 Identity Processing Style and Sex-Role Identity The significant multivariate interactions between IPB scales and sex-role class and between IPB scales and gender were examined at the univariate level by conducting separate one-way ANOVAs for each of the IPB scales. For gender, only the IPB scale of intimacy proved significant, F (1, 305) = 8.85, p = .003, with females (M = 58.45, SD = 7.54) having higher mean scores for intimacy than males (M = 55.82, SD = 7.88). With respect to Bem class, all five IPB scales were significant and in every case, individuals with a sex-role class of androgyny proved to have the higher mean score (see Table 21 for F-scores and p-values). Although individuals with a sex-role class of androgyny provided the highest mean scores for the IPB scales, they were not higher than all other Bem classes in all cases. For basic trust, individuals’ androgyny scores were significanlty higher (M = 60.06, SD = 5.27) than the other three classes (masculine; M = 54.84, SD = 6.14; feminine: M = 57.49, SD = 5.79; undifferentiated: M = 53.53, SD = 6.28). For autonomy, androgyny (M = 51.06, SD = 3.84) was significantly higher than femininity (M = 47.64, 5D = 4.15) and undifferentiated (M = 47.78, SD = 4.06) but was not higher than masculinity. Similarly, for identity, androgyny (M = 55.53, SD = 6.51) was significantly higher than femininity (M = 53.15, SD = 5.45) and undifferentiated (M = 51.01, SD = 5.653), but not significantly higher than masculinity. In the psychosocial stage of intimacy, androgyny (M = 60.34, SD = 6.03) was significantly higher than masculinity (M = 55.65, SD = 7.10) and undifferentiated (M = 53.18, SD = 7.72), but not significantly higher than femininity. Finally, androgyny (M = 62.58, SD = 5.74) did not surpass femininity for the scale of generativity, but was significantly higher than masculinity (M = 58.444, SD = 5.592) and undifferentiated (M = 56.39, 5D = 5.83). 37 Identity Processing Style and Sex-Role Identity Sex-role socialization and identity style subscale o f commitment. A further question of interest corresponded to a subscale of the identity style questionnaire that refers to secondary analyses, namely the commitment component of identity style. To assess level of ego identity style commitment within the sample, a one-way ANOVA was performed with Bem class as the independent variable and the identity style commitment measure as the dependent variable. The univariate results indicated a significant between-subjects difference, F (3, 377) = 16.53, p < .001. Individuals with the Bem class of androgyny (M = 4.04, SD = .61) scored significantly higher on the measure of identity style commitment than those in the other three classes (masculine; M = 3.74, SD = .61, feminine: M = 3.79, SD = .58; undifferentiated: M = 3.45, SD = .60). Those individuals in the undifferentiated class scored significantly lower on this measure than the other three sex-role classes (masculine, feminine, and androgyny). There were no significant differences between masculine and feminine or between masculine and feminine and the other two classes except where already noted (i.e., both lower than androgyny and greater than undifferentiated). Prediction o f Eriksonian Stage as a Function o f Relationship Style The final section of results pertains to a series of standard linear regressions performed with mean relationship style scores as the predictor variables and IPB stage as the criterion in five separate regressions (i.e., one for each IPB stage). All regressions performed were conducted in a standard enter fashion with no assumption that there was any rationale for prioritizing one relationship style over another for either statistical or theoretical reasons. 38 Identity Processing Style and Sex-Role Identity Standard multiple regression on Basic Trust. The regression procedure using relationship style scores as predictor variables and the IPB basic trust score as the criterion variable was significant, F (4, 383) = 40.36, p < .001. The secure relationship style score was significantly and positively related to the basic trust scale of the IPB, whereas the fearful relationship style was significantly and negatively related to the basic trust scale of the IPB (see Table 22 for details). In other words, higher secure relationship style scores were predictive of a higher IPB basic trust score, whereas higher fearful relationship style scores were predictive of lower basic trust scores. The adjusted indicates that nearly 30% of variability in IPB basic trust scores can be accounted for by differences in relationship style scores. Dismissing and preoccupied relationship style scores were not significant although they approached significance as predictors of IPB basic trust scores (p = .054; see Table 22 for details) Standard multiple regression on Autonomy. The second regression was conducted with the relationship style scores as predictor variables with IPB autonomy score as the criterion variable, yielding a significant result, F (4, 383) = 10.84, p < .001. As can be seen in Table 23, all four relationship style scores were significantly related to the criterion. Higher secure and dismissing relationship style scores were predictive of higher IPB scores whereas fearful and preoccupied relationship styles scores were predictive of lower IPB autonomy scores. According to the adjusted R^, 9.2% of variance in IPB autonomy scores could be explained by differences in relationship style scores. Standard multiple regression on Identity. The third set of regressions was conducted with the relationship style variables entered as the predictors and the IPB scale of identity as the criterion. This yielded a significant regression, F (8, 334) = 11.39, p < .001, with the 39 Identity Processing Style and Sex-Role Identity relationship styles of secure, fearful, and preoccupied contributing significantly to the regression equation. The secure relationship style was predictive of higher IPB identity scores whereas fearful and preoccupied were predictive of lower IPB identity scores. The adjusted indicates that 20% of variance in IPB identity scores was attributable to differences in relationship style scores (see Table 24 for details). Standard multiple regression on Intimacy. The fourth regression was conducted with the relationship style scores as the predictor variables and the IPB score for intimacy as the criterion variable. The regression equation was significant, F (8, 328) = 18.86, p < .001. Higher secure relationship style scores were predictive of higher IPB intimacy scores whereas higher fearful relationship style scores were predictive of lower IPB intimacy scores. The adjusted R^ indicated that 26.6% of variance in IPB intimacy scores was accounted for by differences in relationship style scores (see Table 25 for details). Standard multiple regression on Generativity. The final standard regression was performed with the relationship styles as predictor variables and the IPB scores for generativity as the criterion; this regression was also significant, F (4, 383) = 8.51, p < .001. Higher secure and dismissing relationship style scores were predictive of high IPB generativity scores whereas the fearful relationship style scores were predictive of lower IPB generativity scores. The adjusted R^ indicates that 7.2% of variance in IPB generativity scores can be attributed to differences in relationship style scores (see Table 26 for details). 40 Identity Processing Style and Sex-Role Identity Table 2. Means and standard deviations fo r the multivariate interaction o f age groups and relationship style Variable Secure Fearful Preoccupied Dismissing Age group M SD n Young adult 3.223 .526 181 Middle adult 3.109 .537 128 Senior adult 2.912 .446 59 Young adult 2.760 .842 184 Middle adult 2.651 .784 131 Senior adult 2.724 .739 66 Young adult 2.772 .620 182 Middle adult 2.485 .630 131 Senior adult 2.404 .543 68 Young adult 3.365 .519 184 Middle adult 3.378 .553 129 Senior adult 3.429 .593 69 41 Identity Processing Style and Sex-Role Identity Table 3. Means and standard deviations fo r the secure relationship style as a function o f age, gender, and identity style. Age group gender ID style M SD n Young adult male Information 3235 496 23 Normative Z973 .577 22 Diffused 3262 246 21 Information 3229 .502 38 Normative 3T52 .446 29 Diffused 3251 239 43 Information 3264 .557 22 Normative Z941 279 17 Diffused 3.158 223 19 Information Z900 293 16 Normative 3.057 .545 21 Diffused 3.300 .494 28 Information 3125 225 8 Normative 3.125 .413 8 Diffused 2.923 265 13 Information 2256 284 9 Normative 2.800 276 6 Diffused 2208 .514 13 female Middle adult male female Late adult male female Note: N = 3 6 5 42 Identity Processing Style and Sex-Role Identity Table 4 Means and standard deviations fo r the fearful relationship style as a function o f age, gender, and identity style. Age group gender ID style M SD n Young adult male Information 2804 ^2 7 23 Normative 2.591 .915 22 Diffused 2.667 .946 21 Information 2579 ^3 0 38 Normative 3IW6 .914 29 Diffused 2808 .763 43 Information 2682 j3 9 22 Normative 2.971 .712 17 Diffused 2632 .658 19 Information 2.594 .903 16 Normative 2.774 .770 21 Diffused 2 J38 ji76 28 Information 2.472 .824 9 Normative 2656 .916 8 Diffused 2.904 .650 13 Information 2909 J8 5 11 Normative 2821 .641 7 Diffused 2.641 .730 16 female Middle adult male female Late adult male female Note: N = 378 43 Identity Processing Style and Sex-Role Identity Table 5 Means and standard deviations fo r the preoccupied relationship style as a Junction o f age, gender, and identity style. Age group gender ID style M SD n Young adult male Information 2.446 .419 23 Normative 2.682 .524 22 Diffused 2.714 .704 21 Information 2816 .657 38 Normative 2888 .643 29 Diffused 2872 .660 43 Information 2.500 .650 22 Normative 2206 j3 5 17 Diffused 2.500 .666 19 Information 2.703 Normative 2583 .555 21 Diffused 2.455 .581 28 Information 2.417 .545 9 Normative 2.281 .452 8 Diffused 2.643 .507 14 Information 2681 .462 11 Normative 2.187 .776 8 Diffused 2 J3 4 J92 16 female Middle adult male female Late adult male female Note: N = 378 16 44 Identity Processing Style and Sex-Role Identity Table 6 Means and standard deviations fo r the dismissing relationship style as a function o f age, gender, and identity style. Age group gender ID style M SD n Young adult male Information 1287 J28 23 Normative 3.200 .390 22 Diffused 3.229 .511 21 Information 3 J52 .521 38 Normative 3.607 .491 29 Diffused 1367 J60 43 Information 1264 .631 22 Normative 3J53 ^36 17 Diffused 3.316 j9 0 19 Information 3.700 .593 16 Normative 3.505 .472 21 Diffused 1264 Information 1356 Normative 3.475 Diffused 1493 .580 15 Information 3.7545 .537 11 Normative 3.175 .506 8 Diffused 3.312 .619 16 female Middle adult male female Late adult male female Note: 379 28 .371 9 8 45 Identity Processing Style and Sex-Role Identity Table 7 Means and standard deviations fo r the trust psychosocial balance scale as a function o f age, gender, and identity style. Age group gender ID style M SD n Young adult male Information 56.346 4.757 26 Normative 55.905 7.272 21 Diffused 56.200 6366 20 Information 57.437 6T22 32 Normative 54.107 7.181 28 Diffused 56381 ^686 42 Information 57.666 7.783 24 Normative 54.411 7390 17 Diffused 57.722 5323 18 Information 56.769 7393 13 Normative 58.090 6.217 22 Diffused 60.307 6.417 26 Information 57.555 3.644 9 Normative 68.625 6TK9 8 Diffused 55.214 4.995 14 Information 56.400 3334 10 Normative 57.400 2302 5 Diffused 54386 6.091 14 female Middle adult male female Late adult male female Note: N = 364 46 Identity Processing Style and Sex-Role Identity Table 8 Means and standard deviations fo r the autonomy psychosocial balance scale as a function o f age, gender, and identity style. Age group gender ID style M SD n Young adult male Information 51.077 4.307 26 Normative 49.952 4.318 21 Diffused 48.050 4.260 20 Information 49.437 4.557 32 Normative 50.535 5436 28 Diffused 49.047 4.958 42 Information 49.500 3.297 24 Normative 48.882 5.500 17 Diffused 49.500 3.417 18 Information 47.692 5.006 13 Normative 48.773 4.385 22 Diffused 49.461 3.901 26 Information 48.625 2.774 8 Normative 52333 5^38 6 Diffused 49.067 3.058 15 Information 48.800 3.676 10 Normative 49.667 4.761 6 Diffused 49.083 3.175 12 female Middle adult male female Late adult male female Note: N = 362 47 Identity Processing Style and Sex-Role Identity Table 9 Means and standard deviations fo r the identity psychosocial balance scale as a function o f age, gender, and identity style. Age group gender ID style M SD n Young adult male Information 53308 5.221 26 Normative 52.285 7.830 21 Diffused 54.500 5.394 20 Information 54.687 6.402 32 Normative 51642 6395 28 Diffused 52357 63234 42 Information 54.583 5.694 24 Normative 51.764 6.581 17 Diffused 57.000 6.164 18 Information 53.000 6.976 13 Normative 51818 6380 22 Diffused 55.615 1283 26 Information 5L889 6.604 9 Normative 58.143 :L078 7 Diffused 52^43 6.021 14 Information 51.182 2.401 11 Normative 55333 7.840 6 Diffused 52.000 4.378 13 female Middle adult male female Late adult male female Note: N = 370 48 Identity Processing Style and Sex-Role Identity Table 10 Means and standard deviations fo r the intimacy psychosocial balance scale as a function o f age, gender, and identity style. Age group gender ID style M SD n Young adult male Information 55.923 &172 26 Normative 55.761 7.375 21 Diffused 57.200 7.522 20 Information 59.656 6.057 32 Normative 58^43 8.301 28 Diffused 58.286 7.123 42 Information 55.833 8380 24 Normative 51.235 8.452 17 Diffused 57.889 7.070 18 Information 56.307 IL219 13 Normative 59.000 5.380 22 Diffused 59.231 8.608 26 Information 57.222 2.991 9 Normative 5T333 6363 6 Diffused 54.643 6.046 14 Information 59.100 6339 10 Normative 55.800 4.604 5 Diffused 56.357 8.111 14 female Middle adult male female Late adult male female Note: N = 365 49 Identity Processing Style and Sex-Role Identity Table 11 Means and standard deviations fo r the generativity psychosocial balance scale as a function o f age, gender, and identity style. Age group gender ID style M SD n Young adult male Information 57.807 &209 26 Normative 57.190 5.537 21 Diffused 57.750 6^48 20 Information 6Ü625 5.493 32 Normative 61.285 4.965 28 Diffused 57.548 7.068 42 Information 60.667 5.924 24 Normative 59.294 5.565 17 Diffused 6L389 Ï3 9 2 18 Information 60.615 7.309 13 Normative 61.045 6.191 22 Diffused 62.077 5.066 26 Information 5&778 3.701 9 Normative 61.500 2.928 8 Diffused 59.600 5.591 15 Information 62.545 4.865 11 Normative 60.429 4.315 7 Diffused 62.143 4.801 14 female Middle adult male female Late adults male female Note: N = 377 50 Identity Processing Style and Sex-Role Identity Table 12 Means and standard deviations fo r the secure relationship style as a function o f Bern class and gender (for two age groups only) Gender Bern class M SD n Male Masculine 3.109 .463 46 Feminine 3.043 .472 14 Androgynous 3343 .587 35 Undifferentiated 3.035 .567 34 Masculine 3.065 352 34 Feminine 3.272 .461 58 Androgynous 3343 323 49 Undifferentiated 3.021 358 39 Female Note: N = 367 51 Identity Processing Style and Sex-Role Identity Table 13 Means and standard deviations fo r the fearful relationship style as a function o f Bern class and gender (for two age groups only) Gender Bem class M SD n Male Masculine :1878 .689 47 Feminine 2.634 ^28 15 Androgynous 2336 J8 8 35 Undifferentiated 2.907 .772 35 Masculine 2.772 .801 34 Feminine 2.605 .884 62 Androgynous 2536 .776 49 Undifferentiated 3.079 .820 38 Female Note: N = 379 52 Identity Processing Style and Sex-Role Identity Table 14 Means and standard deviations fo r the preoccupied relationship style as a function o f Bem class and gender (for two age groups only) Gender Bem class M SD n Male Masculine 2469 .547 48 Feminine Z350 .611 15 Androgynous 2.479 .654 35 Undifferentiated 2.693 j3 2 35 Masculine 2 328 J3 9 32 Feminine 2.823 .647 62 Androgynous 2823 .660 48 Undifferentiated 2.901 .467 38 Female Note: N - 3 7 9 53 Identity Processing Style and Sex-Role Identity Table 15 Means and standard deviations fo r the dismissing relationship style as a function o f Bem class and gender (for two age groups only) Gender Bem class M SD n Male Masculine 3345 .475 47 Feminine 3.413 .456 15 Androgynous 3.211 .619 35 Undifferentiated 3300 .492 34 Masculine 1.618 384 33 Feminine 3348 .489 62 Androgynous 3.449 .463 49 Undifferentiated 1568 323 38 Female Note; N = 380 54 Identity Processing Style and Sex-Role Identity Table 16 Means and standard deviations fo r the trust psychosocial balance scale as a function o f gender and sex-role (Bern) class Gender Bem class M SD n Male Masculine 54.750 5^02 52 Feminine 58.150 5.566 20 Androgynous 60.000 5.563 46 Undifferentiated 54.390 6.636 41 Total 56.604 6.310 159 Masculine 54.156 8402 32 Feminine 57.243 5.661 74 Androgynous 60.145 5.024 55 Undifferentiated 52.930 5.444 43 Total 56.632 6.519 204 Masculine 54.524 6.712 84 Feminine 57.436 5.623 94 Androgynous 60.079 5.293 101 Undifferentiated 5 3 jü 3 6063 84 Total 56.620 6.149 363 Female Total 55 Identity Processing Style and Sex-Role Identity Table 17 Means and standard deviations for the autonomy psychosocial balance scale as a function o f gender and sex-role (Bem) class Gender Bem class M SD n Male Masculine 50.370 4.015 54 Feminine 47.900 4.038 20 Androgynous 51.174 3.996 46 Undifferentiated 48.051 3 J69 39 Total 49.723 4.147 159 Masculine 51.061 4.690 33 Feminine 47.605 4.403 76 Androgynous 51.036 3.671 55 Undifferentiated 47.605 4.240 38 Total 49.104 .L538 202 Masculine 50.632 4.270 87 Feminine 47.667 4.311 96 Androgynous 51.100 3.804 101 Undifferentiated 47.831 3.988 77 Total 49.377 4 J7 5 361 Female Total 56 Identity Processing Style and Sex-Role Identity Table 18 Means and standard deviations fo r the identity psychosocial balance scale as a function o f gender and sex-role (Bem) class Gender Bem class M SD n Male Masculine 55.204 5.475 54 Feminine 52.800 4.640 20 Androgynous 54.867 7.175 45 Undifferentiated 51.902 5.843 41 Total 53.962 6.117 160 Masculine 52.471 7.659 35 Feminine 53.000 5.799 77 Androgynous 56.071 5.647 56 Undifferentiated 50.195 5.330 41 Total 53.182 6.315 209 Masculine 54.124 &522 89 Feminine 52.959 5^58 97 Androgynous 55.535 6J6 8 101 Undifferentiated 51.049 5.624 82 Total 53.520 6.234 369 Female Total 57 Identity Processing Style and Sex-Role Identity Table 19 Means and standard deviations fo r the intimacy psychosocial balance scale as a function o f gender and sex-role (Bem) class Gender Bem class M SD n Male Masculine 55.453 (i888 53 Feminine 57.950 6.984 20 Androgynous 59.222 6.557 45 Undifferentiated 51.854 7.505 41 Total 55.906 7.464 159 Masculine 55.771 8268 35 Feminine 59.000 &966 76 Androgynous 61.481 5.351 54 Undifferentiated 54.564 7.649 39 Total 58.255 7.476 204 Masculine 55.580 7.644 88 Feminine 58.781 6.947 96 Androgynous 60.454 6.006 99 Undifferentiated 53.175 7.650 80 Total 57226 7.551 363 Female Total 58 Identity Processing Style and Sex-Role Identity Table 20 Means and standard deviations fo r the generativity psychosocial balance scale as a function o f gender and sex-role (Bem) class Gender Bem class M SD n Male Masculine 58.907 4.954 54 Feminine 61.540 41285 20 Androgynous 61.021 5.451 47 U ndifferentiated 56.405 (x293 42 Total 59T84 5.682 163 Masculine 57.176 8.494 34 Feminine 60.900 4.793 80 Androgynous 63.821 5353 56 Undifferentiated 56.442 5.053 43 Total 60.174 6.345 213 Masculine 58.238 6.560 88 Feminine 61.010 4.680 100 Androgynous 62.544 5.551 103 Undifferentiated 56.423 5.666 85 Total 59.745 6.079 376 Female Total 59 Identity Processing Style and Sex-Role Identity Table 21 ANOVA results fo r IPB scales as a function o f Bem class df F-score p-value Basic trust ^330 23J33 <.001 .192 Autonomy 3,301 17.076 <.001 .145 Identity ^306 9.414 <.001 .084 Intimacy 3,303 16.996 <.001 .144 Generativity 3,309 19.852 <.001 .162 Dependent variable 60 Identity Processing Style and Sex-Role Identity Table 22 Standard Multiple Regression o f Relationship Styles on IPB Subscale o f Basic Trust (N - Variables B SE B P sP Secure RS 3.386* .582 .280 .059 Fearful RS -3.106* .375 -J9 8 T38 Preoccupied RS -.886 .459 -.089 .010 Dismissing RS .751 j3 3 .065 .003 R^ = .296 Adjusted R^ =.289 R = .545 *p < .001 Table 23 Standard Multiple Regression o f Relationship Styles on IPB Subscale o f Autonomy (N = 588; Variables B SE B P sP Secure RS 1.241* .446 .151 0.018 Fearful RS -.573* .287 -.108 0.009 Preoccupied RS -1.806** J52 -261 0.061 Dismissing RS 1.339* .409 .171 0.027 .102 Adjusted R^ = .092 R = .319 ■p < .05, **p < .001 61 Identity Processing Style and Sex-Role Identity Table 24 Standard Multiple Regression o f Relationship Styles on IPB Subscale o f Identity (N ■ Variables B SE B Secure RS 2.517* .604 ^12 ^36 Fearful RS -2.087* J8 9 -213 .060 Preoccupied RS -Z563* A ll -262 .060 Dismissing RS jl79 .554 .078 .005 sr R^ = .208 Adjusted R^ = .200 R = .456 *p < .001 Table 25 Standard Multiple Regression o f Relationship Styles on IPB Subscale o f Intimacy (N = Variables B SE B P sr Secure RS 4.160* .696 292 ^68 Fearful RS -3.406* .449 -.371 -.110 Preoccupied RS -.132 .549 -.011 -.000 Dismissing RS .824 ^38 .061 .003 R^ =.273 Adjusted R^ =.266 R = .523 *p < .001 62 Identity Processing Style and Sex-Role Identity Table 26 Standard Multiple Regression o f Relationship Styles on IPB Subscale o f Generativity (N Variables B SE B P s7^ Secure RS 2.108* .642 .180 .026 Fearful RS -1.343* .414 -.178 .025 Preoccupied RS -.764 .507 -.079 .005 Dismissing RS 1.880** j8 8 .169 .024 R^-= .082 Adjusted R^ = .072 R = .286 ^p = .001, **p = .002 63 Identity Processing Style and Sex-Role Identity DISCUSSION The goal of this study was to examine the possible associations among Berzonsky’s concept of ego identity style, Bem’s sex-role identity, and relationship style as conceived of by Bartholomew and Horowitz, within the broader context of Erikson’s psychosocial stages of development. As such, the project yielded mixed results, for although there were data that supported an association between sex-role identity and relationship style, the evidence did not support a connection between ego identity style and relationship style. Similarly, an association between sex-role identity and stage of psychosocial balance emerged, but an association between identity style processing and psychosocial balance was not supported. The hypotheses under examination were that there would be: differences in (1) relationship style and (2) psychosocial balance as a function of age, gender, and identity style category; differences in (3) relationship style and (4) psychosocial balance as a function of age, gender, and sex-role identity; (5) differences in identity style commitment scores as a function of sex-role identity; and, (6) that scores on the four relationship style scales (secure, fearful, preoccupied, and avoidant) predict scores on the psychosocial balance scales. These hypotheses will be discussed from the perspective of the support each garnered from the evidence presented in the results section. Hypothesis 1 The first hypothesis stated that differences in relationship style would vary as a function of age, gender, and ego identity style, with the latter concept a form of socialcognitive processing developed by Berzonsky (1989) as a dynamic alternative to the static concept of ego identity status conceived of by Marcia (1966). This hypothesis arose out of 64 Identity Processing Style and Sex-Role Identity earlier findings by Orlofsky, Marcia, and Lesser (1973) and Fitch and Adams (1983) indicating that the more advanced the individual’s identity status (e.g., achieved status), the greater that individual’s capacity for intimacy. Thus, it was presumed that a similar connection would be found between identity style and the capacity for intimacy in which individuals with a more advanced identity style category (e.g., informational) would demonstrate a greater capacity for intimacy (as measured by the Relationship Style Questionnaire; Bartholomew & Horowitz, 1991). However, this hypothesis was not supported since an analysis of variance found no significant differences in mean relationship style scores as a function of identity style category. The failure to find support for the first hypothesis was puzzling in light of prior research (e.g., Orlofsky, Marcia, & Lesser, 1973; Fitch & Adams, 1983), as well as in light of Erikson’s (1959; 1994) assertion that true intimacy with self or other was dependent upon, and a natural result of, the successful development of ego identity attained through self­ exploration. One possible explanation for the paucity of support for the first hypothesis is that earlier work supporting the ego identity - intimacy link employed the concept of ego identity as a status and not as a style. Identity status is viewed as a trait that is stable over time, whereas identity style might be more accurately seen as a state that is subject to change over time. The importance of this consideration is that the Relationship Style Questionnaire developed by Bartholomew and Horowitz (1994) is predicated on the concept of the attachment style that individuals develop in infancy, which is considered stable over the lifespan. Hence, two measures that purport to assess stable traits are more likely to yield evidence of a relationship between the two constructs (where one exists) than two measures wherein one accesses a stable trait and the other does not. In other words, because identity 65 Identity Processing Style and Sex-Role Identity style describes a social-cognitive processing style that can change over relatively short periods of time (Berzonsky, 1992), there may be no reliable manner for it to map successfully on to a static relationship style as conceived of by Bartholomew and Horowitz (1994). Berzonsky assumed that situational variables, past or potential consequences, and personal preferences constituted motivational factors in the individual’s determination of self-relevant information processing style. As stated by Berzonsky (1994, p. 780), “virtually all normal individuals are capable of employing all three of the social-cognitive processing strategies [italics original]’’. Because of this fluidity, identity style may be too flexible a concept to associate with a stable measure of intimacy. Another possible explanation for the failure to find support for the first hypothesis is that previous research connecting identity and intimacy relied on the Intimacy Interview developed by Orlofsky, Marcia, and Lesser (1973), which was based on Erikson’s concept of intimacy. Erikson’s concept of intimacy is based upon three separate criteria that consist of demonstrable relationship skills or behaviors (see Eitch & Adams, 1983) and not on any underlying attachment style arising from infancy, as illustrated in the two-dimensional model developed by Bartholomew and Horowitz (1994). Judging by the findings in the present research, it appears that differing conceptualizations of intimacy do not yield consistent results when examined as a function of ego identity. Hypothesis 2 The second hypothesis stated that measures of psychosocial balance would vary as a function of age, gender, and identity style category. Although significant multivariate interactions emerged as a function of both age and gender, there were no significant findings 66 Identity Processing Style and Sex-Role Identity as a function of identity style category. Given that Erikson (1959; 1994) suggested that individuals who have successfully resolved the identity/role confusion crisis will have done so through a process of self-exploration, it was assumed that in the context of the current project, a clear connection should have emerged between the more achieved identity style category (i.e., informational identity style) and higher scores for psychosocial balance. A possible explanation for a lack of significant results supporting the second hypothesis might again be found in Berzonsky’s concept of identity style. If ego identity style is conceived of as a means of soeial-eognitive processing, which individuals can select according to situational criteria, there is reason to expect that identity style might not be assoeiated with psychosocial balance. Although Erikson assumed that individuals could successfully resolve earlier developmental crises later in life (e.g., positively resolve a trust crisis as an adult), there is nothing in the literature to indicate that one would ever reverse this process. For example, once an individual has positively resolved the identity/role eonfusion crisis, there is no suggestion that the individual would then revert back to an unresolved state in which identity was still in question. Yet, Berzonsky suggested that an individual with an informational identity style could cease their adaptivity and become entrenched in their social-cognitive proeessing strategies to the extent that their identity style shifts to a normative style in which they become rigid adherents to their own authority. Hypothesis 3 The third hypothesis addressed the question of whether there are differences in relationship style as a function of gender versus sex-role identity (Bem class). Early attachment studies concluded that there were no gender differences in romantic attachment 67 Identity Processing Style and Sex-Role Identity style (e.g., Hazen & Shaver, 1987), which was consistent with the paucity of observable gender differences in childhood attachment (e.g., Ainsworth, Blehar, Waters, & Wells, 1978). Yet, in recent years, researchers have proposed that there may be gender differences in the quality of one’s emotional attachments. For example, a prevalent assumption is that men will more often choose to be more affectively distant and dismissing compared to women (Bem, 1993). The stereotype that men tend to be more distant in relationships has received empirical support, particularly where romantic relationships are concerned (Scharfe & Bartholomew, 1994). Later research using more continuous measures of attachment, such as the four-category model employed in the current project, provided evidence suggesting that men were significantly more dismissing in their relationship style compared to women (Bartholomew & Horowitz, 1991). Similarly, Bartholomew and Horowitz (1991) found that, compared to men, women tended to endorse items on the Relationship Styles Questionnaire indicating a preoccupied relationship style. Finally, Schmitt et al. (2003) found evidence supporting the notion that men are generally more dismissive in their relationship style (again using the four-category rating system of Bartholomew and Horowitz) across a wide array of ethnieities and cultures. The results from the current project are in agreement with some of the above findings, but contradict others. First, the present results support the notion that women are more preoccupied in relationships than men. Females had mean preoccupied relationship style scores that were significantly higher than the mean scores of their male counterparts when categorized in two of the sex-role identity categories: androgynous and feminine (but not masculine or undifferentiated). However, unlike previous research suggesting that men are more dismissing in relationships, the present results point to opposite conclusions suggesting 68 Identity Processing Style and Sex-Role Identity that, in fact, women are more dismissing in their relationship style compared to men. Again, this finding is in conjunction with women’s sex-role identity, wherein females had significantly higher mean scores for the dismissing relationship style than males did when the women could be categorized as undifferentiated or masculine (but not feminine or androgynous). Other results indicate that Bem class on its own plays an important role in relationship style scores. For example, subjects with a sex-role class of androgynous scored significantly higher for the secure relationship style compared to those with a sex-role identity of undifferentiated, but not when compared to those with a feminine or masculine sex-role identity. Finally, individuals with the Bem class of undifferentiated scored significantly higher for the fearful relationship style than those in the class of androgyny, but not compared to those with a masculine or feminine sex-role identity. The consistent opinion concerning the explanation for the relationship style differences in gender is based on perceptions of stereotypical gender-based responses to close or romantic relationships (Collins & Read, 1990). Primarily, men are seen as more strongly oriented towards autonomy and independence than are women (Kirkpatrick & Davis, 1994), whereas women are seen as more nurturing and expressive of the emotions associated with affiliation than are men (Bem, 1974). From a socio-cultural perspective, researchers have expected to find that genderbased differences in relationship style would emerge as a function of sex-role ideology (Schmitt et al., 2003). According to this line of thinking, individuals from cultures with more rigid, traditional structures of sex-role ideology would be more likely to adhere to stricter definitions of male-female comparisons associated with relationship styles. Yet, the 69 Identity Processing Style and Sex-Role Identity evidence indicates that quite the opposite is the case and, indeed, the more liberal and progressive the sex-role culture, the more pronounced the gender-based differences in relationship styles (Costa, Terracciano, & McCrae, 2001). Speculation about this finding rests on the idea that individuals from traditional cultures are far less likely to make acrossgender comparisons, whereas those from more progressive sex-role cultures are more likely to make such comparisons and in the process amplify gender differences in attachment orientations (Schmitt et al.., 2003). Thus, differences in relationship style have been viewed as both a function of gender as well as sex-role socialization; a perspective that is supported by the present research wherein significant interactions emerged between gender and sex-role identity. However, though the available literature contains evidence that can offer an explanation for the current finding that women tend to be more preoccupied in their relationship style compared to men, it does not offer any insight into the finding that women can also be more dismissing in their relationship style compared to men. Another relationship characteristic credited to women that might offer support for the current findings is the notion that women operate as “gatekeepers” for romantic relationships. For example, Kirkpatrick and Davis noted “the well-established observation that women are typically (although, of course, not necessarily) the maintainers and breakers of relationship” (1994, p. 510). It has also been noted that women initiate divorce twice as often as do men (Rice, 1994), suggesting that women experience greater optimism in their opportunities for future relationships compared to men. As has been stated in Bem’s gender schema theory (cited in Hoffman & Borders, 2001), women possess the affective skills essential to manage relationships (e.g., nurturance, willingness to compromise, sensitivity, and affection). 70 Identity Processing Style and Sex-Role Identity Perhaps it is women’s innate sense of possessing these skills that allows them the authority to steer their relationships in whichever direction they decide upon. In this manner, women then fall into two categories of relationship style: those who are preoccupied and those who are dismissive. The first category is marked by a higher degree of feminine sex-role characteristics versus masculine characteristics, whereas the second category is marked by a lower degree of feminine sex-role characteristics versus masculine characteristics. In other words, in the absence of high self-ascribed masculinity, women become the “maintainers” of relationships, but in the absence of high self-ascribed femininity, women become the “breakers” of relationships. Hypothesis 4 The fourth hypothesis examined the issue of psychosocial balance as a function of gender versus sex-role identity. Given the lack of available literature addressing this question, this part of the research is completely exploratory and there were few if any assumptions concerning its outcome. Although both gender and Bem class were significantly related to psychosocial balance, sex-role identity demonstrated the greater association. Concerning gender, only the psychosocial stage of intimacy was significantly associated, with females having the higher mean scores compared to males. Given the reports from research into gender differences in relationship capacities in which there is a consistent observation that women exhibit a greater capacity for intimacy than men, this result is not surprising. As for Bem class, all psychosocial stages proved significant, though not in a uniform manner. For example, although the sex-role identity class of androgyny had the highest mean scores on all psychosocial stages, androgyny was significantly greater than 71 Identity Processing Style and Sex-Role Identity the other three Bem classes (masculine, feminine, and undifferentiated) only in the case of basic trust versus mistrust. Androgyny did not significantly surpass masculinity in the stages of autonomy and identity, nor did it significantly surpass femininity in the stages of intimacy and generativity. Bem’s studies that led to the development of the Bem Sex-Role Inventory were directed more towards the exploration of androgyny as opposed to masculinity and femininity per se (Bem, 1974). She envisioned androgyny as both high masculinity and high femininity present within one individual whereby that individual employs the appropriate trait, whether masculine or feminine, according to circumstances rather than in accordance with a gender schema (Bem, 1977). In this sense, androgynous individuals are equally capable of employing traits of instrumentality as they are traits of expressiveness and communality. Androgyny, then, could be seen as the natural result of adult development in which the individual comes to express a wholeness of personality and capacity, akin to Freud’s claim that a normal well-adjusted adult should direct his or her energy towards love and work (cited in Erikson, 1959; 1980). In Bem (1974, p. 156) Freud’s sentiments are paraphrased nicely when she claims that the central characteristic of masculinity is a “cognitive focus on getting the job done’’, whereas the central characteristic for femininity is “affective concern for the welfare of others”. When these two sets of traits are combined, as they are in androgynous individuals, the outcome is a person with satisfactory adult capacities for “love” and “work”. In other words, androgynous individuals tend to score higher on a measure of psychosocial balance because they have positively resolved the Briksonian stages in a fashion thought to be required for a person to be able to lead a fruitful adult life (Erikson, 1959; 1980). 72 Identity Processing Style and Sex-Role Identity Note, however, that some of Erikson’s stages appear to represent one side of the masculine - feminine dichotomy more than the other. For example, androgynous individuals did not score significantly higher in the psychosocial stages of autonomy and identity compared to those with a masculine Bem category. According to Choi and Fuqua (2003), those who are sex-typed as either masculine or feminine tend to exhibit traits that conform to social expectations. If it is accepted that autonomy and identity are comparable to independence and individuality, traits often attributed to the masculine sex-role, then it is likely that individuals that can be classified as masculine would tend to endorse items on a measure of psychosocial balance that pertain to those qualities. In a similar manner, androgynous individuals did not score higher on the psychosocial stages of intimacy and generativity compared to those who could be classified as feminine. That is, intimacy and generativity are psychosocial stages marked by involvement with and caring for others (Erikson, 1959; 1980); traits/behaviors that are remarkably similar to the Bem femininity characteristics that include “affectionate”, “compassionate”, “sensitive to the needs of others”, and “loves children” (Bem, 1974). Hypothesis 5 The fifth hypothesis concerned an exploration of the commitment subscale of the Identity Style Questionnaire. Of primary interest was the question of whether sex-role identity would prove to be more strongly associated to commitment than was gender. This hypothesis was supported. The sex-role identity class of androgyny was significantly related to identity style commitment to a greater degree than the other three Bem classes (i.e.. 73 Identity Processing Style and Sex-Role Identity feminine, masculine, and undifferentiated), whereas the undifferentiated class scored significantly lower than the other three classes. It is difficult to explain these results, given that there is no prior literature looking at gender differences in the Identity Style Questionnaire commitment subscale, let alone an examination of sex-role identity differences. It is important to remember that the commitment subscale is a measure of one’s commitment to one’s identity style as opposed to commitment in the broader sense (e.g., relationship or career commitments). Indeed, these results appear contradictory since Berzonsky (1992) notes that a high degree of identity style commitment has been observed to suppress an informational identity style, one that has as its operating principle the tendency to respond to eireumstanees by weighing options instead of acting in a patterned manner according to the tenets of one’s identity style. In other words, an individual high for commitment would likely act in accordance to their identity style regardless of the situation due to what Berzonsky refers to as “structural consolidation of one’s self-theory’’ (1992, p. 773). The reason, then, that the results seem contradictory is that androgynous individuals, like those with an informational identity style, are also expected to respond in accordance with situational demands rather than in a patterned fashion dictated by their gender schema (Bem, 1974). Based on the comparison between the informational identity style and the Bem class of androgyny, it might have been expected that individuals classed as androgynous would have measured lower for the commitment subscale. An alternative explanation for results indicating that a measure of androgyny is associated with higher commitment scores, and a sex-role category of undifferentiated is associated with lower commitment scores, rests on the make-up of the commitment scale itself. Although the commitment subseale relates to identity style adherence, items on the 74 Identity Processing Style and Sex-Role Identity subscale refer to broader commitment issues, such as religion, employment, sets of values, and one’s future. Thus, it is conceivable that these items might be interpreted by the individual in the context of a broader definition of commitment that could be incorporated into an individual’s self-schema. Bem (1987) commented about how sex-typed social information constitutes an important aspect of one’s “self-concept”, whereas Berzonsky described self-identity as a “self-constructed theory” (Berzonsky, 1992, p. 771). Given that the commitment subscale, though not an identity style in itself, is still a component of identity style, it is possible that these two identity styles (i.e., ego and sex-role) map on to similar attributes. If this is so, then it might be expected that those who measure on the Bem Sex-Role Inventory as undifferentiated may do so in part because they have been unsuccessful in realizing a clear or distinctive self-concept or self-constructed theory. In other words, for the undifferentiated individual, there is little to commit to. Conversely, individuals who strongly endorse both masculine and feminine characteristies (i.e., androgynous) might be considerably more conseious of and committed to a rich and effective self-theory governing cognition, affect, and motivation. Hypothesis 6 The final hypothesis centered on the questions of whether one’s relationship style would be predictive of one’s level of psychosocial balance. To explore this question, a series of standard regressions were conducted for each of the five Eriksonian stages of interest. With basic trust versus mistrust as the criterion, both the secure and fearful relationship styles proved signifieant predictors though in opposite directions, with secure predicting an increase in the measure of basic trust and fearful predicting a decrease in basic trust. This 75 Identity Processing Style and Sex-Role Identity result is certainly what one might expect judging from the model for adult attachment as conceived of by Batholomew and Horowitz (1991). Erikson (1959; 1980) described the first psychosocial crisis as the judgment of whether the individual’s world (and those in it) is safe or not, whereas Batholomew and Horowitz (1991) state that secure individuals see others as trustworthy and fearful individuals do not. The results concerning the autonomy versus shame and doubt crisis are straightforward. The secure and dismissing relationship styles predict an increase in a measure of autonomy, whereas the fearful and preoccupied styles predict a decrease. The secure relationship style, as proposed by Batholomew and Horowitz (1991, p. 227), is described as “comfortable with intimacy and autonomy”, suggesting that the secure individual has met Erikson’s (1959; 1980) criteria for positively resolving the second psychosocial crisis, which is a positive resolution to the first crisis (basic trust). In a similar fashion, it seems apparent that fearful individuals, after experiencing a negative resolution to the basic trust crisis, are not equipped to positively resolve the autonomy crisis. Similarly, it is quite plain as to how the preoccupied relationship style is negatively related to autonomy since Batholomew and Horowitz (1991) describe the preoccupied style as one in which the individual seeks out relationships with others in an attempt to satisfy a need for dependency. On the face of Batholomew and Horowitz’s description of the dismissing style, which is one of independence and invulnerability, one would logically assume that those with a dismissing style would endorse items that measure higher autonomy, if you assume that autonomy and independence are synonymous. As expected, those with a dismissing style tend to avoid relations with others, perhaps not so much as a result of innate self-sufficiency, but out of a wish to avoid being disappointed and hurt by others (Batholomew and Horowitz, 1991). 76 Identity Processing Style and Sex-Role Identity Thus, the independence that Bartholomew and Horowitz ascribe to dismissing individuals are actually synonymous with autonomy in they are indicative of a self-protective response to intimacy that Batholomew and Horowitz labeled counter-dependence. The results indicated that the secure relationship style scores positively predicted scores on the identity versus role confusion scale. In contrast, scores on the preoccupied and fearful style scales negatively predicted the scores on the identity scale. The positive relationship between the secure relationship style and the criterion is hardly surprising given that the secure style is positively predictive of all five psychosocial stages under examination. Erikson (1959; 1980) repeatedly stressed that the positive resolution of successive stages was largely dependent upon the successful resolution of earlier stages; in this manner, individuals with a secure relationship style appear to demonstrate Erikson’s concept since they are consistent in their endorsement of items that reflect higher psychosocial balance. In contrast, the fearful relationship style was negatively associated with the identity versus role confusion stage, just as it is with all five psychosocial stages. Those with a fearful relationship style are consistent in their endorsement of items reflecting lower psychosocial balance and the negative resolution of successive Eriksonian stages. Finally, the preoccupied style was also negatively predictive of psychosocial balance, a result that possibly reflects the observation that those who score high on the preoccupied relationship style tend to have a lower sense of their own worth and consequently must depend on others to affirm their acceptance (Batholomew & Horowitz, 1991). The results connected to the intimacy versus isolation stage appear quite clear-cut with the secure relationship style positively predictive of intimacy and the fearful style predictive of lower intimacy scores. Once again, Erikson’s missive that the current stage’s 77 Identity Processing Style and Sex-Role Identity positive resolution depends a great deal on the positive resolution of the previous stage(s) is an important factor. Erikson (1959; 1980) stated that true melding with another person requires that one must first be a “true individual” (i.e., posses a clear ego identity) because the act of merging with another means the loss of the sense of self. As noted above, those with a secure style are more likely to have positively resolved the identity versus role confusion stage compared to those with a fearful relationship style. As well, Batholomew and Horowitz (1991, p. 227) state that the secure individual is “comfortable with intimacy”, whereas the fearful individual is “fearful of intimacy”. The final psychosocial stage, generativity versus stagnation, contains an interesting and puzzling finding. Although the positive relationship between the secure relationship style and the criterion and the negative relationship between the fearful style and the criterion seem straightforward, the positive predictive relationship between the dismissing style and generativity is unexpected. If generativity, as described by Erikson (1959; 1980), is predominantly the desire to create and then guide and nurture the subsequent generation, then it is difficult to comprehend the positive relationship between the dismissing style and generativity if the dismissing style is marked by a negative disposition toward others (Batholomew & Horowitz, 1991). However, one possible insight into this matter could be the observation that individuals who have fallen into the stagnation end of the generativity versus stagnation crisis have a tendency to cater to their own whims and desires as if they were their own and only child (Erikson, 1959; 1980). It is possible that a measure of psychosocial generativity might tap into this sentiment, as perverse as it may be in light of what actual generativity means. 78 Identity Processing Style and Sex-Role Identity Limitations o f the Current Study There are three limitations to the present research project. First, due to the nature of the methods used to gather data, which amount to a convenience sample, there is the hazard of self-seleetion for participation that attends all nonrandom samples. Although the sample was fairly large and the breakdown of participants by ego identity style and sex-role classes approximated those groupings found by Berzonsky (1989) and Bem (1974), a truly random sampling procedure would still be preferable. Second, though the sample population was large enough to generate significant results in a number of analyses, some of the analyses fell short of significance, possibly because of insufficient power due to a small sample size. Additionally, other analyses had to be amended because of too few individuals in some cells for SPSS to perform the desired calculations. It is possible that with the addition of another 70 or more participants, power might have been attained in the analyses concerning identity style. A potentially more serious limitation concerns the use of the Bem Sex-Role Inventory. The BSRI has been in use for a quarter century and has become the test of choice for researchers in the field of study of gender and sex-role socialization differences, with references to 795 separate studies involving the BSRI cited in the literature (Hoffman & Borders, 2001). However, its status as the gold standard by which other, similar instruments are measured has not made it invulnerable to criticism. Recently, Hoffman and Borders (2001) and Choi and Fuqua (2003) have authored rather damning critiques of the BSRI with the most salient objection to it focused on the item selection. In her initial research, Bem (1974) used undergraduate students to rate 400 adjectives on their desirability for men and women and then chose the 60 adjectives used in the BSRI (20 each for femininity. 79 Identity Processing Style and Sex-Role Identity masculinity, and neutral). In an effort to ascertain if this procedure would net similar results employing a modern undergraduate cohort, Hoffman and Borders (2001) and Choi and Fuqua (2003) followed Bem’s procedures and obtained very different results. For one thing, both sets of authors found very little agreement within their sample population on the relative femininity and masculinity of the BSRI items purported to measure these characteristics. Hoffman and Borders (2001) and Choi and Fuqua (2003) concluded that terms that were associated with one gender over the other in the early 1970’s did not seem to reflect current views on masculinity and femininity. Secondly, Choi and Fuqua (2003) conducted a factor analysis following Bem’s (1974) description of her research and found that the test items did not factor out in the same way as Bem’s results indicate. This finding led them to conclude that the supposed sex-related characteristics were not as factorally pure as Bem has alluded. For example, the BSRI items “feminine” and “masculine”, assumed by Bem to link closely with their respective sex-role identities, actually formed a separate factor that appeared related to biological as opposed to social concepts. The conclusion reached by Hoffman and Borders (2001) and Choi and Fuqua (2003) was that the BSRI may no longer measure masculinity and femininity as claimed, but instead assess “instrumentality” and “expressiveness”. Those findings highlight the complex issue associated with the use of Bem’s inventory. That is, Bem’s landmark research was conducted nearly 30 years ago when male and female sex-roles were more clear-cut and distinctive (albeit more restrictive as well) than they are now. Indeed, in recent years, Bem (1993; cited by Hoffman & Borders, 2001) has expressed a concern that our tendencies to use the lens of gender that shape our perceptions of society must be overcome in order to eliminate serious restrictions on both sexes. Is the 80 Identity Processing Style and Sex-Role Identity fact that recent studies have failed to find the same results as Bem an indictment of the BSRI or a reflection of the evolution of gender-socialization in which modern young men and women eschew the very notion of gender-based labeling, whether or not these labels are more often associated with one sex or the other? After all, in the present study, the sex-role classes broke down in a manner remarkably similar to that of Bem’s (1974) original norming population, even though more than half of our data set was collected from outside of the student subject pool. If these test items no longer reflect common male and female characteristics, what do they reflect and how is it that our sample yielded the expected frequencies in each class and by gender? Is it not possible that BSRI items labeled instrumentality and expressiveness by Hoffman and Borders (2001) and Choi and Fuqua (2003) are measuring the same (or a very similar) construct conceived of by Bem, but in today’s world, people resist placing gender on such items because they are, in fact, mindful of the same concern assoeiated with that practice as was Bem? Recommendations fo r Future Research The primary recommendation for future research would be a further attempt to isolate an association between ego identity style and adult romantic attachment. It is possible that no signifieant association between identity style and relationship style emerged in the present study because of either too small a sample size or test instruments that were incompatible. In the former case, a larger sample would obviously add power to the analyses, but effect sizes for the relevant analyses were too small to support the idea that increasing the sample size within the realm of practicality would have sufficed. In the latter case, instruments measuring identity style and relationship style that would map on to each other in a more 81 Identity Processing Style and Sex-Role Identity appropriate manner could possibly find the expected results. For example, in the current project, the measure for identity style that was employed had been shown to reveal changes in an individual’s identity style over relatively short periods of time, thus reducing the probability of finding stable associations between Berzonsky’s identity style and the Bartholomew and Horowitz construct of relationship style. However, it is possible that the dynamic ego identity style is simply too ephemeral to be matched to adult attachment style in the same manner as was the static ego identity status. Another recommendation would be a further exploration of the association between relationship style and psychosocial balance. Given the rather impressive proportions of variance in individual level of psychosocial balance explained in some of the regressions (e.g., nearly 30% in basic trust versus mistrust), an attempt to elaborate on the explanation of this phenomenon seems warranted. For example, the causal direction would be particularly interesting: does advanced psychosocial balance result in a secure relationship style or vice versa? Or is it simply a case where a positive resolution to the first crisis, basic trust versus mistrust, leads to further positive resolutions in the subsequent Eriksonian stages while simultaneously yielding a secure infant attachment, which in turn leads to a secure adult attachment? Such are the challenges to other researchers interested in lifespan development. Summary and Conclusion This project was an attempt to explore earlier findings suggesting a connection between ego identity, sex-role identity, and romantic relationships. As such, it met with mixed results, with no significant association between identity style and either relationship style or stage of psychosocial balance. However, a clear association emerged between sex- 82 Identity Processing Style and Sex-Role Identity role identity and both relationship style and psychosocial balance. There were also interesting results found in the analyses of the identity style eommitment subscale and in the prediction of Eriksonian stage from relationship style scores. Although the positive results from this project, most of which stem from associations with sex-role identity, might be called into question by those who have come to doubt the validity of the Bem Sex-Role Inventory, the findings comply with the theories concerning adult attachment, Eriksonian development, and gender socialization. Great strides have been made over the past four decades in an effort to erase social and personal imbalances created by notions of traditional gender differences; is it any surprise that a modern and welleducated research population would balk at gender-specific labeling of characteristics that ideally belong to both sexes? However, if “instrumentality” and “expressiveness” have come to replace “masculine” and “feminine” in the minds of researchers or those who fill out research surveys, does it really change the patterns of cognition, emotion, and behavior associated with adult romantic attachment? According to the Jungian concept of “sex-role convergence” (Hosley & Montemayor, 1997), most adults in middle life experience an emerging androgyny marked by increased “instrumentality” in women and increased “expressiveness” in men, characteristics which were strongly linked to masculinity and femininity, respectively. In other words, whether we refer to these domains as “instrumentality” and “expressiveness” or “masculine” and “feminine”, the connection to other aspects of one’s life, such as romantic attachment, is the same. The importance of this research is twofold. First, the finding that sex-role identity plays an important part in adult romantic relationship style has the potential for application in the field of marriage and relationship therapy. Based on the idea that, for example. 83 Identity Processing Style and Sex-Role Identity stereotypical masculine characteristics are less frequently associated with a secure relationship style in men, our findings might inform an approach to therapy designed to improve men’s long-term relationship stability and satisfaction by helping them develop sexrole androgyny or, at least, question a singular reliance on a masculine sex-role identity. At the least, the connection between the secure relationship style and the successful resolution of the Eriksonian stages of psychosocial development is a finding that should inform childraising practices, if it can be assumed that securely attached children become securely attached adults. Perhaps the most important aspect of this project was the finding that although ego identity style was not assoeiated with relationship style, sex-role identity definitely was. This leads to the temptation to speculate that where adult romantic attachment is concerned, sexrole identity, as a social-cognitive processing system, is more important than ego identity. In other words, sex-role identity is a more fundamental personal construct when romantic relationship issues are involved as compared to identity style or even gender. If so, then it is time to reexamine stereotypical notions about romantic relationships with respect to initiation, maintenance, and termination of these relationships. 84 Identity Processing Style and Sex-Role Identity REFERENCES Ainsworth, M. D., Blehar, M., Waters, E. & Wall, S. (1978). Patterns o f attachment: A psychological study o f the strange situation. Hillsdale, NJ: Erlbaum. Adams, G. R., Ryan, J. H., Hoffman, J. J., Dobson, W. R., & Nielson, E. C. (1985). Ego identity status conformity behavior and personality in late adolescence. Journal o f Personality and Social Psychology, 47, (5), 1091-1104. Adams, G. R. & Shea, J. A. (1979). The relationship between identity status, locus of control, and ego development. Journal o f Youth and Adolescence, 8, 81-89. Bartholomew, K. & Horowitz, L. M. (1991). Attachment style among young adults: A test of the four-category model. Journal o f Personality and Social Psychology, 61, (2), 226244. Bem, S. L. (1974). The measurement of psychological androgyny. Journal o f Consulting and Clinical Psychology, 42, 155 - 162. Bem, S. L. (1977). On the utility of alternative procedures for assessing psychological androgyny. Journal o f Consulting and Clinical Psychology, 45, 196 - 205 Bem, S. L. (1981). A manual fo r the Bem Sex Role Inventory. Palo Alto, CA: Consulting Psychological Press. Bem, S. L. (1987). Gender scheme theory and the romantic tradition. In P. Shaver & C. Hendrick (Eds.), Review o f personality and social psychology (Vol. 7, pp. 251 - 271). Newbury Park, CA: Sage. Bem, S. L. (1993). The lenses o f gender: Transforming the debate on sexual inequality. New Haven, CT: Yale University Press. 85 Identity Processing Style and Sex-Role Identity Berzonsky, M. D. (1989). Identity style: Conceptualization and measurement. Journal of Adolescent Research, 4, 268 - 282. Berzonsky, M. D. (1992). Identity style and coping strategies. Journal of Personality, 60, (4). 771-787. Berzonsky, M. D. (1993). Identity style, gender, and social-cognitive reasoning. Journal o f Adolescent Research, 8, (3), 289-296. Berzonsky, M. D. & Kuk, L. S. (2000). Identity status, identity, processing style, and the transition to university. Journal o f Adolescent Research, 15, (1), 81-98. Berzonsky, M. D. & Neimeyer, G. J. (1994). Ego identity status and identity processing orientation; The moderating role of commitment. Journal o f Research in Personality, 28, 425-435. Brennan, K. A. & Bosson, J. K. (1998). Attachment-style differences in attitudes toward and reaction to feedback from romantic partners: An exploration of the relational bases of self-esteem. Personality and Social Psychology Bulletin, 24, (7), 699-714. Choi, N. & Fuqua, D. R. (2003). The structure of the Bem Sex Role Inventory: A summary report of 23 validation studies. Educational and Psychological Measurement, 63, (5), 8 7 2 -8 8 7 . Clancy Dollinger, S. M. (1995). Identity styles and the five-factor model of personality. Journal o f Research in Personality, 29, 475-479. Collins, N. & Read, S. (1990). Adult attachment relationships, working models and relationship quality in dating couples. Journal o f Personality and Social Psychology, 58, 6 4 4 -6 8 3 . 86 Identity Processing Style and Sex-Role Identity Costa, P. T., Terracciano, A., & McCrae, R. R. (2001). Gender differences in personality traitsacross cultures: Robust and surprising findings. Journal o f Personality and Social Psychology, 81, 322 - 331. Domino, G. & Affonso, D. D. (1990). A personality measure of Erikson's life stages: The Inventory of Psychosocial Balance. Journal o f Personality Assessment, 54 (3 & 4), 576-588. Erikson, E. H. (1959, 1980). Identity and the life cycle^ New York: Norton. Erikson, E. H. (1964). Childhood and society. New York: Norton. Fitch, S. A. & Adams, G. R. (1983). Ego identity and intimacy status: Replication and extension. Developmental Psychology, 19 (6), 839-845. Eromm, E. (1968). The revolution o f hope: Toward a civilized technology. Toronto, ON: Harper-Row. Griffen, D. & Bartholomew, K. (1994). Models of the self and other: Fundamental dimensions underlying measures of adult attachment. Journal o f Personality and Social Psychology, 67 (3), 430-445. Hazen, C. & Shaver, P. R. (1987). Romantic love conceptualized as an attachment process. Journal o f Personality and Social Psychology, 52, 511 - 524. Hoffman, R. M. & Borders, L. D. (2001). Twenty-five years after the Bem Sex-Role Inventory: A reassessment and new issues regarding classification variability. Measurement and Evaluation in Counseling and Development, 54 (1), 39-56. Hosley, C. A. & Montemayor, R. (1997). Fathers and adolescents. In M. E. Lamb (Ed.). The Role o f the Father in Child Development. (5'^^ ed.). New York: Wiley. 87 Identity Processing Style and Sex-Role Identity John,0. P., Donahue, E. M., & Kentle, R. L. (1991). The Big Five Inventory - versions 4a and 54. Berkeley, CA: University of California, Berkeley, Institute of Personality and Social Research. Kirkpatrick, L. A. & Davis, K. E. (1994). Attachment style, gender, and relationship stability: A longitudinal analysis. Journal o f Personality and Social Psychology, 66, (3). 502512. Lenney, E. (1991). Sex roles: The measurement of masculinity, femininity, and androgyny. In J. P. Robinson, P. R. Shaver, & L. S. Wrightman (Eds.), Measures o f Personality and Social Psychological Attitudes (Vol. 1, pp. 573 - 660). San Diego, CA: Academic Press. Marcia, J. E. (1966). Development and validation of ego identity status. Journal o f Personality and Social Psychology, 3, 551-558. Maslow, A. H. (1954). Motivation and personality. New York: Harper & Row. Orlofsky, J. L., Marcia, J. E., & Lesser, I. M. (1973). Ego identity status and the intimacy versus isolation crisis of young adulthood. Journal o f Personality and Social Psychology, 27 (2), 211-219. Read, D., Adams, G. R., & Dobson, W. R. (1984). Ego-identity status, personality, and soeial-influence style. Journal o f Personality and Social Psychology, 46 (1), 169-177. Rice, J. K. (1994). Reconsidering research on divorce, family life cycle, and the meaning of family. Psychology o f Women Quarterly, 18, 559 - 584. Scarfe, E. & Bartholomew, K. (1994). Reliability and stability o f adult attachment patterns. Personal Relationships, 1, 23 - 43. 88 Identity Processing Style and Sex-Role Identity Schwartz, S. J., Mullis, R. L., Waterman, A. S., & Dunham, R. M. (2000). Ego identity status, identity style, and personal expressiveness. Journal o f Adolescent Research, 15 (4). 504-521. Schmitt, D. P., Alcalay, L., Allensworth, M., Allik, J., Ault, L., Austers, 1., Bennett, K. K., Blanchi, G., Boholst, P., Borg Cunen, M. A., Braekman, J., Brainerd Jr., E. G., Caral, L. G., Caron, G., Casullo, M. M., Cunningham, M., Daibo, 1., De Backer, C., De Souza, E., Diaz-Loving, R., Diniz, G., Durkin, K., Echegaray, M., Eremsoy, E., Euler, H. A., Falzon, R., Fisher, M. L., Foley, D., Fowler, R., Fry, D. P., Fry, S., Ghayur, M. A., Girl, V. N., Golden, D. L., Grammer, K., Grimaldi, L., Halberstadt, J., Haque, S., Herrera, D., Hertel, J., Hitchell, A., Hoffman, H., Hooper, D., Hradilekova, Z., Hudek-Keni-Evi, J., Huffcutt, A., Jaafar, J., Jankauskatie, M., Kabangu-Stahel, H., Kardom, 1., Khoury, B., Kwon, H., Laidra, K., Laireiter, A., Lakerveld, D., Lampert, A., Lauri, M., Lavalee, M., Lee, S. J., Leung, L. C., Locke, K., Locke, V., Luksik, L, Magaisa, 1., Marcinkeviciene, D., Mata, A., Mata, R., McCarthy, B., Mills, M. E., Mkhize, N. J., Moreira, J., Moreira, S., Moya, M., Munyae, M., Noller, P., Climat, H., Opre, A., Panayioyou, A., Petrovic, N., Polls, K., Popper, M., Poulimenou, M., P’Yatokha, V., Yaymond, M., Reips, U.-D., Reneau, S., Rivera-Aragon, S., Rowatt, W., Ruch, W., Rus, V. S., Safir, M., Salas, S., Sambataro, F., Sandnabba, K. N., Schleeter, R., Schulmeyer, M. K., Schütz, A., Scrimali, T., Shackelford, T. K., Sharan, M. B., Shaver, P. R., Sichona, F., Simonetti, F., Sineshaw, T., Sookdew, R., Speelman, T., Spyrou, S., Canan-Sumer, H., Sumer, N., Supekova, M.,Szlendeak, T., Taylor, R., Timmermans, B., Turner, A., Vandermassen, G., Vanhoomissen, T., Van Overwalle, F., Vanwesenbeeck, L, Vasey, P. 1., 89 Identity Processing Style and Sex-Role Identity Verissimo, J., Voracek, M., & Wan, W. W. N. (2003). Are men universally more dismissing than women? Gender differences in romantic attachment across 62 cultural regions. Personal Relationships, 10, 307 - 331. Schwartz, S. J., Mullis, R. L., Waterman, A. S., & Dunham, R. M. (2000). Ego identity status, identity style, and personal expressiveness; An empirical investigation of three convergent constructs. Journal o f Adolescent Research, 75(4), 5 4 0 -5 2 1 . Shaver, P. R., & Hazen, C. (1993). Adult romantic attachment: Theory and evidence. In D. Perlman & W. Jones (Eds.). Advances in Personal Relationships. (Vol. 4, 29-70), London: Jessica Kingsley. Streimatter, J. (1993). Identity status and identity style: A replication study. Journal o f Adolescence, 16, 211-215. Statistics Canada (n. d.). Age characteristics of the population of Prince George, British Columbia for the 2001 census. Retrieved September 16, 2003, from http://wwwl2.statcan.ca/english/profil01/Details/detailslpop.cfm Tabachnick, B. G., & Fidell, L. S. (2001). Using Multivariate Statistics, (4^^ ed.). Needham Heights, MA: Allyn & Bacon. Waterman, A. S. (1982). Identity development from adolescence to adulthood: An extension of theory and review of research. Developmental Psychology, 18, 341-358. White, J. M., Wampler, R. S., & Winn, K. I. (1998). The Identity Style Inventory: A revision with a sixth-grade reading level (ISI-6G). Journal o f Adolescent Research, 13, (2), 223 - 245. 90 Identity Processing Style and Sex-Role Identity Appendix A: Demographics Questionnaire The following demographic information is collected to help describe the sample 1. Date of birth;______________________ 2. Sex: male female ____ 3. Mother tongue:____________________ 4. Ethnicity:______ Aboriginal African-Canadian African)________ Caucasian Asian_______Black (other than Other_____________ 5. Marital Status:_________ married/common-law widowed separated single _________ divorced 6. Check your highest education level completed: Elementary school (please specify grade completed)___________ Secondary school (please specify grade completed)____________ High school diploma Trade/technical school (pleasespecify:_______________________ 1 Some college College diploma (please specify:____________________________ ) Some university University degree Other (please specify:_____________________________________1 7. Your occupation or previous occupation:_______________________________ full-time part-time retired currently unemployed 8. Number of children you have: 9. Ages of your children:______ 10. In general, how would you rate your health? Excellent Very Good Good Fair Poor 11. Compared to others your age, how would you rate your health? Excellent Very Good Good Fair Poor 91 Identity Processing Style and Sex-Role Identity Appendix B Informed Consent The psychological research you have been asked to participate in is governed by the Canadian Psychological Association’s regulations governing ethical care and respect for research participants. As research participants, you may be assured that every precaution will be taken to safeguard the security of the research materials, particularly with respect to issues of confidentiality. Your completed test forms may remain anonymous and do not have to have your name on them, but should you put your name on any of the test materials, it will be removed and replaced with a code number. In addition to myself, only my supervisor and a research assistant will have access to the test data. The information you supply will be used for research purposes only and will be stored in a secure, locked filing cabinet. It is important for you to note that you may withdraw from this research project at any time, even after you have returned the completed test forms to me, for any reason. The purpose of this research is to conduct an exploratory examination of a possible association between age and relationships, with the focus on looking at differences between age groups and the nature of their close relationships. I have read the statement above and understand my rights with respect to this research project. I am willing to participate, knowing that I may withdraw my participation at a later date, should I so choose. Name (please print): Signature: Date: 92 Identity Processing Style and Sex-Role Identity Appendix C: Bern Sex-Role Inventory Please circle a number between 1 and 5 that best reflects how much you agree or disagree with the sentence. There are no right or wrong answers. YOUR REACTION TO EACH QUESTION SHOULD BE YOUR ANSWER. 1 Strongly Disagree 2 Somewhat Disagree 3 Unsure Somewhat Agree Strongly Agree 1. I know what I helieve ahout religion. 1 2 3 4 2. I’ve spent a lot o f time thinking ahout what I should do with my life. I 2 3 4 3. I’m not sure what I’m doing with my life. 1 2 3 4 4. I act the way I do because of the values I was brought up with. I 2 3 4 5. I’ve spent a lot o f time reading and/or talking to others ahout religious ideas. I 2 3 4 6. When I talk to someone ahout a problem, I try to see their point of view. I 2 3 4 7. I know what I want to do with my future. I 2 3 4 8. I don’t worry about values ahead of time; I decide things as they happen. I 2 3 4 9. I’m not really sure what I helieve ahout religion. I 2 3 4 10. I was brought up to know what to work for. 1 2 3 4 11. I’m not sure which values I really hold. I 2 3 4 12. I know where the government and country should be going. I 2 3 4 13. If I don’t worry ahout my problems they usually work themselves out. I 2 3 4 14. I’m not sure what I want to do in the future. I 2 3 4 15. I feel that the work I do (or have done in the past) is right for me. I 2 3 4 16. I’ve spent a lot o f time reading ahout and/or trying to understand political issues. I 2 3 4 17. I’m not thinking ahout my future now - it’s still a long way off. I 2 3 4 18. I’ve spent a lot o f time talking to people to find a set of beliefs that works for me. I 2 3 4 19 I’ve never had any serious doubts ahout my religious beliefs. I 2 3 4 20. I’m not sure what job is right for me. I 2 3 4 21. I’ve known since I was young what I wanted to do. 1 2 3 4 22. I have a strong set of beliefs that I use when I make a decision. I 2 3 4 23. It’s better to have a firm set of beliefs than to he open to different ideas. I 2 3 4 24. When I have to make a decision, I wait as long as I can to see what will happen. I 2 3 4 25. When I have a problem, I do a lot of thinking to understand it. 1 2 3 4 26. It’s best to get advice from experts (preachers, doctors, lawyers, teachers) when I I 2 3 4 93 Identity Processing Style and Sex-Role Identity 94 27. I don’t take life too serious; I just enjoy it. 1 2 3 4 5 28. It’s better to have one set of values than to consider other value options. 1 2 3 4 5 29. I try not to think about or deal with problems as long as I can. 1 2 3 4 5 30. My problems can be interesting challenges. 1 2 3 4 5 31. 1 try to avoid problems that make me think. 1 2 3 4 5 32. Once I know how to solve a problem, 1 like to stick with it. 1 2 3 4 5 33. When I make decisions, I take a lot of time to think about my choices. 1 2 3 4 5 34. I like to deal with things the way my parents said I should. 1 2 3 4 5 35. I like to think through my problems and deal with them on my own. 1 2 3 4 5 36. When I ignore a potential problem, things usually work out. 1 2 3 4 5 37. When I have to make a big decision, 1 like to know as much as I can about it. 1 2 3 4 5 38. When I know a problem will cause me stress, I try to avoid it. 1 2 3 4 5 39. People need to be committed to as et of values to live a full life. 1 2 3 4 5 40. It’s best to get advice from friends or family when I have a problem. 1 2 3 4 5 Identity Processing Style and Sex-Role Identity 95 Appendix D: Relationship Style Questionnaire Please read each of the following statements and rate the extent to which you believe each statement best describes your feelings about close relationships. Not at all Somewhat Very much like me like me like me I find it difficult to depend on other people. 1 2 3 4 5 It is very important to me to feel independent. 1 2 3 4 5 I find it easy to get emotionally close to others. 1 2 3 4 5 I want to merge completely with another person. 1 2 3 4 5 I worry that will be hurt if I allow myself to become too close 1 2 3 4 5 I am comfortable without close emotional relationships. 1 2 3 4 5 I am not sure that I can always depend on others to be there 1 2 3 4 5 I want to be completely emotionally intimate with others. 1 2 3 4 5 9. I worry about being alone. 1 2 3 4 5 10. I am comfortable depending on other people. 1 2 3 4 5 11. I often worry that romantic partners don’t really love me. 1 2 3 4 5 12. I find it difficult to trust others completely. 1 2 3 4 5 13. I worry about others getting too close to me. 1 2 3 4 5 14. I want emotionally close relationships. 1 2 3 4 5 15. I am comfortable having other people depend on me. 1 2 3 4 5 16. I worry that others don’t value me as much as I value them. 1 2 3 4 5 17. People are never there when you need them. 1 2 3 4 5 18. My desire to merge completely sometimes scares people away. 1 2 3 4 5 19. It is very important to me to feel self-sufficient. 1 2 3 4 5 20 . I am nervous when anyone gets too close to me. 1 2 3 4 5 21 . I often worry that romantic partners won’t want to stay with 1 2 3 4 5 to others. when I need them. me. 22 . I prefer not to have other people depend on me. I 2 3 4 5 23. I worry about being abandoned. 1 2 3 4 5 24. I am somewhat uncomfortable being close to others. 1 2 3 4 5 25. I find that others are reluctant to get as close as I would like. 1 2 3 4 5 26. I prefer not to depend on others. 1 2 3 4 5 27. I know that others will be there when I need them. 1 2 3 4 5 28. I worry about having others not accept. 1 2 3 4 5 Identity Processing Style and Sex-Role Identity 29. Romantic partners often want me to be closer than I feel comfortable being. 30. I find it relatively easy to get close to others. 96 Identity Processing Style and Sex-Role Identity Appendix E; Identity Style Inventory - Grade 6 Reading Level Please circle a number between 1 and 5 that best reflects how much you agree or disagree with the sentence. There are no right or wrong answers. YOUR REACTION TO EACH QUESTION SHOULD BE YOUR ANSWER. 1 2 3 Strongly Disagree Somewhat Disagree Unsure 5 Somewhat Agree Strongly Agree 1. I know what I believe about religion. 1 2 3 4 5 2. I’ve spent a lot of time thinking about what I should do with my life. 1 2 3 4 5 3. I’m not sure what I’m doing with my life. 1 2 3 4 5 4. I act the way I do because of the values I was brought up with. 1 2 3 4 5 5. I’ve spent a lot of time reading and/or talking to others about religious ideas. 1 2 3 4 5 6. When I talk to someone about a problem, I try to see their point of view. 1 2 3 4 5 7. I know what I want to do with my future. 1 2 3 4 5 8. I don’t worry about values ahead o f time; I decide things as they happen. 1 2 3 4 5 9. I’m not really sure what I believe about religion. 1 2 3 4 5 10. I was brought up to know what to work for. 1 2 3 4 5 11. I’m not sure which values I really hold. 1 2 3 4 5 12. I know where the government and country should be going. 1 2 3 4 5 13. If I don’t worry about my problems they usually work themselves out. 1 2 3 4 5 14. I’m not sure what I want to do in the future. 1 2 3 4 5 15. I feel that the work I do (or have done in the past) is right for me. 1 2 3 4 5 16. I’ve spent a lot of time reading about and/or trying to understand political issues. 1 2 3 4 5 17. I’m not thinking about my future now - it’s still a long way off. 1 2 3 4 5 18. I’ve spent a lot of time talking to people to find a set of beliefs that works for me. 1 2 3 4 5 19 I’ve never had any serious doubts about my religious beliefs. 1 2 3 4 5 20. I’m not sure what job is right for me. 1 2 3 4 5 21. I’ve known since I was young what I wanted to do. 1 2 3 4 5 22. I have a strong set o f beliefs that I use when I make a decision. 1 2 3 4 5 23. It’s better to have a firm set o f beliefs than to be open to different ideas. 1 2 3 4 5 24. When I have to make a decision, I wait as long as I can to see what will happen. 1 2 3 4 5 25. When I have a problem, I do a lot of thinking to understand it. 1 2 3 4 5 26. It’s best to get advice from experts (preachers, doctors, lawyers, teachers) when I have I 2 3 4 5 a problem. 27. I don’t take life too serious; I just enjoy it. 1 2 3 4 5 97 Identity Processing Style and Sex-Role Identity 28. It’s better to have one set of values than to consider other value options. 1 2 3 4 5 29. I try not to think about or deal with problems as long as I can. 1 2 3 4 5 30. My problems can be interesting challenges. 1 2 3 4 5 31. 1 try to avoid problems that make me think. 1 2 3 4 5 32. Once 1 know how to solve a problem, 1 like to stick with it. 1 2 3 4 5 33. When 1 make decisions, 1 take a lot o f time to think about my choices. 1 2 3 4 5 34. 1 like to deal with things the way my parents said 1 should. 1 2 3 4 5 35. 1 like to think through my problems and deal with them on my own. 1 2 3 4 5 36. When 1 ignore a potential problem, things usually work out. 1 2 3 4 5 37. When 1 have to make a big decision, 1 like to know as much as1 can about it. 1 2 3 4 5 38. When 1 know a problem will cause me stress, 1 try to avoid it. 1 2 3 4 5 39. People need to be committed to as et of values to live a full life. 1 2 3 4 5 40. It’s best to get advice from friends or family when 1 have a problem. 1 2 3 4 5 98 Identity Processing Style and Sex-Role Identity 99 Appendix F: Inventory of Psychosocial Balance For each statement please indicate, on the answer sheet, whether you: A = Strongly agree B = Agree D = Disagree E = Strongly disagree C = Are uncertain There are no right or wrong answers - what you honestly think, is the right answer for you (CIRCLE ANSWER THAT YOU WISH TO SELECT). 1. Having friends is important to me. A B C D B 2. I value independence more than financial security. A B C D E 5. Sometimes I wonder who I really am. A B C D E 6. I have experienced some very close friendships. A B C D E 7. I derive-great pleasure in watching a child master a new skill. A B C D E 8. If I could relive my life, I would make few changes. A B C D E 9. I have confidence in my own abilities. A B C D E 10. Those who know me say I am stubhorn. A B C D E 13. When I was a teenager I rarely dated. A B C D E 14. There have been at least several people in my life with whom I have developed a A B C D E very close relationship. 15. 1 have many and varied interests. A B C D E 16. My religious or spiritual beliefs are stronger now than they have ever been. A B C D E 17. 1 can usually depend on others A B C D E 18. It is difficult for me to make up my mind. A B C D E 21. As an adolescent 1 was very shy. A B C D E 22. 1 often feel lonely even when there are others around me. A B C D E 23. My life is or has been a productive one. A B C D E 24. 1 think that certain groups or races of people are inferior to others. A B C D E 25. 1 sometimes have difficulties with what 1 see, hear and feel versus what A B C D E is real - seeing the world as others do. Identity Processing Style and Sex-Role Identity 100 26. When I do something, I do it as carefully as possible. A B C D E 29. It makes me uncomfortable to see little boys play with dolls. A B C D E 30. I am a loner. A B C D E 31. I am often impressed by what young people can achieve nowadays. A B C D E 32. You ean break a person physically but you can never take away their A B C D E human dignity. 33. Suffering can be meaningful for the growth o f the person. A B C D E 34. I am a very organized person. A B C D E 37. My career has changed several times. A B C D E 38. There has been times when I felt extremely close to someone I loved. A B C D E 39. If it were possible, I would greatly enjoy teaching adolescents. A B C D E 40. Life has been good to me. A B C D E 41. Host confliets between people can be resolved by discussion. A B C D E 42. In general, I believe that most people can achieve what they wish to achieve. A B C D E 45. I always have been a confident person. A B C D E 46. There have been people in my life with whom I have been willing to share A B C D E my innermost thoughts. 47. To be a good parent is one of the most challenging taskspeople face. A B C D E 48. I have left my mark on the world. A B C D E 49. I have difficulties dealing with my anger. A B C D E 50. I find it easy to work for future rewards. A B C D E 53. In general, I know what I want out of life. A B C D E 54. There have been several times in my life when I felt left out. A B C D E 55. I often feel that I am not growing as a person. A B C D E 56. There are many things I enjoy in life. A B C D E 57. I find it difficult to express my true feelings. A B C D E 58. I am quite self-sufficient. A B C D E 61. Although outwardly I am at ease, inwardly I amoften unsure. A B C D E Identity Processing Style and Sex-Role Identity 1C 62. I have never met anyone whom I really admired. A B C D E 63. I am very concerned that our children will grow in a polluted world. A B C D E 64. I find little sense in living. A B C D E 65. I am irritable/frustrated most of the time. A B C D E 66. It is important for young people to be independent. A B C D E 69. I feel very comfortable with the values I have. A B C D E 70. I feel inspired when I read about someone who, overcame major A B C D E obstacles and achieved a significant goal. 71. Planning for future generations is very important. A B C D E 72. 1 am o f no use to anyone. A B C D E 73. It is easy for me to believe most people. A B C D E 74. I find it easy to ask favors of others. A B C D E 77. I firmly believe that people should be responsible for their behaviors. A B C D E 78. When I have an orgasm I loose the sense of who and where I am. A B C D E 79. I derive great pleasure in seeing the accomplishments of young people. A B C D E 80. If 1 had the courage 1 would end my life. A B C D E 81. 1 find that most people are helpful. A B C D E 82. We would all be better off if people obeyed the laws we have. A B c D E 85. There are some issues on' which I take a strong stand. A B c D E 86. Overall, my sexual life has been satisfactory. A B c D E 87. With all of our technology, there is no need for anyone to work very hard. A B c D E 89. 1 know what it means to have a strong sense of self. A B c D E 89. I find 1 am open to new ideas. A B c D E 92. When necessary, 1 can devote a lot of energy to a task. A B c D E 93. My adolescence was fairly stormy. A B c D E 94. It would be difficult for me to have sexual intercourse with a person 1 A B c D E c D E did not love. 95. Hard work teaches self-respect. A B Identity Processing Style and Sex-Role Identity 102 96. When I die I will he missed. A B C D E 97. In general I'm an optimistic person. A B C D E 98. I have difficulties expressing my own opinion. A B C D E 101. There are times when I wish I had heen horn of the opposite sex. A B C D E 102. When I was a teenager I had a very close friend with whom I shared A B C D E many experiences. 103. Everyone who can, should have a paid job. A B C D E 104. I have given serious thought to the meaning of life. A B C D E 105. People have the capacity to solve their problems. A B C D E 106. I would prefer a job that pays on commission (i.e. depending upon A B C D E what I do) than one that pays a fixed salary. 109. Friends would describe me as a very, changeable person. A B C D E no. 1 enjoy being with people. A B C D E 111. 1 must admit that 1 am a fairly lazy person. A B C D E 112. When one is old it makes no sense to start new hobbies or activities A B C D E 113. Basically, 1 think I'm an all right person. A B C D E 114. "A place for everything and everything in its place" is my motto. A B C D E 117. I am very uncomfortable with "feminine" men. A B C D E 118. 1 can he friendly to strangers. A B C D E 119. 1 enjoy learning new skills. A B C D E 120. I keep physically active, within my body limits. A B C D E Note. 30 items have heen removed from this scale due to the fact that they are unnecessary to the present research project. The remaining items retain their regular numbering in order to facilitate scoring.