Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Same-sex social support and the enhancement of well-being Wetstein-Kroft, Susan Beth 1986

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1987_A2 W47.pdf [ 7.49MB ]
Metadata
JSON: 831-1.0056024.json
JSON-LD: 831-1.0056024-ld.json
RDF/XML (Pretty): 831-1.0056024-rdf.xml
RDF/JSON: 831-1.0056024-rdf.json
Turtle: 831-1.0056024-turtle.txt
N-Triples: 831-1.0056024-rdf-ntriples.txt
Original Record: 831-1.0056024-source.json
Full Text
831-1.0056024-fulltext.txt
Citation
831-1.0056024.ris

Full Text

SAME-SEX SOCIAL SUPPORT AND THE ENHANCEMENT OF WELL-BEING by Susan Beth Wetstein Kroft B.A., York University, 1978 M.A., The University of British Columbia, 1982 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF EDUCATION in THE FACULTY OF GRADUATE STUDIES Faculty of Education (Counselling Psychology) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA December 1986 © Susan Beth Wetstein Kroft, 1986 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of ^ u ^ . ^ U ^ - A ^ c A ^ ^ The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 DE-6(3/81) Same-Sex Bonds and Well-Being p. ii Abstract The investigator tested propositions derived from theories of male (Lipman-Blumen, 1976) and female (Bernard, 1976) same-sex bonding against propositions derived from theories of male (Bell, 1981; Safilios-Rothschild, 1981) and female (Bell, 1981; Lipman-Blumen, 1976) cross-sex bonding and, against propositions derived from a general theory of social relationships (Weiss, 1974). The purpose of the study was to determine the relevance of the variables, "sex of respondent" and "relationship gender composition" to the attainment of social support and to the relationship between social support and global well-being. The study was divided into three parts. In part one the investigator tested opposing propositions related to differences between the sexes in the overall level of social support provided by same-sex and cross-sex bonds. Lipman-Blumen (1976) has proposed that, in the overall attainment of social support, men's same-sex bonds are stronger than women's same-sex bonds; that men's same-sex bonds are stronger than men's cross-sex bonds; and that women's cross-sex bonds are stronger than women's same-sex bonds. Conversely, Safilios-Rothschild (1981) has suggested that men's same-sex bonds are weaker than women's same-sex bonds; that men's same-sex bonds are weaker than men's cross-sex bonds; and that women's cross-sex bonds are weaker than women's same-sex bonds. In part two, the investigator tested opposing propositions related to differences in the individual dimensions or provisions of social support provided Same-Sex Bonds and Weil-Being p. iii by men's and women's same-sex and cross-sex bonds. Theorists emphasizing same-sex bonds have suggested that women's (Bernard, 1976) or men's (Lipman-Blumen, 1976) same-sex bonds provide higher levels of certain dimensions of social support than do women's or men's cross-sex bonds. Conversely, cross-sex bonding theorists have suggested that women's (Bell, 1981, Lipman-Blumen, 1976) or men's (Bell, 1981, Safilios-Rothschild, 1981) cross-sex bonds provide higher levels of certain dimensions of social support than do women's or men's same-sex bonds. In contrast to both the same-sex and cross-sex bonding theorists, Weiss (197*) has implied that women's and men's same-sex and cross-sex bonds provide equivalent levels of specific dimensions of social support. In part three, the investigator tested opposing propositions related to differences between the sexes in the relationship between the overall attainment of social support to their sense of global well-being. Bernard (1976) and Miller (1976) have proposed that this relationship is stronger for women than it is for men. Conversely, Weiss (197*) has implied that the relationship between social support and well-being is the same for men and women. The investigator also tested opposing propositions related to within sex differences in the relationship between the overall attainment of social support and global wellrbeing. Bernard (1976) and Miller (1976) have proposed that women's same-sex bonds are more strongly related to their sense of well-being than are women's cross-sex bonds. Conversely, Lipman-Blumen (1976) has suggested that women's cross-sex bonds are more strongly related to their sense of well-being than are women's same-sex bonds. Lipman-Blumen (1976) has also proposed that men's same-sex bonds are more strongly related to their sense of well-being than are men's cross-sex Same-Sex Bonds and Well-Being p. iv bonds. Conversely, Safilios-Rothschild (1981) has suggested that men's cross-sex bonds are more strongly related to their sense of well-being than are men's same-sex bonds. In contrast to the same-sex and cross-sex bonding theorists, Weiss (197*) has implied that men's and women's same-sex and cross-sex bonds are equally associated with their sense of well-being. The investigator tested these propositions by having 101 married women and 101 married men, who attended the University of British Columbia summer school session, fill out a series of questionnaires on social support and well-being. All subjects were between the ages of 25 and *5. None of the subjects were married to one another. Respondents completed the Social Provisions Scale (Russell & Cutrona, 198*) designed to assess Weiss's (197*) six dimensions or provisions of social support. These provisions are: attachment, social integration, reassurance of worth, reliable alliance, help and guidance, and, the opportunity for nurturance. The respondents completed the Social Provisions Scale twice, once for their same-sex relationships and once for their cross-sex (non-marital) relationships. Respondents also completed six measures of well-being: the Rosenberg Self-Esteem Scale (Rosenberg, 1969); the Spheres of Control Scale—personal efficacy and interpersonal control dimensions (Paulhus 6c Christie, 1981); the Satisfaction With Life Scale (Diener, 198*); the UCLA Loneliness Scale—Revised-Abbreviated Version (Russell, 1980); and, the Hakstian-McClean Depression Scale (Hakstian & McClean, 1979). The six measures of well-being were statistically combined, forming a molar variable identified as "global well-being". Same-Sex Bonds and Well-Being p. v Concerning the attainment of social support, the results of this study indicate that, overall, women's same-sex bonds provide higher levels of social support than do men's same-sex bonds. Women's same-sex bonds also provide higher levels of social support than do women's cross-sex bonds. Men's same-sex and cross-sex bonds appear to provide equivalent levels of social support, overall. These results indicate partial support for propositions arising from Safilios-Rothschild's (1981) theory and fail to support propositions arising from Lipman-Blumen's (1976) theory. In addition, the women's same-sex bonds provide higher levels of several provisions of social support than do women's cross-sex bonds, which supports Bernard's (1976) theory. Men's same-sex bonds provide higher levels of two dimensions of social support than do men's cross-sex bonds, which indicates only partial support for propositions arising from Lipman-Blumen's (1976) and Weiss's (1974) theories. Concerning the relationship between social support and global well-being, the results of this study indicate that social support appears to be equally related to men's and women's sense of global well-being, supporting propositions implied by Weiss's (1974) theory. Moreover, both same-sex and cross-sex bonds appear to be strongly, but equally associated with men's and women's sense of well-being which again supports propositions implied by Weiss's (1974) theory. In conclusion, the variables of "sex of respondent" and "relationship gender composition" appear to be more relevant in the attainment of social Same-Sex Bonds and Weil-Being p. vi support than in the relationship between social support and well-being. The impact of these results on Weiss's (197*) theory of social relationships, on Bernard's (1976) and Lipman-Blumen's (1976) theories of same-sex bonding, on theories of social support, and on theories of adult development are discussed. The practical implications of these results for counsellors and for future research investigations are outlined. Same-Sex Bonds and Well-Being p. vii TABLE OF CONTENTS ABSTRACT ii LIST OF TABLES x LIST OF FIGURES xiv ACKNOWLEDGEMENT xv CHAPTER I: INTRODUCTION A. Perspectives on Social Support 3 1. Weiss's Theory of Social Relationships 5 2. Same-Sex Friendship 10 3. Theories of Homosociality 11 4. Cross-Sex Friendship • 18 B. Perspectives on Mental Health and Well-Being 21 1. Society 21 2. Mental Health Theorists and Professionals 22 3. Individual 25 C. Statement of the Problem 27 D. Objectives of the Study 28 E. Research Hypotheses • • • 29 F. Significance of the Study 35 G. Limitations of the Study 37 Same-Sex Bonds and Weil-Being p. viii CHAPTER II: REVIEW OF THE LITERATURE A. Weiss's Theory of Social Relationships 39 B. Theories of Homosociality ** 1. Bonds between Men • *6 2. Bonds between Women 51 CHAPTER III: METHODOLOGY A. Sample 60 B. Research Procedures 60 C. Instrumentation 61 1. Social Provisions Scale 61 2. The Spheres of Control Scale 6* 3. The UCLA Loneliness Scale - Revised 67 i*. The Satisfaction with Life Scale 68 5. The Rosenberg Self-Esteem Scale 70 6. The Hakstian-McClean Depression Scale 72 D. Design and Analysis for the Four Research Questions • 73 E. Design and Analysis of the Network Data 75 CHAPTER IV: RESULTS A. Characteristics of the Sample 78 B. Hypotheses relating to the Effects of Sex of Respondent and Relationship Gender Composition on the Overall Level of Social Support 79 Same-Sex Bonds and Weil-Being p. ix 1. ANOVA • 79 2. Hypotheses 80 3. Summary of Hypotheses One through Five • 83 C. Hypotheses relating to the Effects of Sex of Respondent and Relationship Gender Composition on the level of Individual Relational Provisions 84 1. MANOVA 84 2. Hypotheses 85 3. Summary of Hypotheses Six through Eleven 89 D. Hypotheses relating to the Effects of Sex of Respondent and Relationship Gender Composition on the Association between the Overall Level of Social Support and Global Well-being • 91 1. Correlational Analyses 91 2. Hypotheses 92 3. Summary of Hypotheses Twelve through Eighteen 100 E. Sex Differences in Network Characteristics 101 F. Network Characteristics Predicting Global Well-being 102 G. Sex Differences in the Number of Network Members Meeting Relational Provisions 105 H. Supplies of Relational Provisions predicting Global Well-being 109 J. Summary of the Network Data Results 109 Same-Sex Bonds and Weil-Being p. x C H A P T E R V: DISCUSSION, IMPLICATIONS A N D C O N C L U S I O N S A . Discussion of the Findings relating to the Four Research Questions 112 1. The Effects of "Sex of Respondent" and "Relationship Gender Composition" on Overall Level of Social Support: Are men's same-sex relationships stronger than women's same-sex relationships? 113 2. The Effects of "Sex of Respondent" and "Relationship Gender Composition" on the Level of Individual Relational Provisions: Do women's and men's same-sex bonds provide higher levels of the individual relational provisions than do women's and men's cross-sex bonds? 115 3. The Effect of "Sex of Respondent" on the association between Level of Social Support and Global Weil-Being: Are satisfying social bonds more strongly associated with the global well-being of women than with the global well-being of men? 120 *. The Effect of "Gender Composition" on the association between Level of Social Support and Global Weil-Being: Are satisfying same-sex bonds more strongly associated with global Same-Sex Bonds and Well-Being p. xi well-being than satisfying cross-sex bonds? 123 5. Discussion of Questions emerging from the Data 125 B. Limitations of the Study 130 1. Causality 130 2. Generalizability 131 3. Social Provisions Scale 132 C. Theoretical Implications of the Study 133 1. Implications for Weiss's Theory of Social Relationships 133 2. Implications for Bernard's Theory of Homosociality 136 3. Implications for Lipman-Blumen's Theory of Homosociality 138 4. Implications for Theories of Social Support 140 5. Implications for Theories of Adult Development 141 D. Practical Implications of the Study 142 1. Implications for Counselling 142 2. Implications for Future Research 143 E. Conclusions 146 REFERENCES 149 APPENDICES A. Supplementary Statistical Analyses 164 B. Instrumentation 170 Same-Sex Bonds and Weil-Being p. xii LIST OF TABLES 1. Tabular presentation of the variables, design, analysis and instrumentation 77 2. Mean provision scores for women's same-sex and cross-sex bonds 87 3. Mean provision scores for men's same-sex and cross-sex bonds 90 *. Separate intercorrelation matrices of variables forming the first principal component: Men and women 93 5. Pooled intercorrelation matrix of variables forming the first principal component 9* 6. Sex differences in same-sex network composition and frequency of contact 103 7. Global well-being regressed onto network characteristics: Men 10* 8. Number of women's same-sex and cross-sex relation-ships meeting relational provisions 106 9. Number of men's same-sex and cross-sex relationships meeting relational provisions 107 10. Global well-being regressed onto the number of network members supplying each relational provision: Men 110 11. Global well-being regressed onto the number of network members supplying each relational provision: Women 110 Same-Sex Bonds and Well-Being p. xiii 12. Differences in the correlations between women's same-sex and cross-sex bonds with indices of well-being 164 13. Differences in the correlations between men's same-sex and cross-sex bonds with indices of well-being 165 14. Global well-being regressed onto the individual relational provisions: Men 166 15. Global well-being regressed onto the individual relational provisions: Women 166 16. Social provisions scale: Intercorrelation matrix for men's and women's same-sex bonds •••• 167 17. Social provisions scale: Intercorrelation matrix for women's and men's cross-sex bonds 168 18. Full regression analyses: Comparison of dependent multiple R's 169 19. Full regression analyses: Comparison of independent multiple R's 169 Same-Sex Bonds and Weil-Being p. xiv LIST OF FIGURES 1. Graphic presentation of the ANOVA: Main effects and interactions • 81 Same-Sex Bonds and Weil-Being p. xv ACKNOWLEDGEMENT I wish to dedicate this achievement to three people. To my partner in life, Ed, for his unyielding love and loyal support. To my mother, for her great courage and inspiring dignity. To my father, for his gentle wisdom and patient understanding. I take pride in sharing this achievement with them. I would like to thank my research supervisor and mentor, Dr. Lorette Woolsey for bringing out my strengths as an academic and my pride in being a woman. I would also like to thank Dr. Dan Perlman for his insightful comments, encouragement and guidance; Dr. Ralph Hakstian for his warmth, patience and humour, and Dr. Sharon Kahn for her careful attention to the details of this manuscript. I also wish to express my gratitude to my sources of support, my sister, Judy and my brother, Brian and to my friends Janet, Glennie, Kathy, Paul, Anne, Gisela, Linda, Joanne, Marsha and Les for believing in me and for helping me to believe in myself. Same-Sex Bonds and Well-Being p. 1 CHAPTER I: INTRODUCTION In the last decade, converging evidence, from independent sources of theory and research, has drawn attention to the general association between social support and well-being (Diener, 1984; Gottlieb, 1983; Leavy, 1983; Weiss, 1974). Researchers have now been challenged to investigate and to identify the specific factors that contribute to the attainment of social support and to the relationship between social support and well-being (Cohen & Syme, 1985; Cohen & Wills, 1985; Diener, 1984; Reis, 1984). One specific variable, emerging from the literature on homosociality and friendship, that may affect the attainment of social support and clarify the relationship between social support and well-being is the "gender composition" (same-sex, cross-sex) of men's and women's dyadic relationships. To investigate the relevance of "relationship gender composition" as a specific variable affecting the attainment of social support and in the relationship between social support and well-being, the investigator tested propositions derived from three opposing theoretical perspectives. Theories of male (Lipman-Blumen, 1976) and female (Bernard, 1976) bonding were tested against theories of male (Bell, 1981; Safilios-Rothschild, 1981) and female (Bell, 1981; Lipman-Blumen, 1976) cross-sex bonding, and, against Weiss's (1974) theory of social relationships. Theorists of same-sex bonding (Bernard, 1976; Lipman-Blumen, 1976) have posited that same-sex bonds provide higher levels of specific dimensions or Same-Sex Bonds and Well-Being p. 2 provisions of social support and contribute more significantly to well-being than do cross-sex bonds. Conversely, theorists of cross-sex bonding (Bell, 1981; Lipman-Blumen, 1976; Safilios-Rothschild, 1981) have posited that cross-sex bonds provide higher levels of specific dimensions or provisions of social support and contribute more significantly to well-being than do same-sex bonds. In contrast, Weiss (197*) has implied that same-sex and cross-sex bonds provide equivalent levels of particular support dimensions and well-being. Thus, the investigator tested the relevance of "relationship gender composition" from three different perspectives. The investigator tested the theoretical propositions derived from these theories by comparing the different levels of social support and well-being men and women receive through their same-sex and cross-sex bonds. The results of this study contribute to the identification of the specific factors affecting the attainment of social support as well as the specific factors affecting the association between social support and well-being. Chapter One is divided into seven sections. In the first section, literature relating to perspectives on the predictor variable, social support, is presented. These perspectives include: Weiss's (197*) theory of social relationships, same-sex friendship, theories of homosociality, and cross-sex friendship. In the second section, literature relating to perspectives on the outcome variable, well-being, is presented. These perspectives include: the mental health professional, society, and the individual. In the third and fourth sections, the statement of the problem and objectives of the study are presented. In the fifth section, the Same-Sex Bonds and Weil-Being p. 3 research hypotheses a re o u t l i n e d and are f o l l o w e d by a discussion o f the s i g n i f i c a n c e of the study. The c h a p t e r concludes w i t h a discussion of the l i m i t a t i o n s of the study. A. P e r s p e c t i v e s on S o c i a l Support O ver the la s t decade, the concept of s o c i a l support has gained prominence as a focus f o r i n t e g r a t i v e research l i n k i n g the q u a l i t y of p h y s i c a l and p s y c h o l o g i c a l h e a l t h to the adequacy of the s o c i a l e nvironment ( G o t t l i e b , 1983; L e a v y , 1983). The e v o l v i n g i n t e r e s t in s o c i a l support has p a r a l l e l e d and r e f l e c t e d the p a r a d i g m a t i c s h i f t in the m e n t a l h e a l t h professions f r o m i s o l a t i n g the causes of p s y c h o l o g i c a l d i s t r e s s w i t h i n the i n d i v i d u a l , t o the l i n k i n g of i n d i v i d u a l d i f f i c u l t i e s w i t h d e f i c i t s in the c u r r e n t s o c i a l environment (Bloom, 1979; G o t t l i e b , 1981; Henderson, 1982; Weiss, 1973). R e s e a r c h e r s have focused t h e i r i n v e s t i g a t i o n s on l i n k i n g the s t r u c t u r a l p r o p e r t i e s of s o c i a l support ( H i r s c h , 1981; P h i l l i p s , 1981; Wilcox, 1981) or i t s c o n t e n t domain (Cobb, 1976; House, 1981; Kahn Sc A n t o n n u c c i , 1980) to the b u f f e r i n g of l i f e stresses (Leavy, 1983) in c l i n i c a l populations ( M i l l e r Sc Ingham, 1976; T o l s d o r f , 1976), i n samples w i t h s p e c i f i c disorders, notably depression (Brown, B h r o l c h a i n Sc H a r r i s , 1975; F l e t c h e r Sc Rossaby, 1980; P a y k e l <5c Emms, 1980) and in populations e x p e r i e n c i n g s t r e s s f u l l i f e t r a n s i t i o n s ( B a r r e t t , 1978; C a r v e t h & G o t t l i e b , 1979; H i r s c h , 1980; Weiss, 1973; Wilcox, 1981). This model of support, t e r m e d the " b u f f e r i n g model" has proposed t h a t support " i s r e l a t e d to well-being only (or p r i m a r i l y ) f o r persons under stress ... It posits that support Same-Sex Bonds and Weil-Being p. * buffers (protects) persons from the potentially pathogenic influence of stressful events" (Cohen <5c Wills, 1985, p. 310). Despite inconsistencies in definition, research method and theory (Gottlieb, 1981; Leavy, 1983; Wilcox, 1981), investigators have consistently affirmed the general proposition that social support buffers the effects of a diverse range of stressful physical and emotional events and life transitions (Barerra, 1981; Brown, Bhrolchain & Harris, 1978; Cobb, 1978; DiMatteo & Hays, 1981; Gottlieb, 1981; 1983; Henderson, 1982; Weiss, 1973, Wilcox, 1981). Much less attention has been directed towards evaluating the role of social support to well-being in the everyday lives of a non-clinical, non-crisis population (Reis, 198*; Woolsey, 198*). The failure of researchers to consider the benefits of social support to well-being in everyday life may be related to conceptions of healthy adult adjustment. Prevailing conceptions of "mental health" have equated healthy adult adjustment to typically masculine characteristics of autonomy, separateness, independence and self-sufficiency (Broverman, Broverman, Clarkson, Rosenkrantz & Vogel, 1970). Consequently, the hallmarks of healthy adult adjustment have been antithetical to the essence of social support which has emphasized the individual's requirement for interconnectedness. The crisis-oriented conceptualization of social support has implied that individuals only need others in times of physical and emotional distress. Same-Sex Bonds and Well-Being p. 5 In their analysis of the existing literature on social support, Sandler and Barerra (1984) have noted that two independent models of social support have emerged. In the first model, which has dominated the literature, social support has been conceptualized as a buffer or moderator of stress, facilitating the individual's recovery from illness to health. The second model is far less represented in the literature. In it, the dimensions of social support have been conceptualized as basic human needs. This latter model has proposed "that social resources have a beneficial effect irrespective of whether persons are under stress. Because the evidence for this model derives from the demonstration of a statistical main effect of support with no stress x support interaction, it is termed the main effect model" (Cohen & Wills, 1985, p. 310). This model of support has been evident in Weiss's work (1969; 1973; 1974; 1982) and was the perspective taken in this investigation. In the section to follow, the model of social support adopted for use in the present investigation is outlined. 1. Weiss's Theory of Social Relationships Drawing on his work with groups deficient in various aspects of their social networks, Weiss (1969; 1973; 1974; 1982) has developed a typology of six provisions supplied through social relationships. These six provisions are: attachment, social integration, reassurance of worth, reliable alliance, help and guidance, and, the opportunity for nurturance. In the section to follow, each provision is discussed. Same-Sex Bonds and Well-Being p. 6 The first provision is attachment. Weiss has proposed that relationships supplying attachment provide the individual with a sense of security and place. According to Weiss, for most people, attachment is provided through heterosexual bonds, or, in some cases, through relationships with a close friend or kin. Weiss has placed primary importance on the attainment of the attachment provision. As Weiss (197*) has said, "we might think of individuals organizing their lives around whatever relationship provides them with attachment" (p.25). Weiss has observed that deficits in the attachment provision result in emotional loneliness and depression. The second provision is social integration. Weiss has proposed that relationships supplying the social integration provision provide the individual with a sense of belonging to a wider community where there is an opportunity for sharing a common set of concerns, interests and information. He has observed that social integration is provided by both dyadic and network linkages and has hypothesized that deficits in the social integration provision result in social loneliness, marginality, anxiety, and boredom. The third provision is reassurance of worth. Weiss has posited that the reassurance of worth provision provides the individual with a sense of competence in the performance of a social role. He has hypothesized that this provision is supplied through work relationships from colleagues. For homemakers, reassurance of worth can be obtained from spouses, children and friends. Weiss has suggested that deficits in the reassurance of worth provision result in a lowered sense of self-esteem. Same-Sex Bonds and Weil-Being p. 7 The fourth provision, reliable alliance, supplies the individual with assistance in an emergency. He has hypothesized that this provision is primarily supplied through kin relationships and that deficits in the reliable alliance provision result in a sense of vulnerability and isolation. Weiss has reasoned that the fifth provision, the obtaining of help and guidance, provides the individual with an opportunity to seek advice, assistance and problem-solving strategies. This provision is met from relationships with a trusted authority figure. He has hypothesized that deficits in this provision result in a sense of diminished personal efficacy. The sixth provision is the opportunity for nurturance. The opportunity for nurturance provision supplies the individual with a sense of meaning, purpose and life satisfaction. Weiss has hypothesized that this provision is primarily met through the parent-child relationship. The opportunity for nurturance provision is the only relational provision in which support is conceptualized within the context of giving rather than receiving. Several aspects of Weiss's (1974) theoretical ideas are important to highlight. First, as Russell and Cutrona (1984) point out, Weiss did not conceptualize the provisions of social relationships as a theory of social support. Instead, Weiss was "attempting to offer a comprehensive view of what we receive through our relationships with others" (p. 3). However, the provisions of social relationships Weiss has proposed are consistent with the dimensions of social support identified by social support theorists (Caplan, 1974; 1981; Cobb, Same-Sex Bonds and Well-Being p. 8 1976; Kahn <5c Antonnucci, 1980; House, 1981). Therefore, Weiss's theory has been accepted as a theory of social support (Russell et al., 198*). Second, although Weiss has contended that satisfactory supplies of the six relational provisions contribute to well-being, he did not clearly define the concept or the criteria of well-being within a positive mental health framework. Instead, he has focused on linking the notion of "relational deficits" to such negative psychological consequences as loneliness, depression, lower self-esteem, diminished personal efficacy and diminished life satisfaction. Recent research interest in Weiss's theory has focused on validating his concept of "relational deficit" and its relationship to loneliness (Cutrona, 1982; Cutrona, Rose & Yurko, 198*; Shaver & Rubenstein, 1982), and to depression (Bernard, 1976). No known studies have attempted to evaluate whether the satisfaction of Weiss's relational provisions contribute to well-being within a positive mental health framework. One of the purposes of the present investigation was to examine the association between Weiss's dimensions of social relationships to well-being, within a positive mental health context. Finally, although a general association between social support and well-being has been well documented, researchers have now been asked to turn their attention to identifying the "types of social relationships" that contribute to wellrbeing (Diener, 198*). Weiss's typology of relational provisions is rich in clinical insight and has provided testable hypotheses which may clarify the relationship between social support and well-being. However, although Weiss's Same-Sex Bonds and Weil-Being p. 9 theory has addressed the "type of relationship" issue identified by Diener as an important variable for the satisfaction of relational needs and well-being, he has approached this question in terms of the specific role relationships the individual occupies as a spouse, friend, co-worker, family member, employee, and parent. Weiss did not consider the question of differences in the effectiveness of men and women as social partners for supplying specific aspects of social support. In addition, he did not address the question of sex differences in the relationship between social support and well-being nor did he address the question of differences in the effectiveness of men and women as social partners for contributing to well-being. In overlooking these questions, Weiss has made three assumptions: that there are no differences in the effectiveness of men's and women's same-sex and cross-sex bonds for supplying specific dimensions of social support; that men and women benefit to the same degree by social bonds; and, that men's and women's same-sex and cross-sex bonds equally contribute to their well-being. Specific attention to the separate and interacting relevance of "sex of respondent" (male or female) and "relationship gender composition" (same-sex, cross-sex) for the attainment of specific dimensions of social support and to the relationship between social support and well-being have been descriptively highlighted in the literature on friendship and theoretically highlighted in the literature on homosociality. In the sections to follow the literature pertaining to same-sex friendship, theories of same-sex bonding (homosociality) and cross-sex friendship is presented. Same-Sex Bonds and Weil-Being p. 10 2. Same-Sex Friendship The literature on friendship has illuminated the normative persistence and homogeneity of same-sex friendship bonds throughout the life cycle (Dickens & Perlman, 1981). While friends tend to be homogeneous with respect to age, gender, educational attainment, and socioeconomic status (Cohen, 1961) this literature has underscored sex differences in the social styles and effectiveness of men and women as social partners for supplying specific relational needs (Wheeler, Reis & Nezlek, 1983). Gender differences in the quality and style of interaction characterizing same-sex relationships have been widely documented. Descriptive studies of same-sex friendships have consistently affirmed that women's relationships are "affectively richer" (Booth, 1972), reciprocal (Reisman, 1981, Weiss & Lowenthal, 1975), self-disclosing and holistic (Wright, 1982), therapeutic (Davidson & Packard, 1981), self-revealing and accepting (Bell, 1981), and intellectually stimulating (Rose <5c Serafica, 1979). Relationships between men have been described as instrumental and activity-oriented (Bell, 1981), as emphasizing commonality (Weiss & Lowenthal, 1972), as meeting circumscribed needs (Wright, 1982), as stable and enduring (Tiger, 1969), and, as extensive (Wright, 1982). The studies cited so far have taken a descriptive, but largely atheoretical approach to the examination of same-sex bonds. The present study extends Same-Sex Bonds and Weil-Being p. 11 previous work by examining same-sex bonds within a theoretical context. In the section to follow, literature relating to theories of homosociality are presented. 3. Theories of Homosociality Bernard (1976) has defined homosociality as "the different ways that men and women relate to members of their own sex" (p. 227-8). Bernard's formulation of homosociality has focused on outlining "differences" in the relational styles of men and women to account for the "relational deficits" women experience in their cross-sex bonds as well as the "relational deficits" women experience when their relationships with other women are limited or weak. Bernard has described the relational style of women as one which is oriented towards affiliation, attachment, and emotional expressiveness. Conversely, she has described the relational style of men as one which is oriented towards independence, autonomy and emotional inexpressiveness. Consequently, she has argued that women's relational needs are not sufficiently met or validated when relating to members of the opposite sex. She has applied her theory of homosociality as an explanatory concept for understanding the greater incidence of depression among married (as opposed to never married) women over this last century. Bernard has identified her theory as an attempt to understand the "normal" kind of depression, "the kind that is measured by instruments rather Same-Sex Bonds and Weil-Being p. 12 than diagnosed in a clinic, the everyday garden variety of depression, the depression of the woman who is 'just dragging herself around' (p. 21*)." In developing her theory, Bernard has drawn on several sources, including: the literature on depression, Miller's (1976) theory of female psychology, the literature on sex differences, Smith-Rosenberg's (1975) work on women's relationships in the nineteenth century, and Weiss's (197*) concept of relational deficit. Although Bernard has attributed the strains imposed by conformity to the female role (with its encouragement of dependency and discouragement of personal efficacy), as significant factors contributing to depression, she has also linked Miller's (1976) theory of women's affiUative style to Weiss's (197*) concept of relational deficit to further explain the higher incidence of depression among married women over the last century. Bernard has hypothesized that "the female way of being social—involving bonds, affiliation, attachment—renders women, because of their learned as well as actual helplessness, more vulnerable to the stresses of deprivation of such ties, and hence, to depression" (p. 228). Bernard has cited Smith-Rosenberg's (1975) historical account of women's same-sex relationships in the nineteenth century to describe the potential richness available within women's homosocial bonds. Smith-Rosenberg has drawn attention to the importance of women's same-sex bonds for providing both instrumental aid and emotional support. More importantly, women's same-sex relationships in the nineteenth century affirmed their strengths and facilitated a sense of pride in their affiliative style and female identity. Bernard has Same-Sex Bonds and Weil-Being p. 13 commented that in the nineteenth century women's same-sex relationships compensated for the relational deficits they experienced in their cross-sex ties. Bernard has hypothesized that as a consequence of the devaluation of women's same-sex bonds over the last century, women became isolated from each other and ashamed of their affiliative needs. Thus, women were left more vulnerable to the relational deficits experienced in their cross-sex ties, and hence, more vulnerable to depression. Bernard has hypothesized that women's same-sex bonds provide access to an emotionally supportive relational context where a woman's stronger inclination towards affiliation is valued. According to Bernard, women's same-sex relationships have the potential to enhance their sense of self-esteem, personal efficacy, and life satisfaction, and hence, alleviate the cause of their depression. The stronger pull of women towards the establishment of affiliative bonds has also been addressed by Miller (1976). Miller has pointed out that because women are socialized to define their identity in relation to their success in establishing and maintaining social bonds, the erosion of their social bonds results in the erosion of their identities. As Miller has explained, "a woman's sense of self becomes very much organized around relationships. Eventually, threat of disruption of an affiliation is perceived not just as a loss of a relationship but as something closer to a total loss of self" (p. 83). In opposition to Weiss (1974), both Bernard (1976) and Miller have hypothesized that satisfactory social bonds (with both sexes) are more strongly associated with a woman's sense of well-being than a man's sense of well-being. Same-Sex Bonds and Weil-Being p. 14 Lipman-Blumen (1976) has defined homosociality as "the seeking, enjoyment and/or preference for the company of the same-sex" (p. 16). Lipman-Blumen has applied her formulation of homosociality to men in order to account for the exclusion of women from positions of occupational, political and economic power in historical and contemporary society. Lipman-Blumen has viewed the basis of male bonding as an attraction to power. Her concept of power is consistent with the stereotypic male model which is conceptualized as control over resources. She has said that "the pragmatic recognition that males controlled economic, political, educational, occupational and social resources created a situation in which men identified with and sought help from other men" (p. 16). Because men have been the primary controllers of resources in society, she has argued that they are more attractive to both sexes. "Women, recognizing the existential validity of the situation, also turned to men for help and protection" (ibid). Thus, Lipman-Blumen has argued that men's same-sex bonds are stronger than women's same-sex bonds. Moreover, she has hypothesized that women's cross-sex relationships are stronger than their same-sex relationships. Lipman-Blumen's and Bernard's concepts of homosociality have provided both a theoretical complement and a source of opposition to Weiss's theory of social relationships. Bernard has drawn directly on Weiss's concept of relational needs or provisions and relational deficit. She has postulated that same-sex bonds are important for the fulfillment of relational needs which are more effectively met in relationships of similar social styles. From Bernard's theory one would predict that women derive higher levels of the attachment, social integration, reassurance of worth, reliable alliance, help and guidance and the Same-Sex Bonds and Weil-Being p. 15 opportunity for nurturance provisions from their same-sex bonds than from their cross-sex bonds. From Bernard's theory one also would predict that overall satisfaction (the sum of the six provision scores) with same-sex bonds is more strongly and positively associated with well-being than women's overall satisfaction (the sum of the six provision scores) with their cross-sex bonds. While Lipman-Blumen did not address Weiss's typology of relational needs directly, her theory of the relational and psychological benefits derived through men's same-sex bonds complemented that of Weiss's. Within the framework of relational provisions proposed by Weiss, Lipman-Blumen's theory leads one to predict that men derive higher levels of social integration, reassurance of worth, reliable alliance and help and guidance through their association with the male world. Lipman-Blumen's theory also leads one to predict that for men, overall satisfaction (the sum of the six relational provisions) with same-sex bonds is more strongly and positively associated with well-being than overall satisfaction (the sum of the six relational provisions) with their cross-sex bonds. Bernard's and Lipman-Blumen's applications of homosociality have been gender-specific, with Bernard concentrating on the homosocial bonds of women and Lipman-Blumen on the homosocial bonds of men. Nevertheless, there have been significant points of convergence in their positions. Specifically, each theorist has posited that higher levels of specific relational provisions and well-being are more adequately met through same-sex bonds than cross-sex bonds. Points of tension in their arguments have been linked to the basis of the bond. For Bernard, the basis of female bonding has been motivated by the fulfillment I of affiliative needs. For Lipman-Blumen, the basis of male bonding as well as i Same-Sex Bonds and Well-Being p. 16 the basis of female cross-sex bonding has been motivated by the fulfillment of power needs. While both theorists have addressed the issue of power, they conceptualize it differently. Lipman-Blumen has suggested that power over resources is the basis of the bond between men and that association within the male world enhances their personal efficacy, self-esteem and life satisfaction. Bernard has suggested that the basis of female bonding is affiliation and that the outcome of the bond is enhanced personal efficacy, self-esteem, and life satisfaction. The central point of controversy between Weiss (197*) and the homosociality theorists has been whether or not the gender composition of social bonds affects the attainment of specific relational provisions and the association between social support and well-being. Bernard (1976) has hypothesized that higher levels of social support and well-being are more adequately met within relationships of similar social styles, as in same-sex bonds. Lipman-Blumen (1976) has hypothesized that higher levels of specific relational provisions are obtained in relationships with powerful others (same-sex relationships for men and cross-sex relationships for women). In his theory, Weiss has not addressed sex differences in affiliative needs nor has he addressed sex differences in the effectiveness of men and women as social partners for supplying particular dimensions of social support and well-being. The impact of relationship gender composition on the attainment of relational provisions and on the relationship between social support and well-being was the central issue addressed in this investigation. Same-Sex Bonds and Weil-Being p. 17 Although theories of male bonding (Lipman-Blumen, 1976; Tiger, 1969) have asserted that men's same-sex bonds are stronger and more comprehensive than women's same-sex bonds, studies comparing the same-sex relationships of men and women (Booth & Hess, 1972; Gibb, Auerbach & Fox, 1982; Rose, 1985; Wright, 1982) have indicated that women's same-sex bonds are stronger and more comprehensive than men's same-sex bonds. This point of controversy is further addressed in the current investigation. However, the present study extends previous work in two major respects. First, most investigators have restricted their study of same-sex bonds to "friendship", and have overlooked other relevant same-sex bonds with family members, supervisors, co-workers and community affiliations. The present investigation extends past studies by expanding previous conceptualizations of same-sex bonds to encompass a more comprehensive range of same-sex relationships. Second, the majority of past studies have not compared the overall strength and individual character of men's and women's same-sex bonds with their cross-sex bonds. Consequently, there is still some question as to whether men's and women's same-sex and cross-sex relationships are similar in character or whether their same-sex and cross-sex relationships are different in character. By comparing the same-sex and cross-sex relationships of men and women, this study highlights similarities and differences in the overall strength and individual character of men's and women's social bonds. In the next section, the theory and research relating to cross-sex bonds is examined with an attendant focus on outlining opposing propositions derived from theories of same-sex bonding. I I I Same-Sex Bonds and Well-Being p. 18 *. Cross-sex Friendship Sharply contrasting the importance, continuity and homogeneity of same-sex friendships throughout the life cycle, Booth and Hess (197*) have described cross-sex friendships beyond the courtship period as an "anomaly". Because of their infrequency following marriage, the study of cross-sex relationships beyond the courtship period has received very little research attention. Much of the literature in this area has involved theoretical speculation on the qualitative nature and supportive functions served by cross-sex friendship bonds (Bell, 1981; Bernard, 1976; Lipman-Blumen, 1976; Pleck, 1975; Safilios-Rothschild, 1981) and the barriers to their formation (Bell, 1981). The available empirical evidence has focused on the demographic and sociological variables which facilitate and inhibit the development of cross-sex linkages. Booth and Hess (197*) have reported that cross-sex relationships are facilitated by greater educational attainment and white collar occupational status but discouraged by marital status. As Bell (1975) has commented, "the belief in potential sexual involvement between any man and woman not married to one another has been the greatest deterrent to the development of cross-sex friendship" (p. 153). The literature on same-sex friendship bonds has drawn attention to the shared and reciprocal properties of the relationship. Conversely, empirical evidence and theoretical speculation on the nature and supportive functions served by cross-sex bonds have drawn attention to the individual and non-Same-Sex Bonds and Well-Being p. 19 reciprocal properties of the relationship. For example, Bell (1981) has pointed out that "when friendships between men and women do exist, males tend to be older and better educated, placing the male in a superordinate position relative to the woman" (p. 97). Moreover, Booth and Hess (1974) have found that although cross-sex friendships were facilitated through occupational and volunteer work contexts, clear status differences were apparent, with the woman occupying a subordinate position. Bell (1981) and Lipman-Blumen (1976) have suggested that women enjoy relationships with men as a means for gaining access to greater power and status. As Bell (1981) has stated, for women, relationships with men "represents acceptance as to their equality and sense of worthiness as partners and companions" (p. 104). The literature on cross-sex friendship provide sources of opposition to theories of same-sex bonds. For example, from Lipman-Blumen's theory one might predict that within a societal power structure where men are the primary controllers of resources, women also derive specific relational needs and sources of well-being which can only be obtained through their cross-sex bonds. Lipman-Blumen's theory leads one to predict two hypotheses that oppose those predicted from Bernard's theory. The first hypothesis is that women derive higher levels of the reassurance of worth and help and guidance dimensions through their cross-sex bonds than from their same-sex bonds. The second hypothesis is that a woman's overall satisfaction (the sum of the six relational provisions) with cross-sex bonds is more strongly and positively associated with well-being than their overall satisfaction (the sum of the six relational provisions) with their same-sex bonds. Same-Sex Bonds and Well-Being p. 20 Safilios-Rothschild (1981) has suggested that the cross-sex relationships of men supplies for them (but not for the women involved) "the same central, crucial supportive . . . expressive role" (p. 380) that women obtain through their same-sex relationships. Safilios-Rothschild has speculated that relationships with women compensate for the relational needs men are unable to fulfill through their same-sex relationships. Bell (1981) has agreed with Safilios-Rothschild and has commented that men experience relationships with women "as an important means of self-revealing and gaining a greater sense of self-esteem" (p. 110). Safilios-Rothschild's theory gives rise to four hypotheses that oppose those derived from Lipman-Blumen's theory. The first hypothesis is that overall, women's same-sex bonds are stronger than men's same-sex bonds, but overall, men's cross-sex bonds are stronger than women's cross-sex bonds. The second hypothesis is that, overall, men's cross-sex bonds are stronger than men's same-sex bonds. The third hypothesis is that men's cross-sex bonds provide higher levels of Weiss's reassurance of worth and help and guidance support dimensions than do men's same-sex bonds. The fourth hypothesis is that men's overall satisfaction (the sum of the six relational provisions) with their cross-sex bonds is more strongly and positively associated with well-being than their overall satisfaction with same-sex bonds. The perspectives on social support presented so far have all implied that social relationships are beneficial for well-being in terms of alleviating loneliness (Weiss, 197*) and depression (Bernard, 1976; Weiss, 197*), for Same-Sex Bonds and Well-Being p. 21 enhancing self-esteem (Bell, 1981; Bernard, 1976; Lipman-Blumen, 1976; Safilios-Rothschild, 1981; Weiss, 1974), personal efficacy (Bernard, 1976; Lipman-Blumen, 1976; Weiss, 1974) and life satisfaction (Bernard, 1976; Lipman-Blumen, 1976; Safilios-Rothschild, 1981; Weiss, 1974). However, in these theories, the criteria of well-being has only been implied and has not been guided by theory or by empirical evidence. Therefore, in the next section, the theoretical and empirical evidence directly relating to the concept and criteria of well-being is presented. B. Perspectives on Mental Health and Well-Being Considerable attention has been directed towards illuminating both the concept and the criterion of mental health (Jahoda, 1958). Despite the voluminous literature which has emerged, no clear consensus exists (Strupp <5c Hadley, 1977). The multiplicity of meanings that has been associated with the concept of mental health, stems from whether the judgement is made by society, the mental health professional, or the individual. A large body of literature accompanies each vantage point. In the next section, the concept of mental health corresponding to each perspective is reviewed. 1. Society At the societal level, mental health has been conceptualized in terms of the individual's conformity to assigned social roles and prevailing mores (Strupp & Hadley, 1977). This model has defined mental health "as behavior in Same-Sex Bonds and Well-Being p. 22 accordance with the values and role prescriptions of society, particularly sex-role prescriptions. Since cultural role prescriptions and values are different for men and women, this . .. model implies a double standard of mental health" (Rawlings Sc Carter, 1977 p. 25). Numerous studies (Hjeile Sc Butterfield, 1974; Logan Sc Kaschak, 1980; Woolsey, 1977) have reported that conformity to the female role is inversely related to such indices of mental health as self-actualization, self-esteem, ego strength, and life satisfaction. As such, societal conceptualizations of mental health have been rejected for use in this investigation. 2. Mental Health Theorists and Professionals Mental health professionals have evaluated mental health in terms of sound' personality structure, characterized by growth, development, self-actualization, integration, autonomy, environmental mastery, adaptation, absence of symptomatology, and adequacy of interpersonal relationships (Strupp Sc Hadley, 1977). However, as Rawlings and Carter (1977) have pointed out these evaluations of mental health have been based on androcentric or male-centered models. In addressing this issue, Gilligan's (1977) work has focused on analyzing the thematic progression of the major contemporary developmental theories. Drawing on the influential works of Erikson (psychosocial development), Kohlberg (moral development), and Piaget (cognitive development), Gilligan has Same-Sex Bonds and Well-Being p. 23 traced the arc of individual development. She has described it as a progression from infantile dependencies, self-other fusion and contextually determined thought and judgement towards adult independence, self-other differentiations, and autonomous judgement, and reasoning. Research conducted by Broverman, Broverman, Clarkson, Rosenkrantz and Vogel (1970) and, more recently, by Swenson and Ragucci (198*) have dramatically illustrated that this conceptualization of adult development is equated with typically masculine characteristics, but deemed undesirable in women. The widely cited study by Broverman et al. (1970) has provided strong evidence to indicate the existence of a double standard of mental health. Broverman and her colleagues instructed 79 mental health professionals (3.3 women and *6 men) to outline the qualities of a healthy, mature, socially competent, man, woman, and ideal adult (sex unspecified). As predicted, the mental health professionals have characterized healthy women differently from how they have characterized healthy men. A healthy woman has been described as submissive, less independent, less adventurous, more easily influenced, less aggressive, more excitable in minor crises, more emotional, and less objective than her healthy male counterpart. Moreover, a healthy mature man has been characterized as similar to the ideal image for an adult, sex unspecified. Consequently, standards of health have been judged within a masculine framework, placing women in a conflictual position. As Broverman et al. (1970) have concluded: Same-Sex Bonds and Well-Being Acceptance of an adjustment notion of health, then places women in a conflictual position of having to decide whether to exhibit those positive characteristics considered desirable for men and adults and thus have their "femininity" questioned, that is, be deviant in terms of being a woman; or to behave in the prescribed feminine manner, accept second class adult status, and possibly live a lie to boot (p. 6). Paradoxically, the qualities esteemed in women, their tact, gentleness, awareness and concern for the needs of others and facility in emotional expressiveness have not been distinguished as hallmarks of healthy adult adjustment. On the contrary, the qualities esteemed in women are the very qualities which, according to androcentric models of development, maintain their "fusion" with others and hold them in a style of judgement that is essentially contextual. Consequently, Gilligan has charged that women are considered either deviant or deficient in their development. It is apparent that conceptions of healthy adult development have failed to consider both the experience and socialization of women (Gilligan, 1977). In noting the differential socialization patterns of men and women, Bardwick (1971), Chodorow, (1978), and Miller (1976) have pointed out that while all children begin life deeply attached to parents, boys are encouraged to seek independence while girls are encouraged to maintain affiliative attachments. Miller (1976) has recontextualized a woman's greater needs for affiliation and Same-Sex Bonds and Well-Being p. 25 interdependence as strengths which facilitate her individual development. Thus, the establishment and maintenance of attachment bonds have been considered more central to the well-being of women than to the well-being of men. Conversely, success in career and achievement-oriented pursuits has been considered more central to the well-being of men than to the well-being of women (Pearlin, 1975). 3. Individual According to Strupp and Hadley (1977), individuals regard themselves as mentally healthy so long as they experience a sense of well-being and happiness. The individual's evaluation of his/her mental health may or may not be consistent with societal standards of mental health, but is believed to have a validity all its own. Over the last ten years, considerable attention has been directed towards outlining the factors which contribute to subjective well-being. Studies have focused on the demographic, personality, behavioral, and biological correlates of well-being (Diener, 198*). The majority of studies have examined the demographic indices of well-being, such as age, gender, education, affluence, religion, race, marital status, and employment. The findings are generally mixed. More important, these factors are not strongly associated with well-being. Same-Sex Bonds and Well-Being p. 26 Several other variables have demonstrated consistent relationships with subjective well-being. These are: self-esteem (Anderson, 1977; Czaja, 1975); internality (Baker, 1977; Brandt, 1980); perceived choice and control in one's life (Eisenberg, 1981); perceived personal competence (Campbell, 1976); extraversion and sociability (Gorman, 1972; Tolor, 1978); androgyny for women, but not for men (Allan-Kee, 1980); satisfaction in a love relationship (Diener, 1984); and finally, perceived life satisfaction (Diener, 1984). With respect to social behavior, social contact (Wilson, 1967), satisfaction with friends (Anderson, 1977; Campbell, 1976), and social participation (Bradburn, 1969) have emerged as significant factors positively affecting well-being. However, little is known about what types of social contact contribute to well-being (Diener, 1984). At this point, a comprehensive and psychometrically sound instrument designed to measure well-being has not been developed. Therefore, in the present study, the investigator selected six indices of well-being that were both consistent with the components of well-being identified in the theories of Bernard (1976), Bell (1981), Lipman-Blumen (1976), Safilios-Rothschild (1981) and Weiss (1974) and, that were empirically supported in the literature to reflect the subcomponents of well-being. Thus, the following six indices of well-being were selected: self-esteem, personal efficacy, interpersonal control, life satisfaction, loneliness, and depression. In the present study, these dimensions of well-being were statistically combined forming a molar variable identified as "Global Well-Being". Same-Sex Bonds and Well-Being p. 27 C. Statement of the Problem In the last decade, related, but unintegrated areas of the literature have drawn attention to the general association between socially supportive relationships and well-being. Researchers have now been challenged to progress beyond the demonstration of a general association between social support and well-being to a greater refinement of the specific conditions that contribute to this relationship (Cohen & Syme, 1985; Cohen & Wills, 1985; Diener, 198*; Reis, 198*). One variable, emerging from the literature on homosociality and friendship is the "gender composition" (same-sex, cross-sex) of dyadic relationships. However, both the relevance of this variable and the direction of its significance have been debated. For example, theories of male (Lipman-Blumen, 1976) and female bonding (Bernard, 1976) have drawn attention to the greater effectiveness of same-sex bonds than cross-sex bonds for supplying higher levels of social support and well-being. Conversely, theories of female (Lipman-Blumen, 1976, Bell, 1982) and male (Safilios-Rothschild, 1981) cross-sex bonding have drawn attention to the greater effectiveness of cross-sex bonds than same-sex bonds for supplying higher levels of social support and well-being. And finally, in opposition to both same-sex and cross-sex bonding theorists, Weiss has implied that same-sex and cross-sex bonds are equally effective in supplying particular dimensions of social support and well-being. Same-Sex Bonds and Well-Being p. 28 The purpose of the study was to test the theoretical propositions derived from these theories in order to determine whether relationship gender composition influences the attainment of social support and the association between social support and well-being. In the section to follow, the objectives of the study are outlined. D. Objectives of the Study The overall objective of the study was to examine whether the variables "sex of respondent" and "relationship gender composition" are significant factors affecting the attainment of social support and global well-being. In addressing this overall objective, a conceptual distinction was made between the effect that "sex of respondent" and "relationship gender composition" have on the attainment of social support and the effect that "sex of respondent" and "relationship gender composition" have on the relationship between social support and well-being. In addressing this objective, the investigator posed four research questions. The first two questions deal with the attainment of social support and the second two questions deal with the relationship between social support and well-being. 1) Overall, are men's same-sex relationships stronger than women's same-sex relationships? 2) Do women's and men's same-sex bonds provide higher levels of the individual relational provisions than women's and men's cross-sex bonds? Same-Sex Bonds and Well-Being p. 29 3) Are satisfying social bonds more strongly associated with the global well-being of women than with the global well-being of men? *) Are satisfying same-sex bonds more strongly associated with global well-being than satisfying cross-sex bonds? In the section to follow, the hypotheses corresponding to these four research questions are outlined. E. Research Hypotheses The research hypotheses are divided into three parts. In part one, hypotheses relating to the effects of "sex of respondent" and "relationship gender composition" on the overall level of social support are outlined. Part one addresses the first research question. In part two, hypotheses relating to the effects of "sex of respondent" and "relationship gender composition" on the level of individual relational provisions are outlined. Part two addresses the second research question. In part three, hypotheses relating to the effects of "sex of resondent" and "relationship gender composition" on the association between level of social support and global well-being are outlined. Part three addresses the third and fourth research questions. Where appropriate, the hypotheses are organized into oppositional sets, because mutually exclusive truths of the various theories are being tested. Same-Sex Bonds and Well-Being p. 30 Part One; Hypotheses Relating to the Effects of Sex of Respondent and  Relationship Gender Composition on the Overall Level of Social Support 1. The investigator predicted that the total mean provision scores would be higher for the same-sex relationships of men and women than the total mean provision scores for the cross-sex relationships of men and women. 2. It was predicted from Safilios-Rothschild's theory that the total mean provision score for the same-sex relationships of women would be higher than the total mean provision score for the same-sex relationships of men but that the total mean provision score for the cross-sex relationships of men would be higher than the total mean provision score for the cross-sex relationships of women. 3. In opposition to the hypothesis derived from Safilios-Rothschild, it was predicted from Lipman-Blumen's theory that the total mean provision score for the same-sex relationships of men would be higher than the total mean provision score for the same-sex relationships of women, but that the total mean provision score for the cross-sex relationships of women would be higher than the total mean provision score for the cross-sex relationships of men. 4. It was predicted from Lipman-Blumen's theory that for men, the total mean provision score would be higher for their same-sex relationships than for their cross-sex relationships. Same-Sex Bonds and Well-Being p. 31 5. In opposition to the hypothesis predicted from Lipman-Blumen, it was predicted from Safilios-Rothschild's theory that for men, the total mean provision score would be higher for their cross-sex relationships than for their same-sex relationships. Part Two: Hypotheses Relating to the Effects of Sex of Respondent and  Relationship Gender Composition on the Level of Individual Relational Provisions 6. It was predicted from Bernard's theory that for women, the mean scores on the Attachment, Social Integration, Reassurance of Worth, Reliable Alliance, Help and Guidance and Opportunity for Nurturance provisions would be higher for their same-sex relationships than for their cross-sex relationships. 7. In opposition to the hypothesis predicted from Bernard, it was predicted from Lipman-Blumen's theory that for women, mean scores on the Reassurance of Worth and Help and Guidance provisions would be higher for their cross-sex relationships than for their same-sex relationships. 8. In opposition to the hypotheses predicted from Bernard and Lipman-Blumen, it was predicted from Weiss's theory that for the women, there would be no significant differences between their same-sex and cross-sex provisions scores. Same-Sex Bonds and Well-Being p. 32 9. It was predicted from Lipman-Blumen's theory that for men, mean scores on the Social Integration, Reassurance of Worth, Reliable Alliance, and Help and Guidance provisions would be higher for their same-sex relationships than for their cross-sex relationships. 10. In opposition to the hypothesis predicted from Lipman-Blumen, it was predicted from Bell's and Safilios-Rothschild's theories that for men, mean scores on the Reassurance of Worth, and Help and Guidance provisions would be higher for their cross-sex relationships than for their same-sex relationships. 11. In opposition to the hypotheses predicted from Lipman-Blumen, Bell and Safilios-Rothchild, it was predicted from Weiss's theory that for the men, there would be no significant differences between their same-sex and cross-sex scores on the Social Integration, Reassurance of Worth, Reliable Alliance, and Help and Guidance provisions. Part Three; Hypotheses Relating to the Effects of Sex of Respondent and  Relationship Gender Composition on the Association between the Overall Level  of Social Support and Global Well-Being 12. It was predicted from Weiss's theory that higher total mean provision scores (combining total scores on same-sex provisions and total scores on cross-sex provisions) would be equally associated Same-Sex Bonds and Well-Being p. 33 with a significant positive correlation with outcome measures of Global Well-Being for men and women. 13. In opposition to the hypothesis derived from Weiss's theory, it was predicted from Bernard's and Miller's theories that the magnitude of the correlation between total mean provision scores (combining total scores on same-sex provisions and total scores on cross-sex provisions) would be more positively associated with outcome measures of Global Well-Being for women than for men. 1*. It was predicted from Bernard's theory that for women, the magnitude of the correlation between their total same-sex provision scores would be more positively associated with measures of Global Well-Being than would be their total cross-sex provision scores. 15. In opposition to the hypothesis derived from Bernard, it was predicted from Lipman-Blumen's theory that for women, the magnitude of the correlation between their total cross-sex bond provision score and measures of Global Well-Being would be significantly stronger than the magnitude of the correlation between their total same-sex bond provision score and measures of Global Well-Being. Same-Sex Bonds and Weil-Being p. 34 16. It was predicted from Lipman-Blumen's theory that for men, the magnitude of the correlation between their total same-sex bond provision score and measures of Global Well-Being would be significantly stronger than the magnitude of the correlation between their total cross-sex bond provision score and measures of Global Well-Being. 17. In opposition to the hypothesis derived from Lipman-Blumen, it was predicted from Bell's and Safilios-Rothschild's theories that for men, the magnitude of the correlation between their total cross-sex bond provision score and measures of Global Well-Being would be significantly stronger than the magnitude of the correlation between their total same-sex bond provision score and measures of Global Well-Being. 18. In opposition to the hypotheses derived from Bernard, Lipman-Blumen, Bell and Safilios-Rothschild, it was predicted from Weiss's theory that for both men and women the magnitude of the correlation between their total same-sex bond provision score and measures of Global Well-Being would not be significantly different from the magnitude of the correlation between their total cross-sex bond provision score and measures of Global Well-Being. Same-Sex Bonds and Well-Being p. 35 F. Significance of the Study A considerable body of literature from independent sources has indicated a link between social support and well-being. Pioneering social support researchers (Cobb, 1976; Miller <5c Ingham, 1976; Tolsdorf, 1976) have drawn attention to the function of social support as a buffer or moderator of stress during periods of physical and emotional adversity. This crisis-oriented conceptualization of social support has influenced the form and focus of subsequent investigations, and researchers have generally failed to move beyond this initial orientation. Recent literature (Sandler ic Barerra, 1984) has begun to reconceptualize social support within the context of basic adult needs and well-being in a normative sense. Sandler et al. (198*) have cited Weiss's (197*) theory of the provisions of social relationships as an example of this approach. However, an empirical validation of Weiss's conceptualization of social support has not been investigated. If the satisfaction of relational needs indicates a strong association with well-being, in a normative sense, as is hypothesized in this study, then such a result would have implications for theories of adult development and social support. Theories of adult development have generally focused on issues of autonomy, separateness, and self-sufficiency as the hallmarks of healthy adult development (Broverman, Broverman, Clarkson, Rosenkrantz & Vogel, 1970). However, if the satisfaction of relational needs indicates a strong association with well-being in a normative sense, as is hypothesized, then theories of adult development must recontextualize the importance of social relationships to Same-Sex Bonds and Well-Being p. 36 health, rather than to pathology, and as an indication of strength and not as an indication of dependency or weakness. This position has begun to receive recognition in Miller's (1976) theory of female psychology, but so far has not been tested. Therefore, the present investigation is of scholarly significance in that it tests Weiss's theory of social relationships as a theory of well-being. The present investigation also holds potential significance for theory development in social support. It may broaden existing conceptualizations of social support by indicating the contribution of socially supportive relationships to well-being in the everyday lives of a non-crisis, non-clinical sample of men and women. While researchers in the area of well-being have drawn attention to the general association between sociality and well-being (Bradburn, 1969; Diener, 1984), the specific factors influencing this relationship are currently unknown. One critical variable that has emerged from the literature on friendship and homosociality for the fulfillment of relational needs and well-being is same-sex or homosocial bonds. However, the critical importance placed on the heterosexual pair bond in adulthood, both as the basic unit of society and as the basic medium for the fulfillment of adult relational needs, may have influenced theorists, researchers and individuals to overlook the vital importance of same-sex bonds as a valuable resource for meeting individual relational needs and well-being (Bernard, 1976). This investigation draws attention to the importance of gender composition as one specific parameter influencing the attainment of social support and the relationship between social support and well-being. Thus, Same-Sex Bonds and Well-Being p. 37 if same-sex bonds are shown to bear a stronger association to the fulfillment of specific dimensions of social support and well-being than cross-sex bonds as Bernard (1976) and Lipman-Blumen (1976) have hypothesized, then, such a result would have implications for counselling. With respect to individual counselling, recognition of the importance of satisfying same-sex bonds for the fulfillment of specific dimensions of social support and to well-being may serve to encourage the development of such bonds. With respect to marriage counselling, an understanding of how differences in relational styles affect patterns of communication and subsequent need fulfillment may promote, normalize, and validate the importance of same-sex bonds for the fulfillment of needs spouses have not been socialized to provide. If, as hypothesized, higher levels of relational provisions are derived through same-sex bonds, the results of this investigation may encourage the revitalization of same-sex bonds as a valid and important resource for the fulfillment of relational needs necessary for adult well-being. G. Limitations of the Study This investigation was limited in two major respects. First, the generalizability of the results was limited to married men and women between the ages of 25 and 45. Research on social support has indicated that network composition varies as a function of age, marital status, income, and life stage in general (Reisman, 1981; Shulman, 1980). While there has been evidence to suggest that the functions served by social relationships are consistent across the Same-Sex Bonds and Well-Being p. 38 life cycle (Candy, Troll <3c Levy, 1981), it has also been recognized that the functions served by social relationships diminish or increase in importance as a function of life stage (Weiss, 1973). As such, the subjects in the sample were intended to represent a homogeneous population. Thus, generalizations can only be made with respect to the sample characteristics included. Second, this study was also limited in that clear causal relationships between social bonds and well-being could not be made. As other social support investigators have commented (Leavy, 1983; Reis, 198*), it is possible that subjects who score highly on measures of well-being are more socially oriented. Thus, the direction of effects between social support and well-being could not be determined in this investigation. Same-Sex Bonds and Well-Being p. 39 CHAPTER II: REVIEW OF THE LITERATURE This chapter outlines the theory and research evidence related to Weiss's (197*) conceptualization of the provisions of social relationships as well as the theory and research related to the nature of same-sex bonding. A. Weiss's Theory of Social Relationships Drawing on his work with groups deficient in various aspects of their social networks, Weiss (1969; 1973; 197*; 1982) has developed a typology of provisions supplied by social relationships and their importance for individual well-being. These provisions are: attachment, social integration, reassurance of worth, reliable alliance, help and guidance, and the opportunity for nurturance. Weiss has defined the first provision as attachment. Relationships providing attachment give the individual a "sense of security and place". According to Weiss, for most people, attachment is provided through heterosexual bonds, or, in some cases by relationships with a close friend, or kin. In his typology, the fulfillment of emotional integration through the satisfaction of attachment needs occupies "a central importance in the establishment of life organization; we might think of individuals organizing their lives around whatever relationship provides them with attachment" (197*, p. 25). Weiss has reasoned that deficits in this relational provision result in the subjective experience of emotional loneliness and depression. Same-Sex Bonds and Well-Being p. 40 Weiss has defined the second provision as social integration. In this provision, needs for sharing a common set of concerns, exchange of information and services and a shared reality are met. Social integration needs are provided by both dyadic and network linkages, primarily through friendships. Deficits in social integration have been hypothesized to result in social loneliness, a sense of marginality, anxiety, and boredom. Reassurance of worth, the third provision, is supplied through recognition of one's competence in a social role. Weiss has proposed that this provision is supplied for people through work relationships. For homemakers, reassurance of worth can be obtained through spouses, children and friends. Weiss has implied that deficits in this provision result in lowered self-esteem. Weiss has defined the fourth provision as a sense of reliable alliance, which is primarily provided by kin. As he has said, "only within kin ties, expecially those between siblings or lineal kin can one expect continuing assistance whether there is mutual affection or not, whether one has reciprocated for past help or not" (p. 24). Weiss has hypothesized that deficits in this provision contribute to feelings of vulnerability and isolation. The fifth provision, the obtaining of help and guidance, is important during periods of stress. Weiss has reasoned that this provision is obtained from relationships with a trusted authority figure who provides support and problem-solving strategies. The absence of this provision may leave individuals with feelings of powerlessness and anxiety. Same-Sex Bonds and Well-Being p. 41 Opportunity for nurturance is the final provision and is fulfilled by relationships in which an adult takes responsibility for a child. Here, Weiss has focused on the need for meaning and sense of purpose through commitment to another human being. It is the only provision in which support has been conceptualized within a context of giving rather than getting. Drawing on his studies, Weiss (1973) has identified two types of loneliness, which he has defined as emotional and social loneliness. Emotional loneliness "appears in the absence of a close emotional attachment and can only be remedied by the integration of another emotional attachment, or the reintegration of the one that had been lost" (p. 17). "Conversely, . . . the form of loneliness associated with the absence of an engaging social network, the loneliness of social isolation, can only be remedied by access to such a network" (p. 18). Evidence to support Weiss's theory of two distinct forms of loneliness has been drawn from two investigations. The first involved newly separated men and women who sought support through an organization called Parents Without Partners. Weiss found that even though these men and women had a supportive network of friends, they suffered from feelings of loneliness associated with the loss of a spouse. In another investigation, Weiss studied couples who moved two states away, leaying behind supportive friendship ties. The husbands became quickly absorbed into a community of work associates and did not experience deficits in their social relations. However, despite the close relationships between the husbands and wives, Weiss found that, in the absence of a close Same-Sex Bonds and Well-Being p. 42 network of women friends, these women had been vulnerable to loneliness. "They wanted access to a network of women with whom they might establish and then discuss issues of common concern, shopping, home management, the developing lives of their children, and of course one another" (p. 18). These studies have led Weiss to suggest that there are two distinctive forms of loneliness in response to specific relational deficits. Moreover, he has hypothesized that these relational needs can only be satisfied by a specific type of relationship. Weiss has suggested that loneliness is a biological signalling response which contributes to species survival. He has conceptualized it as a biologically adaptive mechanism for maintaining proximity to others, for ensuring safety, and for facilitating well-being. Recent empirically based investigations by Russell, Cutrona, Rose and Yurko (1984) and by Shaver and Rubenstein (1982) have provided additional evidence to support Weiss's distinctions between emotional and social loneliness. In order to test Weiss's theory, Russell et al. (1984) administered the Social Provisions Scale, measures of depression, anxiety and two measures of loneliness to 505 college men and women. Although the results of the study have supported Weiss's formulation of two distinct forms of loneliness, the results have also indicated that the two types of loneliness share a sizable core of common experiences. While deficits in the attachment provision have emerged as the strongest predictor of emotional loneliness, deficits in the social integration provision have not emerged as the most predictive variable of social loneliness. Deficits in the reassurance of worth provision have emerged as the strongest Same-Sex Bonds and Weil-Being p. *3 predictor of social loneliness. The social integration provision has been significantly correlated with the reassurance of worth, reliable alliance and guidance provisions. Consequently all four provisions have been correlated with social loneliness (ranging from -.23 to -.30). "Thus loneliness is apparently related to several different types of relational deficits that result from the lack of a social network" (p. 1320). As expected, social loneliness has been predicted by deficits in friendships and emotional loneliness has been predicted by deficits in romantic or dating relationships. Depression has been significantly related to social loneliness, as Weiss predicted. However, contrary to his theory, depression has been predicted by the degree of emotional loneliness reported, and anxiety has been predicted by social loneliness. Russell, Cutrona, Rose and Yurko (198*) have commented that when sex of respondent was analyzed separately, the differences in loneliness have not been significant. However, they have not included the gender composition of the relationship (same-sex and cross-sex) in their analyses, an important omission in their study. The study by Wheeler, Reis and Nezlek (1983) has pointed to the importance of relationship gender composition as a critical determinant of relational quality and loneliness. Wheeler, Reis and Nezlek (1983) have conducted a study concerning the relationship of actual social behaviour to loneliness. More specifically, Wheeler et al. have examined sex differences in relationship quality, a variable which has been neglected in past investigations. Consistent with Bernard's (1976) theory of homosociality, these authors have reasoned that males and females differ with Same-Sex Bonds and Well-Being p. 44 respect to their social styles and in their effectiveness as social partners, with women the more socially responsive, empathic, and intimate. Consequently, they have hypothesized that the quantity of time spent with women and the quality of time spent with men would be inversely related to loneliness. These hypotheses were confirmed. The quantity of time spent with females was inversely related to loneliness for both sexes. Moreover, women who allocated more of their social time with males were more lonely. This finding is consistent with Bernard's application of Weiss's (1974) relational deficit thesis. Moreover, for both males and females, relationship quality or meaningfulness with males (as a composite score of disclosure, intimacy, pleasantness and satisfaction) was related to less loneliness. "The implication is that both sexes need the same qualities in their interactions to avoid loneliness but that females are more adept at providing them" (p. 947). However, germane to the present study was the finding that meaningfulness with males was more important than meaningfulness with females for warding off loneliness. These researchers have suggested that because intimacy with males is encountered less frequently, it may contain greater reward value. Since loneliness correlates with low self-esteem and powerlessness Uones, Freeman «5c Goswick, 1981), this finding indicates support for Lipman-Blumen's thesis, that both women and men derive a sense of self-esteem and power through their relationships with men. B. Theories of Homosociality Recent interest in homosociality has been stimulated by the controversial anthropological work of Tiger (1969). Tiger has developed the thesis that men Same-Sex Bonds and Well-Being p. *5 have a genetically determined and culturally reinforced predilection to bond with one another. According to Tiger, bonds between men have been stronger and more stable than those between women. He has speculated that the early evolutionary history of men as hunters and warriors necessitated co-operative bonding among men in order to ensure their survival. This early structure of male bonding further evolved and became linked to political and economic dominance hierarchies. In Tiger's view, female organizations have been less numerous and less stable because women lack a biological prediliction towards bonding. He has said that when female organizations do exist their relative obscurity "and their apparent unimportance for the macro-life of the community is striking . . . nor do they appear to be persistent over time" (pp. 93-9*). Tiger has invoked a biological explanation to account for the exclusion of women from societal power structures. He has contended that females have an "underlying inability—at the ethological level of pattern releasing behavior to affect the behavior of subordinates" (p. 96). Tiger's work has stimulated subsequent empirical investigations into the greater strength, stability and superiority of male bonds. However, research comparing the same-sex bonds of men and women have generally failed to support Tiger's propositions (Booth, 1972; Booth & Hess, 197*; Gibb, Auerbach, Fox, 1980; Rose, 1985; Wright, 1982). Although the generalized exclusion of women from positions of political and economic stature is generally acknowledged, subsequent theories, notably that of Lipman-Blumen (1976) have rejected biological explanations for this occurrence. Rather, Lipman-Blumen has identified the generalized non-Same-Sex Bonds and Well-Being p. 46 participation of women from positions of leadership to be a culturally learned, culturally maintained and culturally reinforced phenomenon. In the sections to follow, Lipman-Blumen's (1976) and Bernard's (1976) theories of homosociality are presented. 1. Bonds between Men Lipman-Blumen (1976) has defined "homosocial" as the "seeking, enjoyment and/or preference for the company of the same sex" (p. 16). Lipman-Blumen has distinguished homosociality from homosexuality, but has added that under some circumstances relationships between members of the same sex may include an erotic element. Although Lipman-Blumen has conceptualized the term "homosocial" to describe the social patterns of both sexes, the application of her theory has focused on the homosocial world of men and the needs men have fulfilled through their bonds with one another. She has focused on male homosocial bonds in order to account for the generalized exclusion of women from positions of power and purveyors of valuable resources in historical and contemporary society. She has argued that sexual segregation and the exclusion of women from occupational, political and economic domains can be understood within a theory of male homosocial bonds. She has identified her theory as one of the "middle range", a term coined by Merton (1949) to describe theories which are "applicable to a limited range of data" (p. 9). Merton has encouraged the use of middle range theories for the intensive study of specific social phenomena. Same-Sex Bonds and Well-Being p. *7 The central proposition of Lipman-Blumen's theory has stated that "men are attracted to, stimulated by and interested in other men" (p. 15). In her view, the basis of male bonding lay in the recognition of men as the primary controllers of power and resources. She has suggested that association with this world satisfies the need for power, status and self-esteem. As Lipman-Blumen (1976) has stated: The pragmatic recognition that males controlled economic political, educational, occupational, legal, and social resources created a situation in which men identified with and sought help from other men. Women, recognizing the existential validity of the situation, also turned to men for help and protection. By now, . it is practically a psychological truism that individuals identify with others whom they perceive to be the controllers of resources in any given situation (p. 16). She has also stated that men satisfy the majority of their needs through their relationships within the male homosocial world. "Men can and commonly do seek satisfaction for most of their needs from other men. They can derive satisfaction for their intellectual, physical, political, economic, occupational, social, power and status needs and in some circumstances their sexual needs--from other men" (p. 16). According to Lipman-Blumen, the only restriction imposed by the male homosocial world is that it cannot supply men with what she refers to as the ultimate validation of their masculinity—paternity. She has suggested that the desire for offspring is a basic need. Same-Sex Bonds and Well-Being p. 48 Moreover, because women traditionally do not hold positions of power, Lipman-Blumen has considered them to be less attractive, less interesting and less stimulating to both sexes. As Lipman Blumen (1976) has commented: This uneven array of resources systematically made men more interesting to women, women less interesting and useful to other women, and women fairly often unnecessary and/or burdensome to men. This disparity of resources made it apparent that men were the most valued social beings (p. 18). Lipman-Blumen has noted that in recent years the embryonic redistribution of power has witnessed the emergence of a "new girl network". Lipman-Blumen has predicted that as women lobby for and occupy greater representation in positions of power, women will become more interesting to women and eventually more interesting to men. However, according to Lipman-Blumen, the existence of a female homosocial network outside the domestic sphere is still in its infancy. Lipman-Blumen has identified her theory as an extension and application of existing psychological theory drawn from the analytic model. The basis of male bonding as the attraction to power is reminiscent of "identification with the aggressor", a defense mechanism the early analytic writers have identified and the "will to power" concept Adler (1927) has identified. Additional support has been drawn from the social learning theorists. Bandura's (1963) studies of children have indicated that from a young age both boys and girls identify with Same-Sex Bonds and Well-Being p. *9 and model the behavior of those perceived to be the controller of resources. Lipman-Blumen has considered the attraction to power the central motivating factor in male bonding and in female cross-sex bonding. She has cited evidence to indicate that an interest in power through the possession of material objects characterizes the socialization process of males to a greater extent than that of females and that women only derive a sense of power vicariously through their association with the male world. Lipman-Blumen has speculated, as did Tiger (1969), that "the origin of male control over valued resources lay in their early roles as hunters and warriors, roles which placed men in positions of protecting and acquiring territory and food—fundamental resources in any society" (p. 17). She has contended that these roles created dominance hierarchies which have continued long after technology eliminated their necessity. However, she has failed to account for the reason why greater value was placed on the roles occupied by men than the child-bearing and child protecting roles occupied by women. Consistent with Lipman-Blumen's (1976) conceptualization of male bonding, Shaver and Buhrmester (1983) have commented on the benefits men derive from integrated group involvement. They have cited evidence to suggest that "males derive much of their self-esteem, feelings of belonging and sense of meaning and engagement from participation in organized groups" (p. 272). Group involvement provides a sense of being needed "not as a nurturer, or unique personality but as a competent, talented occupant of certain social roles" (p. 267). Same-Sex Bonds and Well-Being p. 50 The results of a recent study conducted by Rose (1985) are particularly relevant to the present investigation. In the only known study designed to test Lipman-Blumen's theory, Rose has examined the homosocial preferences and friendship patterns of a group of married and single men and women. Contrary to Lipman-Blumen's (1976) theory, Rose has concluded that women are more homosocial than men and that same-sex friendship preferences are more evident among the married than the single respondents. However, Rose restricted her study of homosociality to friendships, and did not include other significant homosocial ties with family members, work colleagues, supervisors, and community members. The literature on male bonding cited in the previous sections has focused on outlining both the value and strength of male same-sex bonds. However, attention has also been directed towards outlining the factors which create barriers to the development of male same-sex bonds. Lewis (1978) and Pleck (1976) have suggested that the male sex-role emphasizes manliness, competition and emotional control. Consequently, closeness between men ignites homophobic fears, and the expression of vulnerability is experienced and viewed as a betrayal of one's masculinity. In a study concerning need for affiliation, Shipley and Veroff (1952) have discovered that men who express strong affiliative needs are negatively judged as 'dependent' and 'approval seeking' by other men. Thus, male same-sex relationships appear to be rule-governed and constrained by sex-role socialization. Pleck (1975) has identified these issues and suggested that the inherent constraints in patterns of male sex-role socialization be re-evaluated and challenged. Same-Sex Bonds and Well-Being p. 51 2. Bonds between Women Bernard (1976) has defined homosociality "as the different ways that men and women relate to members of their own sex" (pp. 227-8). Bernard has linked her theory of homosociality with Weiss's (197*) concept of "relational deficit" to account for the higher incidence of depression among women over this last century. Bernard has characterized the social style of men as one which is oriented towards autonomy, separation, and emotional inexpressivity and the social style of women as one which is oriented towards attachment, affiliation, and emotional expressivity. She has suggested that because women are socialized towards meeting their needs through relationships in general and through heterosexual relationships in particular, "they are more vulnerable to the stresses of deprivation of such ties and hence to depression" (p. 228). Bernard has cited Radloff's (1975) application of Seligman's (197*) concept of "learned helplessness" to the phenomenon of female depression. As Bernard has said, "the feminine role prescribes for women crippling attitudes, beliefs and behaviors which in and of themselves generate depression" (p. 225). While Bernard has suggested that role theory does not offer a complete explanation of female depression, she has extracted the concepts of power/mastery and a lack of control in one's life as important components to female depression. Other studies have found that women who do not conform to the female role are correspondingly higher on measures of self-actualization (Hjelle & Butterfield, 197*); well-being (Logan & Kaschak, 1980) and ego strength (Gump, 1972). Same-Sex Bonds and Well-Being p. 52 Bernard has drawn attention to the importance of female homosocial bonds in order to compensate for and to supply the relational needs men are not socialized to provide. To support her thesis, Bernard has drawn on the work of Carroll Smith-Rosenberg. Smith-Rosenberg (1975) has conducted an historical analysis of female homosocial bonds in the nineteenth century. Her analysis has revealed the critical role that female bonds had served for providing the emotional support and understanding men were unable to give. In commenting on Smith-Rosenberg's work, Bernard (1976) has written: Women in the nineteenth century could find bonding, attachment, affiliation and intimacy with other women, so that the deficit in response from men was compensated for by attachments with women. Such homosocial affiliative ties were approved of, even encouraged, and women were not embarrassed by or ashamed of them (p. 229). Several factors have encouraged female homosociality, a term which Smith-Rosenberg has used to describe the female world of love and ritual in the nineteenth century. Smith-Rosenberg has focused on the shared biological experiences of pregnancy, childbirth, nursing and menopause as the underpinnings of female homosociality. As Smith-Rosenberg (1975) has stated: The biological realities of frequent pregnancies, childbirth, nursing and menopause bound women together in physical and emotional intimacy. It was within just such a social framework . . . Same-Sex Bonds and Well-Being that a specifically female world did indeed develop, a world built around a generic and unselfconscious pattern of single sex or homosocial networks. These supportive networks were institutionalized in social conventions or rituals which accompanied virtually every important event in a woman's life from birth to death (p. 2*). The basis of female homosociality in the nineteenth century was a celebration and validation of female abilities, capacities and concerns free from the subordination and inferiority imposed by the male world. "In this secure and empathic world women shared sorrows, anxieties and joys with certain knowledge that other women had experienced them also." (Bernard 1976, p. 229). In this world emotional expressiveness was cherished and considered a positive quality. Moreover, women considered themselves different from but not lesser than men (Cott, 1970). Bernard (1976, 1981), has traced some of the salient forces leading to the historical decline of female homosociality. First, opportunities for geographic mobility and immigration prevented and eroded the maintenance of strong bonds. Moreover, the impact of Freudian psychiatry had a significant bearing on the norms governing homosocial ties: strong homosocial ties became sexually suspect, and cultural taboos against homosexuality discouraged the development of same-sex bonding. Consequently, from an early age girls were encouraged to seek out heterosexual relationships. Psychiatry also discouraged talking to Same-Sex Bonds and Well-Being p. 54 friends about personal problems. As a consequence, Bernard (1981) has commented that: the loss of the nineteenth century type of friendship left women in a particularly vulnerable position since the male world did not offer the same emotional support that women friends had supplied in the past . . . The very concept of female friendship became downgraded (p. 105). Bernard's theory of homosociality has underscored the importance of female bonding for supplying essential relational provisions of support, mutual exchange, and the sharing of a common reality. However, Bernard has also drawn attention to the critical link between the satisfaction of relational needs met through homosocial bonds and its effect on self-esteem and on a sense of control and mastery over one's life. She has cited Silverman's (1976) conclusion that "such mutual help increases the average individual's self-reliance and his (her) ability to take charge of his (her) own life" (p. 234). Bernard has agreed with Miller's (1976) contention that women derive a sense of personal power and effectiveness through their relationships with other women. The work of Davidson (1983) and of Candy, Troll, and Levy (1981) have complemented Bernard's propositions regarding the importance of female bonds. Davidson has described the bond between women as "the felt connection" based on the shared reality of daily life concerns. As Davidson (1983) has said: Same-Sex Bonds and Well-Being p. 55 Friendship between women contributes to the personal growth, support and change of the women who share it. Depression, an experience all too common to many women, is often related to feelings of powerlessness. The fact that women identify their friendships as a context where they give and receive advice, understand and are understood, actively communicate and stimulate, suggests that women in their friendships with other women experience some degree of impact and control—power, personal power (p. 10). Candy, Troll, and Levy (1981) have investigated the functions served by friendship between women across six age groups. These researchers extracted seven salient dimensions of friendship that have been noted in the literature. These are: gregariousness, intimacy, commonality, assistance, compatibility, status and power. Women from six age groups were asked to describe their relationships with their five closest friends in terms of the dimensions outlined. A principle-axis solution with varimax rotation was run across all subjects for best friend. Three orthogonal factors emerged across all age groups. These are intimacy-assistance, status and power. These three factors accounted for *2% of the variance. However, these researchers did not investigate the cross-sex relationships of women. Consequently, it is unclear whether these functions are unique to female friendship bonds since no comparisons were made. A recent study conducted by Goodenow (1985) has examined the association of women's same-sex friendships to their well-being. Goodenow has Same-Sex Bonds and Well-Being p. 56 assessed the differential contribution of Wright's (1982) relationship dimensions (friendship strength, maintenance difficulty, utility value, and affirmation value) as well as several other dimensions (role similarity, attitude similarity, status value, and perceived equality) to indices of well-being (self-esteem, depression, identity integration, and life satisfaction) in a sample of married and unmarried women. Using a backward elimination procedure, Goodenow has found that for the married women the dimensions of affirmation value, perceived equality, and low maintenance difficulty have emerged as the most predictive components of well-being, respectively. Paralleling Weiss's (1974) reassurance of worth provision, affirmation value involves "the extent to which friends were emotionally encouraging, treated the participant as worthwhile, supported her individuality and enabled her to be herself" (p. 5). Further, contrary to expectation, Goodenow has found that the same-sex relationships of married women demonstrated a stronger relationship to indices of well-being than the same-sex relationships of single women. To account for this finding, Goodenow has suggested that women's same-sex relationships may become more significant once the need for a stable heterosexual relationship has been satisfied. The literature which has been cited so far has focused on the more positive aspects of women's same-sex relationships. However, attention has also been directed towards delineating the factors which undermine the development of female homosocial bonds. Caplan (1981) has illuminated the prevailing psychological and social myths which create and perpetuate barriers to the formation of female homosocial bonds. Caplan has identified four myths: first, that males are better than females; second, that females are naturally Same-Sex Bonds and Well-Being p. 57 constrained and tidy; third, that females have unmeetable needs for love and nurturance; and, finally, that females are naturally and endlessly nurturant. Caplan has explored how these myths influence the development of the mother-daughter bond, the most fundamental homosocial relationship. She has focused on the daughter's unmet needs for nurturance (the mother witholds nurturance because of her own homophobic fears) and the daughter's ambivalence in accepting the mother as a role model (because of her disparaged status as a female). Caplan has articulated how these early conflicts and sources of ambivalence continue to characterize the daughter's subsequent homosocial relationships. Caplan has suggested that these issues create barriers in the formation of strong affiliative same-sex relationships between women and has recommended that these myths be recognized and challenged in order to enhance women's same-sex relationships. Despite gender differences in the social styles of men and women and in the nature of the bond, Bernard's theory of female homosociality and Lipman-Blumen's theory of male homosociality have indicated critical points of convergence and of tension. Both theorists have emphasized the importance of homosocial bonds for providing specific relational needs which cannot be adequately supplied through cross-sex bonds. Moreover, each theorist has conceptualized homosocial bonds as a vital resource for enhancing self-esteem and a sense of personal efficacy. The critical point of tension between Bernard's and Lipman-Blumen's argument has been on the basis of the bond between members of the same sex. Same-Sex Bonds and Well-Being p. 58 Bernard has viewed the basis of same-sex bonding between women as affiliation. Within a context of affiliation, Bernard has hypothesized that specific relational and psychological benefits are derived. Lipman-Blumen's theory has centered on the attraction to and need for power as the basis of bonding. Lipman-Blumen has hypothesized that need fulfillment is derived through relationships where power is achieved. In Lipman-Blumen's thesis, the concept of power has been defined in a circumscribed manner and conforms to Winters' (1973) model. Winters has defined the power motive as the striving "to produce effects on the behavior or feelings of another person" (p. 10). The ability to produce these effects, Lipman-Blumen has argued, is through occupying positions of status. Since women are generally excluded from positions of status, she does not recognize women as having power, and, thus, relationships with women are not desirable for either sex. Moreover, she has contended that women only derive a sense of their own value through relationships with powerful others. Hence, based on this concept of power, Lipman-Blumen has undervalued same-sex bonds between women. In contrast to Winter's male-oriented concept of power, McClelland (1975), has argued that the goal of power motivation is to feel powerful and that influencing others is only one means for achieving this. McClelland's concept of power is more congruent with those of feminist psychologists (Lips, 1981). Feminist psychologists concerned with the issue of power have recognized that power can be contextualized in other ways than having power over another person (Goodchilds, 1979; Johnson; 1979; Lips 1979; Miller, 1976). Same-Sex Bonds and Well-Being p. 59 Both Lipman-Blumen and Bernard have addressed the issue of power in relationships, but have differed with respect to whether power is the basis of the bond or whether it is the outcome of the bond. Moreover, they have differed in their definition of power. For Lipman-Blumen, power is defined through material acquisitions and positional status. It conforms to the "power over" concept espoused by Winter (1973). Bernard has suggested that, although women feel a greater sense of power in their personal relationships through influence, she would see power as a lesser facet in the establishment of the bond, but a significant facet in the outcome of the bond. Bernard has predicted that what women derive from their same-sex bonds is a greater sense of power over themselves by sharing a common reality, and by having their needs met and affirmed in a context of affiliation. Same-Sex Bonds and Well-Being p. 60 CHAPTER III: METHODOLOGY In this section, the sample, research procedures, instrumentation, design and analysis of the data are presented. A. Sample The sample consisted of 101 married men and 101 married women between 25 and 45 years of age. Respondents were enrolled in UBC summer session courses. B. Research Procedures 1. All professors who taught courses during the 1985 summer session period were contacted and asked if student volunteers could be recruited from their classes. 2. If permission was obtained, a brief announcement was made by the investigator requesting the participation of married students in a study about social relationships in adult life and the benefits derived from them. Students were told that their participation was voluntary and they were free to withdraw from the study at any time without academic penalty. Students were also told that participation in the study involved completing a series of questionnaires which required approximately one hour to fill out. Same-Sex Bonds and Well-Being p. 61 3. Interested students were given an envelope containing the questionnaires with instructions for their completion, and told that the completion of the questionnaires assumed their permission to participate had been given. Students were asked to return the envelopes with the completed questionnaires to class a week later where they were picked up by the investigator. All questionnaires were numerically coded and presented in a counter balanced order to ensure subject anonymity and to guard against response set bias. Subjects were given the name and phone number of the investigator in case they wanted to discuss the study further. 4. Completed questionnaires were scored and coded on arrival to ensure that equal numbers of married men and women were included in the sample. Questionnaire packages containing incomplete or incorrectly fill out data were not included in the analysis. C. Instrumentation 1. The Social Provisions Scale (Russell <5c Cutrona, 1984) Development of the Scale. The Social Provisions Scale (Russell Sc Cutrona, 1984) was designed to assess the usefulness of Weiss's (1974) typology of relational needs fulfilled by social relationships. The Social Provisions Scale contains 24 items, with four items keyed on each provision, two in a positive direction and two in a negative direction. Subjects are asked to respond to each question on a four-point scale ranging from strongly agree to strongly disagree. Same-Sex Bonds and Well-Being p. 62 In the present study, subjects were asked to complete the Social Provisions Scale twice, once for their relationships with women and once for their relationships with men. The wording for each form of the questionnaire is expressed appropriately. In order to prepare respondents for completing the Social Provisions Scale, subjects were provided with a structured table and asked to list their network members. Respondents were instructed to define their network members as those people who "are important in your life at this moment, whether you like them or not". In addition, information as to the age of each network member, their relationship to them, their frequency of contact and the number of years they have known each person was also requested. This information was requested twice, once for the respondents' same-sex relationships and once for their cross-sex relationships. Next, on a separate table, respondents were asked to indicate the number of relationships fulfilling each of the six relational provisions. This table was also completed twice, once for their same-sex relationships and once for their cross-sex relationships. Scoring. A score for each provision was derived separately such that a high score indicates the respondent is receiving that provision. Negatively keyed items were reversed before scoring. The maximum score which can be obtained for each provision is 16, the minimum score is *. A total of six separate scores (one for each provision) as well as a composite score with maximum values of 96 and minimum values of 2* were obtained for each subject. Reliability. Test-retest and internal consistency reliability coefficients were derived from two investigations. The first investigation conducted by Cutrona, Russell and Rose (198*) involved a prospective study of stress and Same-Sex Bonds and Well-Being p. 63 social support among a sample of 100 community elderly residents. The Social Provisions Scale was administered twice over a four to six month period. Internal consistency reliabilities for the individual social provisions were all above .70. Test-retest reliabilities were not as high and ranged from .36 to .66. However, Russell and Cutrona (1984) noted that because social support tends to fluctuate over time, high test-retest correlations would not be expected. Internal consistency estimates based on a second study of 300 teachers conducted by Russell, Altmaier and Van Velzen (1984) on the relationship between social support and teacher burnout, found reliability coefficients of .60. Validity. In order to test the factor structure of the instrument, a confirmatory factor analysis based on the data derived from the teacher study was used. An oblique six factor model was tested using the LISREL VI program, with each of the factors corresponding to the six social provisions. The factor loadings indicated a "fairly good fit to the data" (p. 6). The Goodness of Fit Index was .857. Additional evidence supporting the validity of the measure was obtained through the two samples discussed earlier. One set of analyses examined the relationship between scores on the Social Provisions Scale and other measures of personal relationships. Scores on the Social Provisions Scale were found to correlate with measures of social networks (number of relationships and frequency of contact) as well as measures of satisfaction in different types of social relationships among the elderly. Moreover, scores on the social support measure developed by House (1981) were also correlated with the Social Provisions Scale. The pattern of correlations was consistent with those Same-Sex Bonds and Well-Being p. 6* predicted by Weiss. For example, spousal support was correlated with attachment, help and guidance and opportunity for nurturance. Supervisor support was associated with reassurance of worth and reliable alliance. Finally, Russell and Cutrona (1984) examined the ability of the Social Provisions Scale to predict the physical and mental health among the two samples. Social provision scores were significantly related to scores on the UCLA Loneliness Scale, with attachment as the most predictive index of emotional loneliness among the elderly (-.6*7, p_ < .001). Social Provision Scores were also found to be predictive of loneliness (-.371, £ < .001); depression (-.193, £ < .01) and health status (.166, £ < .05) among the teacher sample. In summary, the Social Provision Scale appears to demonstrate adequate psychometric properties. Moreover, it appears to be a valid measure of the different relational provisions described by Robert Weiss and was therefore adopted for use in this investigation. 2. The Spheres of Control Scale (Paulhus & Christie, 1981) Development of the Scale. The Spheres of Control Scale (Paulhus & Christie, 1981) provides a theoretical extension of existing conceptions of the locus of control construct developed by Rotter (1966). The model developed by Paulhus and Christie involves "partitioning the individual's life space in terms of primary behavioral spheres. Thus, the individual's confrontation with the world is decomposed into three distinct theaters" (p. 1253). The model is based on an interactionist perspective and reasons that perceived control varies as a function Same-Sex Bonds and Well-Being p. 65 of the different expectancies of control the individual exerts within the three domains of interaction with the world. The three domains of control are conceptualized as personal efficacy, which is defined as control within the nonsocial world in situations of personal achievement; interpersonal control which the authors define as interaction with others in dyads and group situations, for example "defending his or her interests at meetings, attempting to develop social relationships, or maintaining harmony in the family" (p. 1254). The third facet of the model is defined as sociopolitical control, which is conceptualized as perceived control in situations where the individual's needs conflict with those of the political and social system. The conceptual model underlying the scale views these spheres of control as independent dispositions. The first two scales, personal efficacy and interpersonal control were adopted for use in this investigation. The third sphere, sociopolitical control was not germane to this study and was therefore omitted. The Spheres of Control subscales were developed and refined over a period of two years. Information concerning the reliability, validity and factor structure of the scales was obtained in a series of five psychometric studies. Each subscale consists of 10 items, and the number of positively and negatively keyed items are balanced for each scale. Subjects respond to the items on a 7-point Likert scale. Reliability The authors report that alpha reliabilities for the subscales range between .75 and .80. Test-retest correlations at four weeks are above .9 and at six months are above .70. Same-Sex Bonds and Well-Being p. 66 Validity. Confirmatory factor analysis (LISREL) was applied to the data in order to test the validity of the Spheres of Control multidimensional factor structure over Rotter's unidimensional model. As predicted, the confirmatory factor analysis established the validity of the three factor model, and found that the unidimensional model is overly restrictive. Support for the validity of the Spheres of Control Scale was obtained through three studies. The first study conducted by Paulhus, Molin and Schuchts (1979) investigated the control profiles of varsity football players, varsity tennis players and non-athletes. "The study provides further support for the construct validity of the Spheres of Control Scales in that the researchers were able to predict the control pattern unique to each population on the basis of general character descriptions made by raters familiar with such athletes." (p. 1258). The second study sought to establish the relationship between the various dimensions of the Spheres of Control Scale with specific behavioural outcomes. First, it was hypothesized that high social political scores were predictive of voting behaviour; that high personal efficacy scores would predict preference for a bargaining game and that high interpersonal control would predict assertiveness in a telephone interview study. Overall, the predicted patterns were confirmed thus supporting the concurrent and convergent validity of the scales. Paulhus and Christie (1983) conclude that "on the whole the research reported here reflects favourably on the reliability, validity and utility of the Spheres of Control Scales" (p. 1262). Same-Sex Bonds and Well-Being p. 67 3. The UCLA Loneliness Scale - Revised (Russell, Peplau & Cutrona, 1980) The UCLA Loneliness Scale was developed in order to create "a psychometrically adequate, easily administered and generally available scale that would serve as a stimulus for empirical research on loneliness. At present, it is the most widely used loneliness measure" (Russell, 1980, p. 90). Development of the Scale. Initial development of the scale began with 25 items borrowed from Sisenwein's (1964) loneliness measure. Extreme items from Sisenwein's measure were eliminated. The initial set of items was administered to two groups, a clinical sample, and a sample of undergraduates attending UCLA. All participants responded to the items in terms of how lonely they were in comparison to others. Subjects were also asked to rate the intensity of their affective experience on ratings of depression, emptiness, restlessness and boredom. Reliability. The final scale, consisting of 20 items all demonstrated an inter-item correlation above .50. The internal consistency of the measure is .96. Reliability was also demonstrated in subsequent investigations. Solano (1980) reported an alpha of .89. With respect to test-retest reliability, Jones (1978) reported a correlation of .73 over a two month period and Cutrona (1980) reported a test-retest reliability of .62 over a 7-month period. These findings suggest that there is both stability and variation in loneliness raising the issue as to whether loneliness is a state or a trait. Same-Sex Bonds and Well-Being p. 68 Validity. The validity of the measure was assessed in three ways. First , a correlation of .79 was found between total scores on the loneliness measure and a single item, self-report of loneliness. Second, a comparison of the clinical and student samples indicated a significant difference between these two populations with the clinical group scoring two standard deviations higher than the student group. Finally, loneliness correlated with measures of depression, anxiety, dissatisfaction, unhappiness and shyness. Other studies have reported correlations with other measures of loneliness. Solano (1980) reported a correlation of .7* between the Bradley and U C L A Scales. Abbreviated Loneliness Scale. Russell, Peplau and Cutrona (1980) also developed a *-item survey version of the U C L A Loneliness Scale, consisting of two positively keyed items and two negatively keyed items. Through the use of regression analysis, "the set of four items that best predicted scores on the self-labeling loneliness index were selected." (p. *73). The *-item loneliness scale has a coefficient alpha of .75. Ellison and Paloutzian (1979) reported a correlation of .79 between the Abbreviated Loneliness Scale and the U C L A Scales. "We recommend that investigators who want a shortened version of the loneliness scale use these four items" (p. 473). Therefore, the abbreviated scale was adopted for use in this investigation. 4. The Satisfaction With Life Scale (Diener, 198*) Development of the Scale. In recent years, three separate components of the well-being construct have been identified, positive affect, negative affect Same-Sex Bonds and Well-Being p. 69 and general life satisfaction (Andrews <5c Withey, 1976). Researchers have focused on assessing the positive and negative affective components of well-being (Bradburn, 1969) but have generally failed to measure the third component, general life satisfaction. Scales of general life satisfaction usually consist of a single item. The Satisfaction With Life Scale assesses general life satisfaction which is conceptualized as "a global assessment of a person's quality of life according to his chosen criteria" (p. 3). The 5-item scale is predicated on the assumption that "the hallmark of . .. subjective well-being . . . centers on a person's own judgements, and not upon some criterion which is set by the researcher" (p. 3). The scale was initially constructed from a list of 48 self-report items which consisted of questions about satisfaction with one's life. A factor analysis was conducted, yielding three factors, positive affect, negative affect and satisfaction. The affect items and satisfaction items with loadings of less than .60 were eliminated. The final scale consists of five items. Scoring. Subjects are asked to respond to each of the five items on a 7-point Likert Scale ranging from strongly disagree to strongly agree, yielding a maximum score of 35 and a minimum score of five. Reliability. In order to test the reliability of the scale, 176 undergraduate students were given the Satisfaction With Life Scale and 76 students were retested two months later. The inter-item correlations ranged from .44 to .71 with a mean value of .57. The two month test-retest correlation coefficient was .82. Same-Sex Bonds and Well-Being p. 70 Validity. The validity of the measure was assessed through two additional studies. The first examined the relationship between the Satisfaction With Life Scale and other measures of subjective well-being. The results indicated moderately strong correlations with all measures of subjective well-being (.58-.75). The second study examined the correlation between the scale to measures of self-esteem (Rosenberg, 1965); temperament (Buss & Plomin, 1975); neuroticism (Eysenck, 196*); a symptom checklist (Derogatis, 197*) and the Crowne-Marlowe Scale of Social Desirability (196*). The correlations for the second validity study were self-esteem .5*; symptom checklist -.*1; neuroticism -.*8; emotionality, -.25; activity, .08; sociability, .20; impulsivity -.03 and .02 with the Crowne-Marlowe measure, indicating that the scale is not evoking a socially desirable response set. Overall, "it appears that individuals who are satisfied with their lives are in general well-adjusted and free from psychopathology" (p. 7). The author concludes that the scale has favourable psychometric properties. He maintains that the conceptual strength of the scale lies in the ability of the respondents to weigh the various domains of their life in whatever way chosen. 5. The Rosenberg Self-Esteem Scale (Rosenberg, 1965) Development of the Scale. The Rosenberg Self-Esteem Scale (Rosenberg, 1965) is a 10-item Guttman Scale designed to measure the "totality of the individual's thoughts and feelings having reference to himself as an object" (Rosenberg, 1969, p. 7) Subjects are instructed to respond to the items on a *-point scale ranging from strongly agree to strongly disagree. Same-Sex Bonds and Well-Being p. 71 Scoring. The items are grouped into six categories and each category is scored on the basis of its component item or items, yielding a maximum score of six for each category. Category one is contrived from the combined responses to items 3, 7 and 9; category two from items 4 and 5; category three from item 1; category four from item 8; category five from item 10; and category six from items 2 and 6. Rosenberg (1969) notes that the 10 items can be added together to derive a total score. This method of scoring was used in the current investigation. Reliability. Internal reliability coefficients of .92 were reported by Rosenberg (1969). Test-retest reliability coefficients over a two week period of .85 were reported by Silber and Tippett (1965). Validity. The construct validity of the scale was assessed through its relationship to depressive affect, anxiety and peer group reputation (Rosenberg, 1969). Rosenberg reports that in his New York State study there was a clear relationship between the self-esteem scale and a 6-item scale measuring depressive effect. "Only 4 percent of those with the highest self-esteem scores compared with 80 percent of those with the lowest scores were rated as 'highly depressed' r = .30" (p. 292). Moreover on measures of anxiety, 69 percent of those reporting low self-esteem also reported high on anxiety. In a sample of 272 high school seniors, those students who reported high on the self-esteem scale also obtained high sociometric ratings from their peers on the basis of their perceived leadership abilities. Same-Sex Bonds and Well-Being p. 72 Silber and Tippett (1965) examined the convergent validity of the scales. Using a sample of college students, subjects were assessed on two traits, global self-esteem and stability of self-concept by means of four different methods. These were: the Rosenberg Scale, the Kelley Repertory Test, the Health Self-Image Questionnaire and a psychiatric rating. The Rosenberg Self-Esteem Scale correlated .67 with the Kelley Repertory Test; .83 with the Health Self-image Questionnaire and .56 with the psychiatric rating. Other evidence of convergent validity was reported by Crandall (1973). A correlation of .60 was found between the Rosenberg Scale and the Coopersmith Self-Esteem Inventory (Coopersmith, 1967). 6. The Hakstian-McClean Depression Scale (Hakstian Sc McClean, 1979) Development of the Scale. The Hakstian-McClean Depression Scale (Hakstian Sc McClean, 1979) is a *-item scale designed to assess clinical depression. The scale was developed during a differential treatment outcome study of a moderately depressed and non-depressed population. The scale is comprised of the four most discriminating items from the Client Assessment Questionnaire which were found to be the best predictors of group membership. These predictors were then subjected to discriminant analysis using a double cross-validation technique to determine hit rates (Peter McClean, personal communication). Scoring. The first three items are scored on a 10-point scale. The fourth item is scored on a 5-point scale. The subjects' responses to the first three items Same-Sex Bonds and Well-Being p. 73 are added together. The subjects' response on the last item is multiplied by four and then added to the sum of the first three. Scores range from 7 to 50 (Peter McClean, personal communication). Reliability. The test-retest reliability of the scale was assessed over a three month interval among a group of depressed (n = 196) and non-depressed (n = 161) people. The test-retest reliability for the non-depressed group was .54 and the alpha coefficient was .63. The alpha coefficient for the depressed group was .65. A test-retest reliability coefficient was also assessed for a group of students (n = 49), over a one week period. The test-retest reliability coefficient for this group was .73. D. Design and Analysis for the Four Research Questions The analysis of the data was divided into three sections. The first research question involved testing the overall strength of men's and women's same-sex and cross-sex bonds. A 2 X 2 univariate analysis of variance (ANOVA) was used with sex of respondent and relationship gender composition (same-sex, cross-sex) as the independent variables, with a repeated measure on the last factor. The dependent measure was the total score obtained on the Social Provisions Scale. A simple main effects analysis was then performed for the significant interactions. The second research question involved testing differences in the individual relational provisions supplied through same-sex and cross-sex bonds. A 2 x 2 multivariate analysis of variance (MANOVA) was used with sex of respondent and relationship gender composition (same-sex, cross-sex) Same-Sex Bonds and Well-Being p. 7* as the independent variables. The dependent measures were Attachment, Social Integration, Reassurance of Worth, Reliable Alliance, Help and Guidance and the Opportunity for Nurturance. The overall MANOVA was followed up with a series of 2 x 2 univariate analyses of variance (ANOVA) on the six dependent measures. These analyses were ultimately followed up with a simple main effects analysis for the significant interaction effects. An experiment-wise error rate of 5% was maintained by the apportionment of alpha through the Bonferroni procedure. The third and fourth research questions involved testing the effects of 'sex of respondent' and 'relationship gender composition' in the association between the overall strength of bonds (total score from the Social Provisions Scale) with global well-being. In preparation for the correlational analyses a principal components analysis (PCA) was performed on the six well-being variables: Self-Esteem, Personal Efficacy, Interpersonal Control, Depression, Loneliness and Life Satisfaction. One principal component was extracted and identified as "global well-being". Tests of independent correlations were used to answer the third research question and tests of dependent correlations were used to answer the fourth research question. Because of the number of analyses, alpha was set at .005 to control for the increased probability of Type 1 errors. Same-Sex Bonds and Well-Being p. 75 E. Design and Analysis of the Network Data In order to explore sex differences in network structure as well as sex differences in those aspects of network structure contributing to global well-being, separate multivariate, univariate and regression analyses were performed. First, a 2 X 2 multivariate analysis of variance (MANOVA) with one between subjects factor (sex of respondent) and one within subjects factor (relationship gender composition) with a repeated measure on the second factor was performed on the four dependent measures: Number of Kin and Non-Kin Network Members, and Frequency of Contact with Kin and Non-Kin Network Members. The omnibus MANOVA was followed up with a series of univariate analyses of variance (ANOVA). Alpha was again proportioned to counteract the risk of Type 1 errors. Next, in order to explore sex differences in those aspects of network structure that contribute to global well-being, stepwise regression analyses were performed. Global well-being was regressed onto the four same-sex and cross-sex network variables separately, by sex. In order to explore sex differences in the number of network members supplying each of the six relational provisions, independent of their strength, a second 2 X 2 multivariate analysis of variance (MANOVA) with one between subjects factor (sex of respondent) and one within subjects factor (relationship gender composition) was performed on the six dependent variables: Attachment, Same-Sex Bonds and Well-Being p. 76 Social Integration, Reassurance of Worth, Reliable Alliance and Help and Guidance. The omnibus MANOVA was followed up with a series of univariate analyses of variance (ANOVA). A simple main effects analysis was performed on the significant interactions. An experiment-wise error rate of .05 was maintained by the apportionment of alpha through the Bonferroni procedure. In order to explore whether the number of network members supplying each of the relational provisions contributes to men's and women's global well-being, stepwise regression analyses were performed. Global well-being was regressed onto the number of network members supplying Attachment, Social Integration, Reassurance of Worth, Reliable Alliance, Help and Guidance and Opportunity for Nurturance, separately, by sex. Same-Sex Bonds and Well-Being p. Table 1 Tabular Presentation of the Variables, Design, Analysis and Instrumentation Variables Design & Analysis Measures Independent Dependent Question One sex of respondent gender composition total provision score 2X2 ANOVA and simple main effects analysis Social Provisions Scale Question Two sex of respondent gender composition attachment social integration reassurance of worth reliable alliance help and guidance opportunity for nurturance 2X2 MANOVA followed-up with 10 ANOVAS and simple main effects analysis Social Provisions Scale Questions Three and Four total provision score global well-being correlation and tests of dependent and independent correlations Social Provisions Scale Rosenberg Self-Esteem Scale Spheres of Control Scale Life Satisfaction Scale UCLA Loneliness Scale-R Hakstian-McClean Depression Scale Same-Sex Bonds and Well-Being p. 78 CHAPTER IV: RESULTS The presentation of the research findings is divided into eight sections. In the first section, a description of the sample is presented. The next three sections are organized around the research hypotheses outlined in Chapter One. Each section begins with a description of the method of analysis, followed by a restatement of the opposing research hypotheses, corresponding findings and, summary of the results. The last four sections contain a discussion of differences in the network characteristics of men and women and their contribution to well-being. A. Characteristics of the Sample During the summer of 1985, all professors teaching University of British Columbia summer school courses were approached by the researcher and asked if student volunteers could be recruited from their classes. If permission was obtained, the researcher attended the class and made a brief announcement about the nature of the study, and requested student participation. To ensure a strong response rate, the researcher returned to the class within the following week (summer school classes are held daily). Approximately three hundred students agreed to participate in the study and 101 women and 101 men completed the questionnaires, yielding a response rate of 66%. All respondents were married students attending the University of British Columbia summer session. Respondents ranged between 25 and *5 years of age and reported an average of one child. Seventy-five per cent of the men and 67% of the women Same-Sex Bonds and Well-Being p. 79 were employed full-time. Seventy-one percent of the men were employed in a managerial/professional capacity as teachers and school administrators, while 59% of the women were employed in managerial/professional positions as teachers, nurses and school administrators. Approximately one-quarter of the sample were unemployed. Seventy-three percent of the men were graduate students enrolled in masters or doctoral programmes in the Faculty of Education and 27% were undergraduates. Of the women, 48% were enrolled in masters or doctoral programmes and the remaining 52% were undergraduates. Information regarding the socioeconomic status of the respondents was not collected. With respect to the ethnic composition of the sample, approximately 50% of the respondents identified themselves as Canadian, 4% as Chinese, 25% as European, 1% as East Indian, 1% as Native Indian and 1% as West Indian. The remaining 18% of the respondents did not identify their ethnic origin. B. Hypotheses relating to the Effects of Sex of Respondent and  Relationship Gender Composition  on the Overall Level of Social Support 1. ANOVA In order to answer Hypotheses 1-5, a 2 x 2 between-within univariate analysis of variance (ANOVA) with one between subjects factor (sex of respondent) and one within subjects factor (gender composition) was performed on the total scores obtained from the six subscales of the Social Provisions Scale. Same-Sex Bonds and Well-Being p. 80 A significant main effect was obtained for relationship gender composition, F(l,200) = 28.85, £ < .001, but not for sex of respondent, F(l,200) = 2.41, £ > .05. A significant interaction effect was also obtained F(l,200) = 9.56, £ < .005. 2. Hypotheses Hypothesis One The investigator predicted that the total mean provision scores would be higher for the same-sex relationships of men and women than the total mean provision scores for the cross-sex relationships of men and women. Although the results of the ANOVA indicated a significant main effect for gender composition, F(l,200) = 28.85, £ < .001, a significant interaction effect was also obtained, F(l,200) = 9.56, £ < .00.5. When the significant interaction was analyzed through a simple main effects analysis procedure, the results indicated that the effect was attributable to the results obtained for the women. Women's total mean provision score was significantly higher for their same-sex relationships (M = 85.09, s.d. = 8.88) than for their cross-sex relationships (M_ -78.8*, s^ cL = 10.97), F(l,200) = 35.80, £ < .001. However, there was no significant difference in the total mean provision score for the same-sex relationships of men (M = 80.9*, s.d. = 9.*8) and the cross-sex relationships of men (M = 79.25, s.d. = 10.57), F(l,200) = 2.60, £ > .05. Thus, the significant main effect for gender composition was due to the results for the women only and the investigator's prediction was partially supported. Same-Sex Bonds and Well-Being p. 81 Figure 1 Graphic Presentation of the ANOVA: Main Effects and Interactions men women same-sex cross-sex Same-Sex Bonds and Well-Being p. 82 Hypothesis Two It was predicted from Safilios-Rothschild's theory that the total mean provision score for the same-sex relationships of women would be higher than the total mean provision score for the same-sex relationships of men but that the total mean provision score for the cross-sex relationships of men would be higher than the total mean provision score for the cross-sex relationships of women. Hypothesis Three In opposition to the hypothesis derived from Safilios-Rothschild, it was predicted from Lipman-Blumen's theory that the total mean provision score for the same-sex relationships of men would be higher than the total mean provision score for the same-sex relationships of women, but the total mean provision score for the cross-sex relationships of women would be higher than the total mean provision score for the cross-sex relationships of men. The total mean provision score was significantly higher for the same-sex relationships of women (M = 85.09, s.d. = 8.88) than the same-sex relationships of men (M = 80.9*, s^ d. = 9.*8), F(l,332) = 8.66, £ < .005. However, there was no significant difference in the total mean provision score for the cross-sex relationships of women (M = 78.8*, s.d. = 10.97) and the total mean provision score for the cross-sex relationships of men (M = 79.25, s.d. = 10.57), F(l,332) = .086, £ > .05. Predictions derived from Safilios-Rothschild's theory were therefore partially supported. Predictions derived from Lipman-Blumen's theory were not supported. Same-Sex Bonds and Well-Being p. 83 Hypothesis Four It was predicted from Lipman-Blumen's theory that for men, the total mean provision score would be higher for their same-sex relationships than for their cross-sex relationships. Hypothesis Five In opposition to the hypothesis predicted from Lipman-Blumen, it was predicted from Safilios-Rothschild's theory that for men, the total mean provision score would be higher for their cross-sex relationships than for their same-sex relationships. A non-significant difference was obtained between the total provision score for the same-sex relationships of men (M = 80.94, s.d. = 9.48) and the cross-sex relationships of men (M 79.25, s^ d. = 10.57), F(l,200) = 2.59, £ > .05. Predictions derived from the theories of Lipman-Blumen and Safilios-Rothschild were not supported. 3. Summary of Hypotheses One through Five In this section, a series of predictions derived from the theoretical positions of the investigator, Lipman-Blumen, and Safilios-Rothschild on the overall level of social support provided by men's and women's same-sex and cross-sex bonds were examined through a univariate analysis of variance Same-Sex Bonds and Well-Being p. 8* procedure and followed up with a simple main effects analysis for the significant interactions. Overall, women's same-sex relationships provided signficantly higher levels of social support than their cross-sex relationships, supporting Safilios-Rothschild's theory. Moreover, women's same-sex relationships provided higher levels of social support than men's same-sex relationships, supporting Safilios-Rothschild's theory and failing to support Lipman-Blumen's theory. Further, there were no significant differences between the overall level of social support provided by men's same-sex and cross-sex bonds, failing to support Lipman-Blumen's and Safilios-Rothschild's theories. Finally, there were no significant differences between the overall level of social support provided by men's and women's cross-sex bonds, again failing to support Lipman-Blumen's theory. The implications of these -results are discussed in Chapter Five. C. Hypotheses relating to the Effects of Sex of Respondent  and Relationship Gender Composition on the level of Individual Relational Provisions 1. MANOVA In order to answer Hypotheses 6-11, a 2 x 2 between-within multivariate analysis of variance (MANOVA) with one between subjects factor (sex of respondent) and one within subjects factor (gender composition) was performed on the individual provision scores obtained from the six subscales of the Social Provisions Scale. The six dependent measures were Attachment, Social Integration, Reassurance of Worth, Reliable Alliance, Help and Guidance, and Same-Sex Bonds and Well-Being p. 85 the Opportunity for Nurturance. By the likelihood ratio test, a significant main effect was obtained for sex of respondent, F(l,200) = 5.60, £ < .001; for relationship gender composition, F(l,200) = 12.30, £ < .001 and for their interaction effect F(l,200) = 11.02, £ < .001. To investigate differences in the individual provisions, ten univariate analyses of variance (ANOVA) were computed. Significant interaction effects were analyzed by a simple main effects analysis (Winer, 1971, p. 529-532). An experiment-wise error rate of 5% was maintained by the apportionment of alpha through the Bonferroni procedure. Because ten analyses were performed, alpha was set at .005. 2. Hypotheses Hypothesis Six It was predicted from Bernard's theory that for women, the mean scores on the Attachment, Social Integration, Reassurance of Worth, Reliable Alliance, Help and Guidance and Opportunity for Nurturance provisions would be higher for their same-sex relationships than for their cross-sex relationships. Hypothesis Seven In opposition to the hypothesis predicted from Bernard, it was predicted from Lipman-Blumen's theory that for women, mean scores on the Reassurance Same-Sex Bonds and Well-Being p. 86 of Worth and Help and Guidance provisions would be higher for their cross-sex relationships than for their same-sex relationships. Hypothesis Eight In opposition to the hypotheses predicted from Bernard and Lipman-Blumen, it was predicted from Weiss's theory that for the women, there would be no significant differences between their same-sex and cross-sex provision scores. The following provision scores were significantly higher for the same-sex relationships of women than the cross-sex relationships of women: Attachment (F(l,200) = 23.23, £ < .001), Social Integration (F(l,200) = 16.6*, £ < .001), Reassurance of Worth (F(l,200) = 17.50, £ < .001), Help and Guidance (F(l,200) = 20.28, £ < .001), Opportunity for Nurturance (F(l,200) = 12.7*, £ < .001). However, there was no significant difference in the scores on the Reliable Alliance provision for the same-sex and cross-sex relationships of women (F(l,200) = 7.30, £ > .005). These results are illustrated in Table 2. A majority of the predictions derived from Bernard's theory was supported; only one prediction from Weiss's theory was supported; and, predictions derived from Lipman-Blumen' theory were not supported. Same-Sex Bonds and Well-Being p. 87 Table 2 Mean Provision Scores for Women's Same-Sex and Cross-Sex Bonds Provision Same-Sex Cross-Sex Attachment M S.D. 14.30 2.00 12.80 2.70 23.23 < .001 Social Integration M S.D. 14.00 2.00 12.90 2.40 16.64 < .001 Reassurance of Worth M S.D. 14.20 1.70 13.00 1.90 17.50 < .001 Reliable Alliance M S.D. 15.00 1.30 14.60 2.00 7.30 > .005 Help and Guidance M S.D. 14.70 1.80 13.60 2.60 20.28 < .001 Opportunity for Nurturance M S.D. 12.70 2.60 11.00 2.80 12.74 < .001 Hypothesis Nine Same-Sex Bonds and Well-Being p. 88 It was predicted from Lipman-Blumen's theory that for men, mean scores on the Social Integration, Reassurance of Worth, Reliable Alliance and Help and Guidance provisions would be higher for their same-sex relationships than for their cross-sex relationships. Hypothesis Ten In opposition to the hypothesis predicted from Lipman-Blumen, it was predicted from Bell's and Safilios-Rothschild's theories that for men, mean scores on the Reassurance of Worth and Help and Guidance provisions would be higher for their cross-sex relationships than for their same-sex relationships. Hypothesis Eleven In opposition to the hypotheses predicted from Lipman-Blumen, Bell and Safilios-Rothschild, it was predicted from Weiss's theory that for the men, there would be no significant differences between their same-sex and cross-sex scores on the Social Integration, Reassurance of Worth, Help and Guidance and Reliable Alliance provisions. The following provision scores were significantly higher for the same-sex relationships of men than the cross-sex relationships of men: Social Integration (F(l,200) = 47.40, p. < .001) and Reliable Alliance (F(l,200) = 12.32, £ < .001). Same-Sex Bonds and Well-Being p. 89 However, there were no significant differences between the scores on the Reassurance of Worth (F(l,200) = 2.50, £ > .005) and Help and Guidance provisions (F(l,200) = 2.65, £ > .005). These results are illustrated in Table 3. Predictions derived from Lipman-Blumen's and Weiss's theories were partially supported. Predictions derived from the theories of Bell and Safilios-Rothschild were not supported. 3. Summary of Hypotheses Six through Eleven In this section, predictions derived from the theoretical positions of Bernard, Lipman-Blumen, Bell and Safilios-Rothschild and Weiss on the relational provisions obtained through men's and women's same-sex and cross-sex bonds were tested through a multivariate analysis of variance procedure (MANOVA) and followed up with a series of univariate analysis of variance (ANOVA) procedures, and ultimately, with a simple main effects analysis for the significant interaction effects. Women's same-sex relationships were found to provide higher levels of the Attachment, Social Integration, Reassurance of Worth, Help and Guidance and Opportunity for Nurturance provisions than their cross-sex relationships. These findings lend further, but not complete support to Bernard's theoretical formulations and generally fail to support Weiss's theory. In addition, men's same-sex relationships were found to provide higher levels of the Social Integration and Reliable Alliance provisions than were their cross-sex relationships. However, non-significant differences were obtained for the remaining tested provisions. These results indicated partial support for Lipman-Same-Sex Bonds and Well-Being p. 90 Table 3 Mean Provision Scores for Men's Same-Sex and Cross-Sex Bonds Provision Same-Sex Cross-Sex Social Integration M 14.40 12.60 47.40 < .001 S.D. 1.70 2.40 Reliable Alliance M 14.90 14.20 12.32 < .001 S.D. 1.70 2.30 Reassurance of Worth M 14.30 14.10 2.50 > .005 S.D. 1.70 1.80 Help and Guidance M 14.10 13.70 2.65 > .005 S.D. 2.10 2.50 Same-Sex Bonds and Well-Being p. 91 Blumen's and Weiss's theories and did not support the theories of Bell and Safilios-Rothschild. The implications of these results are discussed in Chapter Five. D. Hypotheses relating to the Effects of Sex of Respondent  and Relationship Gender Composition on the  Association between the Overall Level of Social Support  and Global Well-Being 1. Correlational Analyses Hypotheses 12-18 were correlational in nature and were analyzed by means of correlational procedures. Since directional hypotheses were predicted, one-tailed tests of significance were used. Hypotheses 12 and 13 involved testing sex differences in independent correlations. Because hypotheses 16-18 involved testing differences in relationship gender composition, tests of dependent correlations were used. Because of the number of analyses, alpha was set at .005 to control for the increased probability of Type I errors. In preparation for the correlational analyses, a principal components analysis (PCA) was performed on the six outcome variables: Self-Esteem, Personal Efficacy, Interpersonal Control, Depression, Loneliness, and Life Satisfaction. This preliminary step was taken for both conceptual and psychometric reasons. Conceptually, these variables have been identified in the literature as subcomponents of well-being (Andrews «5c Withey, 1976; Diener, Same-Sex Bonds and Well-Being p. 92 198*; Weiss, 1969). However, a comprehensive and psychometricaily sound instrument for measuring well-being has not been developed. A principal components analysis was undertaken to evaluate the coherence of these variables for a conceptual formulation of well-being. These variables were found to be significantly intercorrelated, providing initial support for their common structure (Table *). Psychometricaily, the principal components analysis was performed to create an optimal linear composite of the measures. The homogeneity of the variance-covariance matrices for men and women was tested using the Bartlett-Box procedure (Harris, 1975). The raw scores were first mean-deviated by sex to give a common zero mean for all the variables. Non-significant differences between the variance co-variance matrices were obtained (F = .79, £ > .10) enabling the combined analysis of the sexes (Table 5). One principal component was extracted which accounted for *7.296 of the variance, thus suggesting a common structure among these variables. The principal component was identified as "Global Well-Being." 2. Hypotheses Hypothesis Twelve It was predicted from Weiss's theory that higher total mean provision scores (combining total scores on same-sex provisions and total scores on cross-sex provisions) would be equally associated with Global Well-Being for men and women. Same-Sex Bonds and Well-Being p. 93 Table 4 Separate Intercorrelation Matrices of Variables forming the First Principal  Component: Men and Women Dep. Self- Life Per. Int. Lone. Est. Sat. Eff. Con. Dep. — -.427 -.347 -.177 -.266 .216 Self-Est. -.523 — .504 .310 .311 -.401 Life Sat. -.509 .558 — .268 .153 -.385 Per. Eff. -.172 .343 .203 — .224 -.232 Int. Con. -.454 .471 .291 .238 -.422 Lone. .539 -.491 -.518 -.124 .510 — Note: Correlations for the men are above the diagonal. Correlations for the women are below the diagonal. Same-Sex Bonds and Weli-Being p. 9* Table 5 Pooled Intercorrelation Matrix of Variables forming the First Principal Component Dep. Self-Est. Life Sat. Per. Eff. Int. Con. Lone. First i Com Dep. -.*80 - . W O -.172 -.381 A07 - . 7 H Self-Est. ~ .532 .325 .*01 -A52 .792 Life Sat. ~ .235 .232 -.458 .730 Per. Eff. — .229 -.175 .450 Int. Con. — -.47* .650 Lone. -.7*0 Note; Due to the non-significant differences between the variance co-variance matrices, the intercorrelation matrices for the men and women were pooled. Same-Sex Bonds and Well-Being p. 95 Hypothesis Thirteen In opposition to the hypothesis derived from Weiss's theory, it was predicted from Bernard's and Miller's theories that the magnitude of the correlation between total mean provision scores (combining total scores on same-sex provisions and total scores on cross-sex provisions) would be more positively associated with outcome measures of Global Well-Being for women than for men. For men, total mean provision scores demonstrated a significant, positive correlation with Global Well-Being (r = .50, £ < .001). For women, total mean provision scores demonstrated a significant, positive correlation with Global Well-Being (r = .52, £ < .001). However, these correlations were not significantly different (z = .190, £ > .05). The prediction derived from Weiss's theory was supported. However, the prediction derived from the theories of Bernard and Miller was not supported. Hypothesis Fourteen It was predicted from Bernard's theory that for women, the magnitude of the correlation between their total same-sex provision score would be more positively associated with measures of Global Well-Being than would be their total cross-sex provision scores. Same-Sex Bonds and Well-Being p. 96 Hypothesis Fifteen In opposition to the hypothesis derived from Bernard, it was predicted from Lipman-Blumen's theory that for women, the magnitude of the correlation between their total cross-sex bond provision score and measures of Global Well-Being would be significantly stronger than the magnitude of the correlation between their total same-sex bond provision score and measures of Global Well-Being. The women's total mean same-sex provision scores demonstrated a significant, positive correlation with Global Well-Being (r = .46, p_ < .001). Moreover, women's total mean cross-sex provision scores also demonstrated a significant positive correlation with Global Well-Being (£ = .44, £ < .001). However, these rs were not significantly different (t(200) = .27, £ > .05). Therefore, the prediction derived from Bernard's and Lipman-Blumen's theories was not supported. However, the prediction derived from Weiss's theory was supported. Hypothesis Sixteen It was predicted from Lipman-Blumen's theory that for men, the magnitude of the correlation between their total same-sex bond provision score and measures of Global Well-Being would be significantly stronger than the magnitude of the correlation between their total cross-sex bond provision score and measures of Global Well-Being. Same-Sex Bonds and Well-Being p. 97 Hypothesis Seventeen In opposition to the hypothesis derived from Lipman-Blumen, it was predicted from Bell's and Safilios-Rothschild's theories that for men, the magnitude of the correlation between their total cross-sex bond provision score and measures of Global Well-Being would be significantly stronger than the magnitude of the correlation between their total same-sex bond provision score and measures of Global Well-Being. Hypothesis Eighteen In opposition to the hypotheses derived from Bernard, Lipman-Blumen, Bell and Safilios-Rothschild, it was predicted from Weiss's theory that for both the men and the women, the magnitude of the correlation between their total same-sex bond provision score and Global Well-Being would not be significantly different from the magnitude of the correlation between their total cross-sex bond provision score and Global Well-Being. The men's total mean same-sex provision scores demonstrated a significant, positive correlation with Global Well-Being (r = .43, p_ <.001).. Moreover, men's total mean cross-sex provision scores also demonstrated a significant, positive correlation with Global Well-Being (r = .41, £ <.001). However, these rs were not significantly different (t(200) = .22, £ >.05). Therefore, the predictions derived from the theories of Lipman-Blumen, Bell and Safilios-Rothschild were not supported. However, the prediction derived from Weiss's theory was supported. Same-Sex Bonds and Well-Being p. 98 In order to investigate the association between relationship gender composition to men's and women's Global Well-Being more fully, additional correlational and regression analyses were performed. For the correlational analyses, the molar variable, Global Well-Being was separated into its constituent variables. These variables were: Interpersonal Control, Personal Efficacy, Self-Esteem, Life Satisfaction, Depression and, Loneliness. The scores on these individual variables were then correlated with the men's and the women's total same-sex provision scores as well as with the men's and the women's total cross-sex provision scores. With the exception of Personal Efficacy, all correlations between men's and women's total same-sex provision scores and all correlations between men's and women's total cross-sex provision scores were significantly different from zero. However, tests of dependent correlations again revealed no significant differences between the correlations involving the total same-sex provision scores and those involving the total cross-sex provision scores with measures of Interpersonal Control, Personal Efficacy, Self-Esteem, Life Satisfaction, Depression and, Loneliness. These results are illustrated in Tables 12 and 13 in Appendix A. Thus, there is a non-significant difference between same-sex and cross-sex bonds with well-being when both global and individual measures of well-being are used. Regression analyses were also performed. First, in a step-wise regression analysis procedure, Global Well-Being was regressed onto the six individual same-sex and the six individual cross-sex provision scores: Attachment, Social Integration, Reassurance of Worth, Reliable Alliance, Help and Guidance, and, the Opportunity for Nurturance, separately, by sex. Same-Sex Bonds and Well-Being p. 99 For the men, three provisions entered the stepwise equation, Cross-Sex Reassurance of Worth, Cross-Sex Opportunity for Nurturance and Same-Sex Social Integration, respectively. Cross-Sex Reassurance of Worth entered as the first step (R 2 = .30; F(l,99) = 42.32, £ < .001) followed by Cross-Sex Opportunity for Nurturance (R 2 = .35; F(2,98) = 26.28, £ < .001) and Same-Sex Social Integration (R 2 = .39; F(3,97) = 20.70, £ < .001). For the women, two provisions entered the equation. At the first step Same-Sex Reassurance of Worth was entered (R 2 = .20; F(l,99) = 24.84, £ < .001) and at the second Cross-Sex Help and Guidance was entered (R 2 = .28; F(2,98) = 18.99, £ < .001). These results are illustrated in Tables 14 and 15 in Appendix A. Although these results suggest that different dimensions of same-sex and cross-sex bonds contribute to men's and women's Global Well-Being, the substantial intercorrelations between the relational provisions (Table 16 and 17 in Appendix A) prevents a clear and accurate interpretation of the data. Because the results of the step-wise analysis seem to suggest that men obtain more of their global well-being through their cross-sex relationships, a full regression analysis was conducted in order to clarify the relationship between relationship gender composition and Global Well-Being. In the full regression analysis, Global Well-Being was regressed onto the six individual same-sex relational provisions, separately, by sex, yielding four regression analyses in all. Thus, four comparisons were made: 1) men's same-sex multiple R_ was compared with men's cross-sex multiple Rj 2) women's same-sex multiple R_ was compared with women's cross-sex multiple Rj 3) men's same-sex multiple R was compared with women's same-sex multiple R; and 4) men's cross-sex multiple R was Same-Sex Bonds and Well-Being p. 100 compared with women's cross-sex multiple R. The first two multiple R/s were compared through a procedure developed by Steiger and Browne (1984). The results of the four pairwise R comparisons again indicate that there are no significant differences in the amount of Global Well-Being men and women obtain through their same-sex and cross-sex bonds. These results are illustrated in Tables 18 and 19 in Appendix A. Collectively, the results of these analyses indicate support for Weiss's theory and fail to support theoretical propositions that predict differences in the relationship between the gender composition of social bonds with Global Well-Being. 3. Summary of Hypotheses Twelve through Eighteen In this section, a series of predictions derived from the theories of Bernard, Lipman-Blumen, Bell, Safilios-Rothschild and Weiss examined the relationship between sex of respondent and relationship gender composition with measures of Global Well-Being. The hypotheses were tested through correlational analyses, including tests on independent and dependent correlations. Overall, there were no significant differences in the correlations between men's and women's social bonds with Global Well-Being. Moreover, there were no significant differences in the relationships between same-sex or cross-sex bonds with measures of Global Well-Being, supporting Weiss's theory and failing to support all of the other theoretical propositions. The implications of these results are discussed in Chapter Five. Same-Sex Bonds and Well-Being p. 101 E. Sex Differences in Network Characteristics In order to explore sex differences in the structural characteristics of men's and women's networks, a 2 X 2 multivariate analysis of variance (MANOVA) with one between subjects factor (sex of respondent) and one within subjects factor (relationship gender composition) was performed on the scores obtained from the four dependent variables: Number of Kin and Non-Kin Network Members, and, Frequency of Contact with Kin and Non-Kin Network Members. The overall MANOVA was followed up with a series of univariate ANOVAS. By the likelihood ratio test a significant main effect was obtained for sex of respondent, F(l,200) = 6.22, £ < .001 and relationship gender composition, F(l,200) = 29.02, £ < .001, but not for their interaction, F(l,200) = 2.50, £ > .05. Means and standard deviations for the two cells in the design and the four dependent variables appear in Table 6. The significant main effect for sex was attributable to two variables. The women reported larger Kin composed networks, F(l,200) = 12.88, £ < .001 and greater Frequency of Contact with Kin F(l,200) = 9.94, £ < .005 than the men. Non-significant differences between the sexes were obtained for the non-kin variables. The significant main effect for relationship gender composition was attributable to the greater number of Same-Sex Non-Kin network members reported by the respondents F(l,200) = 98.79, £ < .001 as well as the greater Frequency of Contact with Same-Sex Non-Kin network members F(l,200) = 101.11, £ < .001 reported by the respondents. Same-Sex Bonds and Well-Being p. 102 F. Network Characteristics Predicting Global Well-Being In order to explore sex differences in the structural network characteristics contributing to well-being, stepwise regression analyses were performed. Global well-being was regressed onto the same-sex and cross-sex network characteristics: Number of Kin and Non-Kin Network Members, and, Frequency of Contact with Kin and Non-Kin Network Members as the independent variables. For the men, three network characteristics entered the equation. At the first step, Frequency of Contact with Cross-Sex Non-Kin was entered (R 2 = .07; F(l,99) = 7.70, £ < .01). At the second step, the Number of Cross-Sex Non-Kin Network Members was entered (R 2 = .11; F(2,98) = 6.31, £ < .005). At the third step, Frequency of Contact with Same-Sex Kin was entered (R 2 = .15; F(3,97) = 5.92, £ < .001) (Table 7). For the women, there were no structural network variables that were significantly related to global well-being, thus no stepwise analysis was performed. Same-Sex Bonds and Well-Being p. 103 Table 6 Sex Differences in Same-Sex Network Composition and Frequency of Contact Network Variable Women Men Number of Kin M. 3.30 2.20 4.50 < .001 S.D. 1.50 1.90 Number of Non-Kin M 4.90 4.50 1.00 > .05 S.D. 2.40 2.50 Frequency of Contact - Kin M 10.40 6.50 4.80 < .005 S.D. 5.80 5.80 Frequency of Contact - Non-Kin M 16.09 16.10 .00 > .05 S.D. 9.14 8.60 Same-Sex Bonds and Well-Being p. 104 Table 7 Global Well-Being Regressed onto Network Characteristics: Men Step Variable Entered R.2 F £ 1 frequency of contact with cross-sex non-kin .07 7.70 <.01 2 number of cross-sex non-kin network members .11 6.31 <.005 3 frequency of contact with same-sex kin .15 5.92 <.001 Same-Sex Bonds and Well-Being p. 105 G. Sex Differences in the Number of Network Members Meeting Relational Provisions Hypotheses 6 - 1 3 explored sex differences in the strength of men's and women's same-sex and cross-sex bonds for supplying the Attachment, Social Integration, Reassurance of Worth, Reliable Alliance, Help and Guidance, and Opportunity for Nurturance provisions, independent of the number of people in their social networks supplying each provision. In this section, sex differences in the number of same-sex and cross-sex network members supplying each of the six relational provisions, independent of their strength, is explored. A multivariate analysis of variance (MANOVA) with one between subjects factor (sex of respondent) and one within subjects factor (relationship gender composition) was performed on the six dependent variables Attachment, Social Integration, Reassurance of Worth, Reliable Alliance, Help and Guidance and Opportunity for Nurturance. By the likelihood ratio test, a significant main effect was obtained for sex' of respondent, F_(l,200) = 8.13, £ < .001, for relationship gender composition F(l,200) = 43.01, £ < .001 and for their interaction effect, F(l,200) = 12.58, £ < .001. Means and standard deviations are illustrated in Tables 8 and 9. Same-Sex Bonds and Well-Being p. 106 Table 8 Number of Women's Same-Sex and Cross-Sex Relationships Meeting Relational Provisions Provision Same-Sex Cross-Sex Attachment M S.D. Social Integration M S.D. Reassurance of Worth M S.D. Reliable Alliance M S.D. Help and Guidance M S.D. Opportunity for Nurturance M S.D. 4.40 2.60 4.40 2.40 4.50 2.80 5.30 3.00 4.03 2.50 3.30 2.40 2.90 2.30 3.00 2.00 3.00 2.00 4.00 2.50 2.60 2.20 2.30 2.10 7*34 < .001 6.67 < .001 5.34 < .001 6.30 < .001 6.65 < .001 4.61 < .001 Same-Sex Bonds and Well-Being p. 107 Table 9 Number of Men's Same-Sex and Cross-Sex Relationships Meeting Relational Provisions Provision Same-Sex Cross-Sex Attachment M S.D. Social Integration M S.D. Reassurance of Worth M S.D. Reliable Alliance M S.D. Help and Guidance M S.D. Opportunity for Nurturance M S.D. 3.00 2.40 3.40 2.30 3.70 2.40 4.20 2.70 3.30 2.30 2.40 2.50 2.60 2.10 2.50 2.10 3.40 2.40 3.40 2.20 2.60 2.00 2.20 2.10 2.08 > .05 4.13 < .001 1.35 > .05 4.01 < .001 3.78 < .001 1.49 > .05 Same-Sex Bonds and Well-Being p. 108 A simple main effects analysis was conducted in order to test the effects of each significant interaction. The significant interaction effect was attributable to the greater number of same-sex relationships supplying women with the Attachment F(l,384) = 33.86, £ < .001 and Opportunity for Nurturance F(l,372) = 33.81, £ < .001 provisions in comparison to men's same-sex relationships. There were no significant differences between men and women on the number of network members supplying the remaining provisions. Univariate dependent sample t-tests were conducted in order to test within sex differences. These results are also illustrated in Tables 8 and 9. An experiment-wise error rate of .05 was maintained through the aportionment of alpha. The results indicate that women report receiving supplies of the following provisions from more people in their same-sex networks than from people in their cross-sex networks: Attachment, t(l,200) = 7.34, £ < .001; Social Integration, _t(l,200) = 6.67, £ <.001; Reassurance of Worth jt(l,200) = 5.34, £ < .001; Reliable Alliance, _t(l,200) = 6.30, p < .001; Help and Guidance, t(l,200) = 6.65, £ < .001 and the Opportunity for Nurturance, t(l,200) = 4.61, £ < .001. Further, the men report receiving supplies of the Social Integration _t(l,200) = 4.13, £ < .001, Reliable Alliance _t(l,200) = 4.01, £ < .001 and Help and Guidance t(l,200) = 3.78, £ < .001 provisions from more people in their same-sex networks than from people in their cross-sex networks. Non-significant differences were obtained for the remaining three provisions. Same-Sex Bonds and Well-Being p. 109 H. Supplies of Relational Provisions predicting Global Well-Being In order to explore whether the number of same-sex and cross-sex network members supplying each of the relational provisions contributes to global well-being, a stepwise regression analysis was performed. Global well-being was regressed, separately, by sex, onto the number of same-sex and cross-sex network members supplying: Attachment, Social Integration, Reassurance of Worth, Reliable Alliance, Help and Guidance and Opportunity for Nurturance, as the independant variables. For the men, one variable entered the equation: Cross-Sex Opportunity for Nurturance, resulting in a bivariate Pearson j_of .22 (R 2 = .05; F(l,99) = 5.48, £ < .05), illustrated in Table 10. For the women, one variable entered the equation, Same-Sex Help and Guidance, resulting in a bivariate Pearson r_ of .30 (R 2 = .09; F(l,98) = 10.11, £ < .005), illustrated in Table 11. I. Summary of the Network Data Results In this section, two research questions were explored. The first was concerned with sex differences in the structural characteristics of the respondents networks. An analysis of the network data indicated that overall, women reported a larger number of and greater frequency of contact with same-sex and cross-sex network members than did the men. However, when kin and non-kin network characteristics were analyzed separately, these differences were largely attributable to the greater number and frequency of contact with Same-Sex Bonds and Well-Being p. 110 Table 10 Global Well-Being Regressed onto the Number of Network Members Supplying Each Relational Provision: Men Step Variable Entered R 2 1 cross-sex opportunity for nurturance .05 5.48 <.05 Table 11 Global Well-Being Regressed onto the Number of Network Members Supplying Each Relational Provision: Women Step Variable Entered R 2 F_ £ 1 same-sex help and guidance .09 10.11 <.005 Same-Sex Bonds and Well-Being p. 111 same-sex kin reported by the women. There were no differences in the number and frequency of contact with same-sex non-kin and cross-sex non-kin. A separate analysis was conducted on the number of network members providing each of the six provisions. The women report receiving all six of the relational provisions from more people within their same-sex networks than from people within their cross-sex networks. The men report receiving supplies of the social integration, reliable alliance and help and guidance provisions from more people within their same-sex networks than from people within their cross-sex networks. The second research question was concerned with sex differences in the network characteristics contributing to global well-being. Two separate stepwise regression analyses were conducted separately, by sex. The first examined the network characteristics contributing to global well-being and the second examined whether the number of network members supplying each provision contributed to global well-being. For the men, three network characteristics were predictive of global well-being: frequency of contact with cross-sex non-kin, number of cross-sex non-kin network members and frequency of contact with same-sex kin. There were no significant structural network characteristics contributing to women's global well-being. In the second analysis, the number of cross-sex network members supplying the opportunity for nurturance provision was predictive of global well-being for the men. For the women, the number of same-sex network members supplying the help and guidance provision was predictive of global well-being. The implications of these results are discussed in Chapter Five. Same-Sex Bonds and Well-Being p. 112 CHAPTER V: DISCUSSION AND IMPLICATIONS This chapter is divided into five sections. In the first section, the discussion is focused on the findings relating to the four research questions outlined in Chapter One. In the second section, the limitations of the study are outlined. In the third and fourth sections, the theoretical and practical implications of the study are examined. In the closing sections of the chapter, the conclusions drawn from the study are outlined. A. Discussion of the Findings relating to the  Four Research Questions In developing the conceptual framework for this study, the investigator made a distinction between the factors that contribute to the attainment of social support and the factors that contribute to the relationship between social support and global well-being. The first two research questions focused on assessing the relevance of the variables "sex of respondent" and "relationship gender composition" to the attainment of social support. Questions three and four focused on assessing the relevance of the variables "sex of respondent" and "relationship gender composition" to the association between social support and global well-being. The four research questions are listed below. The results of this study appear to support the validity of this conceptual distinction. 1) Overall, are men's same-sex relationships stronger than women's same-sex relationships? Same-Sex Bonds and Well-Being p. 113 2) Do women's and men's same-sex bonds provide higher levels of the individual relational provisions than do women's and men's cross-sex bonds? 3) Are satisfying social bonds more strongly associated with the global well-being of women than with the global well-being of men? 4) Are satisfying same-sex bonds more strongly associated with global well-being than satisfying cross-sex bonds? In the sections to follow, the findings relating to these questions are discussed.- These findings only pertain to the relationships of married men and women. 1. The Effects of "Sex of Respondent" and "Relationship Gender  Composition" on Overall Level of Social Support: Are men's same-sex  relationships stronger than women's same-sex relationships? In this section, data relevant to differences in the overall level of social support provided by men's and women's same-sex and cross-sex bonds are discussed. Four findings of the study are relevant to this discussion. First, contrary to the theories of Lipman-Blumen (1976) and Tiger (1969), overall, worqen's same-sex bonds appear to provide higher levels of social support than do . men's same-sex bonds, supporting the investigator's and Safilios-Rothschild's (1981) predictions. Same-Sex Bonds and Well-Being p. 114 Second, there do not appear to be significant differences in the overall level of social support provided by men's same-sex and cross-sex bonds, failing to support Lipman-Blumen's (1976) theory as well as the prediction of Safilios-Rothschild (1981) and the investigator. Third, women's same-sex relationships appear to provide higher levels of social support than do their cross-sex relationships, supporting the predictions of Safilios-Rothschild and the investigator and again failing to support Lipman-Blumen's theory. Finally, contrary to the predictions derived from Safilios-Rothschild's and Lipman-Blumen's theories, there do not appear to be significant differences in the overall level of social support provided by men's and women's cross-sex bonds. These results draw attention to a central theme organizing the data, namely, that women's overall same-sex or homosocial relationships appear to provide higher levels of social support than do men's same-sex or homosocial relationships. This theme both confirms and extends the findings of past investigations. Consistent with past studies comparing the same-sex bonds of men and women (Booth, 1972; Gibb, Auerbach & Fox, 1981; Rose, 1985; Wright, 1982), the results of this investigation support the greater strength of women's same-sex bonds and fail to support theories purporting to the greater strength of men's same-sex bonds. The results of this investigation also serve to extend the findings of past investigations. This study extends the results of previous investigations by comparing the overall level of social support provided by men's and women's same-sex bonds with their cross-sex ties. The incorporation of this comparison Same-Sex Bonds and Well-Being p. 115 in the present study serves to underscore the greater importance and higher level of social support provided by women's same-sex relationships both in relation to men's same-sex bonds and to women's cross-sex bonds. This comparative dimension also serves to highlight the relatively less important role of men's same-sex relationships compared to both women's same-sex bonds and to men's cross-sex bonds, in terms of overall social support. The present investigation also extends the results of past studies by conceptualizing same-sex bonds to encompass a more comprehensive range of homosocial relationships than were covered by previous studies. Thus, the results of this investigation extend the generalizability of past research investigations by pointing to the higher level of social support and importance of women's same-sex bonds in comparison to men's same-sex bonds across a broader range of relationships and contexts. In the following section differences in the level of individual relational provisions supplied through same-sex and cross-sex bonds are discussed. 2. The Effects of "Sex of Respondent" and "Relationship Gender  Composition" on the Level of Individual Relational Provisions: Do  women's and men's same-sex bonds provide higher levels of the individual  relational provisions than do women's and men's cross-sex bonds? In this section, data relevant to differences in the level of individual provisions supplied through men's and women's same-sex and cross-sex bonds are discussed. Two major findings are relevant to this discussion. First, with respect to the same-sex relationships of women, the research evidence indicates strong, but not complete support for Bernard's theory and fails to support Same-Sex Bonds and Well-Being p. 116 Lipman-Blumen's and Weiss's theories. Women's same-sex relationships appear to provide significantly higher levels of Weiss's provisions of attachment, social integration, reassurance of worth, help and guidance, and opportunity for nurturance than do their cross-sex relationships. These findings are further supported by the analysis of the network data. Not only do women's same-sex networks appear significantly larger in number than their cross-sex networks, but women also report receiving supplies of each provision (including reliable alliance) from more people within their same-sex networks than from people within their cross-sex networks. These findings extend and elaborate upon the theme identified in the previous section of this discussion by further underscoring the higher levels of social support as well as the greater comprehensiveness of women's same-sex bonds. Contrary to Weiss's theory, the findings draw attention to the importance of gender composition as a specific variable contributing to the women's higher attainment of specific dimensions of social support. Qualitatively, these results characterize the provisions which women receive through their same-sex relationships as an intimate, secure and emotional relationship (attachment) that provides a sense of integrated group involvement, for sharing similar concerns, social activities and interests (social integration), affirmation, respect, admiration and recognition for their skills, abilities and competencies (reassurance of worth). Women's same-sex relationships also appear to provide the opportunity to discuss personal concerns, decisions and problem-solving strategies (help and guidance). Finally, women's same-sex relationships appear Same-Sex Bonds and Well-Being p. 117 to provide them with a sense of being needed for the care and well-being of others (opportunity for nurturance). Consistent with the findings of Goodenow's (1985) investigation, the results of this study affirm the importance of women's same-sex relationships as a vital resource for obtaining specific dimensions of social support, serving to complement or to supplement the marital bond. Further, these results fail to support Lipman-Blumen's conceptualization of women's same-sex bonds as essentially irrelevant, unfulfilling and impoverished. The results of this study also serve to extend the results of past investigations. Past studies of women's same-sex relationships have consistently documented the emotional-expressive qualities of intimacy (Bell, 1981; Booth & Hess, 1972; Davidson <3c Packard, 1978; Safilios-Rothschild, 1981), involvement (Seiden & Bart, 1975; Smith-Rosenberg, 1975; Wright, 1982), affirmation and encouragement (Davidson, 1983; Goodenow, 1985; Harris, 1984), but have neglected to integrate the emotional domain with the more instrumental task-oriented component of women's relationships. In the present study the measure of social support administered, Social Provisions Scale, contained both emotional-expressive and instrumental relationship dimensions. The provisions for attachment, reassurance of worth, help and guidance, and opportunity for nurturance appear to comprise the more emotional-expressive domain and the provisions for social integration, reliable alliance, and help and guidance appear to comprise the more instrumentally oriented domain. The overall picture of these results would seem to suggest that women's same-sex relationships fulfill Same-Sex Bonds and Well-Being p. 118 Weiss's (1974) emotional-expressive relationship dimensions and some of the more instrumental task-oriented relationship dimensions. It can be speculated that the combination and interfacing of these dimensions in women's same-sex relationships provide women with an opportunity for meeting their instrumental needs within an affiliative context, lending further support to Bernard's (1976) and Smith-Rosenberg's (1975) characterization of women's same-sex bonds. The present investigation extends the results of previous investigations by comparing the dimensions of social support women receive through their same-sex bonds with their cross-sex bonds. This comparison serves to accent and sharpen the stronger and more vital character of women's same-sex relationships as well as to highlight their importance for supplying higher levels of relational provisions not as strongly obtained through their cross-sex bonds. The second major finding relevant to this discussion concerns the relational provisions supplied through men's same-sex and cross-sex relationships. The research evidence indicates only partial support for Lipman-Blumen's (1976) and Weiss's (1974) theories and fails to support the predictions suggested by the work of Bell (1981) and Safilios-Rothschild (1981). Men's same-sex relationships appear to provide higher levels of the social integration and reliable alliance provisions than do their cross-sex bonds, indicating only partial support for Lipman-Blumen's theory. However, there do not appear to be significant differences between men's same-sex and cross-sex relationships for supplying the reassurance of worth and help and guidance provisions, supporting Weiss's (1974) theory and failing to support Lipman-Blumen's theory. These findings are Same-Sex Bonds and Well-Being p. 119 corroborated by the analysis of the network data. Men also report receiving supplies of the social integration and reliable alliance provisions from more people within their same-sex networks than from people within their cross-sex networks. These findings are consistent with the results of past investigations. Paralleling Weiss's social integration and reliable alliance provisions, past studies of men's same-sex bonds have also characterized their relationships as collegial and task-oriented (Bell, 1981; Dickens <5c Perlman, 1982; Shaver Sc Buhrmester, 1983; Yoon, 1978). Further, the non-significant difference between men's same-sex and cross-sex bonds for supplying the reassurance of worth and help and guidance provisions seem to support conceptualizations of the less intimate character of male socialization patterns and relational style (Bernard, 1976; Pleck, 1976; Reisman, 1981). This finding is consistent with the results of Rose's (1985) study. Rose found that men's same-sex and cross-sex bonds did not differ in supplying relationship intimacy. In contrast to the findings for the women, men's instrumental needs do not appear to be as strongly met within an affiliative context. Further, the results suggest that gender composition is a more significant variable affecting the attainment of specific dimensions of social support for women than for men. In conclusion, the data do not appear to support Lipman-Blumen's conceptualization of male same-sex bonds as a vehicle for meeting the majority of their needs. On the contrary, the data indicate that, consistent with male socialization, men's same-sex relationships appear to be focused on the attainment of collegial and instrumental needs. Same-Sex Bonds and Well-Being p. 120 3. The Effect of "Sex of Respondent" on the association between Level of  Social Support and Global Well-Being: Are satisfying social bonds more  strongly associated with the global well-being of women than with the  global well-being of men? In this section, data illustrating the overall relationship between level of social support (same-sex and cross-sex bonds combined) and global well-being are discussed. The first part of the discussion considers the overall relationship between level of social support and global well-being, then, sex differences in this relationship are examined. With respect to the overall relationship between level of social support and global well-being, the correlational relationship between these two variables is highly significant. This finding points to an important theme characterizing the data, namely, that satisfying social bonds is strongly associated with global well-being during the course of everyday life. Moreover, this finding suggests strong support for the major theoretical proposition arising from Weiss's (1974) theory. Weiss has conceptualized socially supportive relationships as an important source of well-being to individuals within the context of their daily lives. While previous studies have established that social support buffers stress, this study indicates that social support also enhances global well-being during daily life, as Weiss has proposed. The implications of this finding in relation to Weiss's theory, for theories of social support and for theories of adult development are discussed more fully in a later section of this chapter. Same-Sex Bonds and Well-Being p. 121 Second, with respect to the question of sex differences in the relationship between level of social support and global well-being, the data appear to indicate that satisfaction with social bonds is strongly associated with global well-being for both sexes. That is, there do not appear to be sex differences in the relationship between level of social support and global well-being. This finding points to another important theme characterizing the data, namely, that satisfaction with social bonds is equally relevant for the global well-being of men and women. Both Bernard (1976) and Miller (1976) have suggested that a woman's sense of well-being is more strongly affected by relationship satisfaction than is a man's sense of well-being, while Pearlin (1975), has suggested that a man's sense of well-being is more strongly affected by career satisfaction than is a woman's sense of well-being. These findings tend to suggest that the importance of satisfying social bonds may have been an underestimated component of a man's sense of well-being. Thus, this result appears to indicate support for propositions which seem to arise from Weiss's (1974) theory, but fail to support propositions derived from Bernard's (1976) and Miller's (1976) theories. Weiss (1974) has not commented on the question of sex differences in the relationship between satisfying social bonds and well-being. Because he does not address sex differences in his theory, he seems to have implied, by omission, that there are no sex differences in the relationship between satisfying social bonds and well-being. Alternatively, Bernard (1976) and Miller (1976) have suggested that a woman's sense of well-being is more strongly effected by satisfying social Same-Sex Bonds and Well-Being p. 122 bonds than is a man's sense of well-being because women are more strongly socialized towards defining their sense of self in relation to their success in establishing and maintaining satisfying affiliative ties. Although the results of this study do not appear to support the idea of sex differences in the overall relationship between satisfying social bonds and global well-being, the results of the study do indicate sex differences in the specific characteristics of men's and women's social networks that contribute to their global well-being. Generally speaking, the structural characteristics of men's social networks appear to contribute more strongly to their global well-being than do the structural characteristics of women's social networks. For the men, four structural network characteristics appear to predict global well-being. These characteristics are: frequency of contact with cross-sex kin; the number of cross-sex non-kin network members; frequency of contact with same-sex kin and, the number of women in their social networks supplying the opportunity for nurturance provision. Collectively, these results appear to suggest that, independent of relational quality, contact with women contributes to men's global well-being. This interpretation is consistent with the findings of the Wheeler, Reis and Nezlek (1983) study. Wheeler, Reis and Nezlek have suggested that because women are more skilled in providing an affectively rich type of social exchange, contact with women contributes to men's global well-being. In addition, consistent with patterns of male sex-role socialization, these results can be interpreted as suggesting that contact with women may provide men with the opportunity to fulfill a protective, care-giving role that affirms their sex-role socialization. Same-Sex Bonds and Well-Being p. 123 For the women, only one structural network characteristic appears to predict global well-being—the number of women in their social networks supplying the help and guidance provision. These results suggest that, consistent with their relational style, women appear to benefit from having numerous same-sex social contacts in which personal concerns, problem-solving strategies and decisions are discussed. Thus, while there do not appear to be sex differences in the relationship between the overall quality of men's and women's social support to global well-being, there do appear to be sex differences in the contribution of network structure variables to men's and women's global well-being. 4. The Effect of "Gender Composition" in the association between Level of  Social Support and Global Well-Being: Are satisfying same-sex bonds  more strongly associated with global well-being than satisfying cross-sex  bonds? In this section, data relating to the question of differences in the amount of global well-being provided by same-sex and cross-sex bonds is discussed. The data appear to indicate that although same-sex and cross-sex bonds are strongly related to global well-being for both sexes, same-sex and cross-sex bonds are also equally related to global well-being for both sexes. That is, there do not appear to be differences in the association between relationship gender composition and global well-being. This finding points to an important theme characterizing the data, namely, that global well-being is related to both same-Same-Sex Bonds and Well-Being p. 124 sex and cross-sex relationships. This finding tends to suggest support for theoretical propositions implied by Weiss's (1974) theory of social relationships but fails to indicate support for propositions arising from Bernard's (1976) and Lipman-Blumen's (1976) respective theories of female and male same-sex bonding. Moreover, the findings also fail to support propositions arising from Safilios-Rothschild's (1981) and Lipman-Blumen's (1976) respective theories of male and female cross-sex bonding. Overall, the findings suggest a limitation to Bernard's (1976) theory of female same-sex bonding and Lipman-Blumen's (1976) theory of male same-sex bonding. These theorists have failed to recognize that benefits to well-being can be derived from cross-sex, non-marital relationships. While Bernard seems to have undervalued the importance of men's task-oriented relational style to women's well-being, Lipman-Blumen seems to have overvalued the importance of relating to powerful others for men's well-being. It would appear that Bernard and Lipman-Blumen have both underestimated the contribution of cross-sex bonds to men's and women's global well-being. In order to investigate the effect of relationship gender composition to global well-being more fully, additional correlational and regression analyses were conducted. Although these analyses further corroborate the equal contribution of men's and women's same-sex and cross-sex bonds to their global well-being, the results of the step-wise regression analyses do suggest the possibility that women's and men's global well-being are predicted by different relational provisions. Women's global well-being appears to be predicted by Same-Sex Bonds and Well-Being p. 125 obtaining reassurance of worth from their same-sex bonds and obtaining help and guidance from their cross-sex bonds. Alternatively, men's global well-being appears to be predicted by obtaining reassurance of worth and the opportunity for nurturance from their cross-sex bonds and obtaining the social integration provision from their same-sex bonds. While these results are intriguing, the substantial intercorrelations among these provisions interferes with the ability to make a fuller interpretation. The findings of the regression analyses, however, do raise the possibility that men's and women's same-sex and cross-sex bonds contribute in different ways to their global well-being, a topic for future researchers to consider. 5. Discussion of Questions Emerging from the Data An overall picture of the findings indicates the relevance of conceptually distinguishing the effect that sex of respondent and relationship gender composition has on the attainment of social support from the effect that sex of respondent and relationship gender composition has on the association between social support and global well-being. With respect to the level or quality of social support, both the sex of the respondent and the gender composition of the relationship appear to contribute to the overall level of social support and to the level of the individual relational provisions. However, gender composition appears to contribute to the quality of women's social support more strongly than to the quality of men's social support. Same-Sex Bonds and Well-Being p. 126 With respect to global well-being, neither the sex of the respondent nor the gender composition of the relationship appear to affect the overall association between social support and global well-being. For the men, the gender composition of the relationship does not appear to be a strong factor influencing the quality of social support, nor does it appear to be a strong factor influencing the overall relationship between social support and global well-being. Conversely, for the women, the gender composition of the relationship does appear to be a strong factor affecting the quality of social support, but it does not appear to be a strong factor affecting the overall relationship between social support and global well-being. This overall picture of the results raises two important questions. First, if women receive higher levels of the critical ingredients that contribute to global well-being (social support) then why aren't women higher in global well-being than men? Second, if women's same-sex bonds provide higher levels of the critical ingredients that contribute to global well-being than do their cross-sex bonds, then why aren't women's same-sex bonds more strongly associated with global well-being than women's cross-sex bonds? In the sections to follow, a discussion of the statistical and theoretical propositions arising from these questions is considered. With respect to the question of sex differences in the relationship between the overall quality of social support and global well-being, it is first important to examine the theoretical assumptions underlying this relationship. The relationship between social support and well-being is assumed to be Same-Sex Bonds and Well-Being p. 127 positively and linearly related. That is, with increasing levels of social support there are, correspondingly, increasing levels of well-being. However, this assumption may be somewhat faulty or limited. The results of the study suggest the proposition that there may be a "ceiling effect" operating in the gradient between social support and well-being. That is, there may be an optimal level of social support required for well-being and increases beyond this level may not further elevate one's sense of well-being. Thus, although the quality of women's social support exceeds the quality of men's social support, it may not increase the amount of global well-being women obtain through their social relationships. It is also important to emphasize that social relationships provide only one source of well-being. Other factors in the individual's life such as satisfaction in a love relationship, satisfaction with physical health and, satisfaction in work provide additional sources of well-being. Thus, social relationships may only provide a finite amount of well-being. We now turn our attention to the question of differences in the relationship between the quality of social support provided through women's same-sex and cross-sex bonds and global well-being. One interpretation of the findings is that although women receive higher levels of social support from other women and weaker levels of social support from men, women may undervalue the supportiveness available within their same-sex relationships and overvalue the supportiveness available within their cross-sex relationships. This proposition is empirically consistent with the results of Wheeler, Reis and Same-Sex Bonds and Well-Being p. 128 Nezlek's (1983) study and is theoretically consistent with the work of Bernard (1976), Caplan (1981) and Lipman-Blumen (1976). In the Wheeler, Reis and Nezlek (1983) study, meaningful interactions with men were inversely related to loneliness among a sample of women. They concluded that because intimate, self-disclosing and meaningful interactions with men are encountered less frequently, they contain greater reward value. In a related vein, Bernard (1976), Caplan (1981) and Lipman-Blumen (1976) indicate that the cultural devaluation of women and valuation of men affects the rewards derived from same-sex and cross-sex bonds. Bernard (1976) has commented on how societal devaluation of women and corresponding overvaluation of men have affected women's same-sex relationships. She has suggested that devaluation of women in general and of women's same-sex relationships in particular created a context in which women were ashamed of their affiliative style and sought to disassociate themselves from their female identities. Caplan (1981) and Lipman-Blumen (1976) have suggested that women sought relationships with men in order to elevate themselves from their disparaged status as women. Caplan (1981) has cited Kennedy's (1976) notion of "horizontal hostility" to account for this phenomenon. Kennedy (1976) has said that: If you have a sense of your own worthlessness, then somebody else from your class or race or religion is clearly not to be looked up to. This is one of the bases for the pathology of women saying, "I don't Same-Sex Bonds and Well-Being p. 129 get along with women. I get along with men; they're superior, so if I get along with them I'm superior . . ." (p. 51). The discrepancy between the level of social support and the level of global well-being women derive through their same-sex and cross-sex bonds may be interpreted within this context. That is, women may devalue the intimacy, assistance and nurturance they receive from other women and overvalue the intimacy, assistance and nurturance they receive from men. This proposition also suggests that the relationship between social support and well-being may be more complex than previously thought. It may be that in order for social support to be maximally effective as a mental health enhancing process, the recipient of support may need to value and respect the provisions of support offered by the giver of support. Thus, the affective quality of the relationship, including women's attitudes towards other women may further clarify the relationship between social support and well-being. A second interpretation of this result may be related to the negative aspects of intimacy. The data indicate that women's same-sex relationships are more intimate than their cross-sex relationships. Hatfield (1984) has observed that greater interpersonal intimacy heightens the risks of interpersonal conflict and may elicit fears of rejection, abandonment, vulnerability and engulfment. Thus, although the present investigation focuses on drawing attention to the more positive resources available within women's same-sex relationships, the noted discrepancy may reflect the more stressful aspects of intimate same-sex bonds. The results of this study suggest the possibility that the stresses Same-Sex Bonds and Well-Being p. 130 associated with affectively rich and intimate relationships may diminish the strength of their more positive components. Several factors, working separately, or in combination may account for the discrepancy between the different levels of social support supplied through women's same-sex and cross-sex relationships and global well-being. Further studies are needed to clarify the factors responsible for this intriguing finding. B. Limitations Of The Study The results of the study appear to be limited by three factors: The inability to derive causal relationships; the inability to generalize the results beyond the characteristics of the sample; and by questions concerning the interpretation of the amended Social Provisions Scale items. In the section to follow, these three issues are discussed. 1. Causality Research in social support has largely been focused on the use of correlational methods to establish a link between social support and well-being. As such, the independent variable, social support, is measured rather than experimentally controlled (Dooley, 1985). Correlational designs are vulnerable to two important threats to their internal validity, spuriousness and reverse causation. As Dooley explains "an association between two variables is said to be spurious when it is produced by a third variable" (p. 111). In the present study Same-Sex Bonds and Well-Being p. 131 variables such as marital satisfaction, social class, or personal characteristics such as social competence may be responsible for greater well-being rather than satisfaction with same-sex or cross-sex bonds. Reverse causation occurs when "the observed association is causal but the direction of causation is opposite from that hypothesized" (p. 111). For example, in this study, it may be that respondents who are higher on measures of well-being tend to have stronger and more satisfying interpersonal relationships. While a laboratory simulation design might be suggested as a means of ameliorating and unravelling the confounding effects of spuriousness and reverse causation, the artificiality of experimentally manipulating social support fails to capture the long-term, in-depth relationships characterizing the essence of social support. 2. Generalizability The generalizability of these findings is limited to men and women who are compatible with the characteristics of the sample. Therefore, the generalizability of the results is limited to married men and women between 25 and 45 years of age, who are well-educated (at least a B.A.) and professionally trained. The issue of sexual orientation was not addressed in this study, except insofar as married respondents can be assumed to be heterosexual. However, in a recent study, Henderson (1986) investigated differences in the same-sex relationships of heterosexual and lesbian feminist women. The findings indicate that women experience their same-sex relationships similarly, Same-Sex Bonds and Weil-Being p. 132 regardless of their sexual preference. As such, there is some evidence to suggest that the generalizability of the findings may also extend to the close relationships of lesbian women. The sample comprised respondents from several ethnic groups including those of Chinese, European, East Indian, Native Indian and West Indian descent. However, differences among these groups were not analyzed separately. Consequently, the generalizability of the findings to specific ethnic groups cannot be ascertained. Moreover, information on the socioeconomic status of the respondents was not collected. As such, the generalizability of the findings across socioeconomic groups cannot be determined. 3. Social Provisions Scale For the present investigation, all the items in the Social Provision Scale (Russell & Cutrona, 1984) were amended in order to assess differences in the relational provisions supplied through men's and women's same-sex and cross-sex bonds. For example, in the original formulation, item twelve states: "There are people I could talk to about important decisions in my life". The amended item states: "There are men/women I could talk to about important decisions in my life". The results of the study do not indicate if these items were interpreted differently within each relational context. For example, within the context of women's same-sex relationships this item may reflect a more emotionally-Same-Sex Bonds and Well-Being p. 133 expressive dimension but within the context of their cross-sex ties, this item may reflect a more problem-solving, task-oriented authority figure context. Thus, the influence of sex-role stereotypes may have affected how the respondents reacted to the items. The relevance of this issue cannot be clearly determined from the results of the study, and suggests a possible limitation to the interpretation of the results. It also suggests the need for measures specifically designed to measure same-sex and cross-sex bonds, with content validity for each. In addition, the six dimensions of the Social Provisions Scale are substantially intercorrelated. This raises some questions as to the validity of the six-factor model proposed by the scale developers. Further work on the factor structure of the scale through exploratory and confirmatory methods might clarify the factors underlying the scale. C. Theoretical Implications of the Study 1. Implications for Weiss's Theory of Social Relationships The results of the study indicate a confirmation, extension, and limitation to Weiss's theory of social relationships and well-being. First, an important result of the study suggests support for an association between the satisfactory fulfilment of Weiss's relational provisions and well-being. This result extends the current application of Weiss's theory. Recent research interest in Weiss's theory has focused on validating his concept of "relational deficit" and its Same-Sex Bonds and Well-Being p. 134 relationship to loneliness (Cutrona, 1982; Russell, Cutrona, Rose & Yurko, 1984; Shaver & Rubenstein 1982). However, in presenting his theory Weiss has hypothesized that the satisfactory attainment of the six relational provisions contributed to positive mental health (well-being) outcomes. Researchers have focused their attention on the deficit component of Weiss's theory and its mental health consequences. The results of this investigation support Weiss's conceptualization of the relational benefits obtained from social relationships and their contribution to positive mental health. Further, the results of the study also indicate support for viewing the importance of social support within the context of everyday life. The results of the study also suggest a partial limitation to Weiss's theory. Weiss has not addressed the question of differences in the effectiveness of men and women as social partners for supplying particular dimensions of social support. In effect, Weiss has assumed that same-sex bonds are as equally effective as cross-sex bonds for supplying the individual relational provisions. The results of the present investigation suggest that relationship gender composition is an important variable affecting the attainment of specific dimensions of social support, particularly for the women. The inclusion and recognition of the variable gender composition in the present study therefore serves to extend and enhance Weiss's typology of relational needs. In addition, Weiss has not addressed the question of sex differences in the relationship between social support and well-being. He has implied, by omission, that the relationship between social support and well-being is similar for both Same-Sex Bonds and Well-Being p. 135 sexes. The results of this study seem to indicate some support for this proposition. Similarly, Weiss has not addressed the question of differences in the association between same-sex and cross-sex bonds with well-being. He has implied that same-sex bonds contribute to well-being as much as cross-sex bonds. Although the results of this study suggest that the overall relationship between same-sex bonds and global well-being is similar to the overall relationship between cross-sex bonds and global well-being, there is also evidence to suggest that different aspects of same-sex and cross-sex bonds may contribute to global well-being. In conclusion, the results of this study do not indicate support for overlooking the question of gender composition in the attainment of social support. However, the results of this study do indicate support for overlooking the question of sex differences in the relationship between social support and well-being and only mixed support for overlooking the role of gender composition in the relationship between social support and global well-being. The results of the study also point to a more general limitation to Weiss's theory. Although Weiss has hypothesized that relationships become specialized in the type of relational provision fulfilled, the individual dimensions of the Social Provisions Scale were found to be highly intercorrelated. This suggests that relationships are more multifaceted, providing the individual with a variety of relational provisions perhaps simultaneously. This finding is consistent with Same-Sex Bonds and Well-Being p. 136 Cutrona and Russell's (1986) results and raises the question as to whether there is a more general factor characterizing social relationships than the model proposed by Weiss. The results suggest that relationships appear to be broader in scope and appear to fulfill several support dimensions simultaneously. 2. Implications for Bernard's Theory of Homosociality The results of the study confirm, limit and extend Bernard's theory of homosociality. Bernard has defined homosociality as "the different ways that men and women relate to members of the same sex" (p. 227-8). Bernard's theory has concentrated on articulating differences in the relational styles of men and women to account for the "relational deficits" women experience in their cross-sex relationships and when access to same-sex relationships is limited. She has characterized the relational style of women as affiliative and emotionally expressive and has maintained that women's relational needs are met within an affiliative context. This style is contrasted with men's relational style which Bernard has characterized as more emotionally inexpressive and instrumental. The results of the study confirm Bernard's conceptualization of "differences" in the relational styles of men and women. Women's same-sex relationships appear to be more strongly oriented towards an emotionally expressive quality while men's same-sex relationships appear to be more strongly focused on instrumental activities. These results also confirm Bernard's vision of women's same-sex bonds as a vital and important contributor to their well-being. Same-Sex Bonds and Well-Being p. 137 The results of the study, however, also suggest limitations to Bernard's theory. Bernard has not recognized that a potent source of women's well-being is met within a less affiliative relational context and, that access to a different relational style may complement what is less effectively obtained through relationships with one's own sex and social style. Bernard focused only on the relational deficits women experience in their cross-sex relationships without recognizing the relational benefits. The results of the study also cast an interesting perspective on Bernard's (1976) theory of homosociality and female depression. Bernard has suggested that the societal devaluation of women's same-sex relationships and the corresponding overvaluation of women's cross-sex relationships has prevented the development of strong bonds between women. This, in turn, has led to relational deficit and hence, to depression. The results of this study indicate that women obtain higher levels of social support from other women than from men. This suggests that the devaluation of women's relationships does not appear to have prevented them from establishing strong same-sex bonds. However, the results of this study also indicate that the amount of global well-being women derive from their same-sex and cross-sex bonds is not commensurate with the level of social support women receive from them. This finding leads to the proposition that while the devaluation of women's same-sex relationships does not appear to prevent women from establishing stronger same-sex than cross-sex bonds, it may prevent women from fully valuing and benefitting from the social support they receive from Same-Sex Bonds and Well-Being p. 138 other women. Thus, rather than directly linking relational deficit to depression, as Bernard has suggested, there may be another factor involved. That is, women who tend to devalue the supportiveness, encouragement and affirmation they receive from other women may not be deprived at the level of getting support, but may be deprived at the level of fully accepting its influence and impact. Future researchers might consider assessing whether the variable, women's attitudes towards other women, affects the attainment of social support as well as the association between social support and well-being. 3. Implications for Lipman-Blumen's Theory of Homosociality Overall, the results of the study do not seem to support Lipman-Blumen's theory of male homosociality. Lipman-Blumen (1976) has defined homosociality as the "seeking, enjoyment and/or preference for the company of the same-sex" (p. 7). Whereas Bernard has conceptualized homosociality within the context of sex differences in social styles, Lipman-Blumen has conceptualized homosociality within the context of differences in relationship strength. The results of this study do not appear to support Lipman-Blumen's formulation of male same-sex bonds as stronger and as more comprehensive than women's same-sex bonds. Conversely, the results of the study seem to suggest that men's same-sex relationships are weaker and less comprehensive than women's same-sex relationships. Men's same-sex relationships appear to be restricted to a collegial and task-oriented focus, consistent with patterns of male sex-role socialization. Further, the results of the study do not appear to support Lipman-Blumen's conceptualization of men's cross-sex bonds as "burdensome" and limited to the Same-Sex Bonds and Well-Being p. 139 fulfillment of paternity needs. To the contrary, the results of the study seem to indicate that, independent of the spousal relationship, an important source of men's well-being is obtained through their cross-sex relationships. Lipman-Blumen appears to have underestimated the vital importance and contribution of cross-sex relationships with mothers, sisters, daughters, co-workers and supervisors as a valuable source of men's well-being. Moreover, the results of the investigation fail to support Lipman-Blumen's definition of homosociality as "the preference for the company of the same-sex". Although Lipman-Blumen does not clarify the meaning of "preference", perhaps one possible index of preference is time spent socializing with members of the same-sex. The results of the network data indicate that for men, frequency of contact with women and the number of women in their networks contributes to their well-being more strongly than socializing with men. This finding may suggest that men enjoy and prefer the company of women. Further, Lipman-Blumen's theory of male same-sex bonding and female cross-sex bonding is predicated upon the "attraction to powerful others" concept borrowed from the social learning theorists. The results of this study do not appear to support the power concept as an explanatory variable for male bonding or for female cross-sex bonding. Overall, the results of this study do not support Lipman-Blumen's theory of male homosociality, but different results may have been obtained if different relationship dimensions such as status, power, dominance, intellectual stimulation, and utility value had been assessed. Alternatively, Lipman-Blumen's theory may have greater applicability within a specific sphere of men's relationships, such as work relationships. Same-Sex Bonds and Weil-Being p. 140 4. Implications for Theories of Social Support The results of the study have implications for theory development in social support. Although social support has been identified as a conceptual focus for the integration of theory and research (Gottlieb, 1983), researchers have generally failed to incorporate the knowledge gleaned from related disciplines in designing their research studies. The integration of these related areas in the present study, including the literature on friendship, homosociality, sex differences and well-being serves to enhance a more refined understanding of the specific parameters influencing the attainment of social support and well-being. Specifically, the results of the study draw attention to the importance of gender composition as a critical variable affecting the quality and the structure of social support and therefore should be included in future investigations. The results of the study also suggest a re-examination of the overly constraining and value-laden "crisis-oriented" conceptualization of social support characterizing much of the literature. Research in social support has failed to recognize and to validate the importance of social relationships during the course of everyday life. The results of this study highlight the naturally instrumental and emotionally sustaining elements of social relationships and point to the reconsideration and theoretical development of social support as an important mental health enhancing resource within the context of everyday life. Same-Sex Bonds and Well-Being p. 141 5. Implications for Theories of Adult Development The results of the study have implications for theories of adult development. Theories of adult development have focused on non-affiliative qualities such as autonomy and self-sufficiency as the hallmarks of healthy adult development and functioning (Broverman, Broverman, Clarkson, Rosenkrantz & Vogel, 1970). Correspondingly, affiliative needs have been treated as an indication of pathology, dependency and weakness. The pathologizing of women's affiliative needs has already been challenged by Gilligan (1977) and Miller (1976), who have drawn attention to women's greater needs for affiliation and ethic of caring as basic strengths. Further, an early study conducted by Shipley and Veroff (1952) on men's affiliative needs, has pointed out that men who express high need for affiliation are negatively judged by other males as 'dependent' and 'approval seeking.' The results of the present investigation suggest that affiliative relationships with others significantly contribute to adaptive functioning and well-being within the context of daily life. This result would suggest that theories of adult development recontextualize and incorporate affiliative social relationships as an indication of health rather than of pathology. Such affiliative relationships can be understood as an important resource contributing to adaptive functioning and well-being rather than as an indication of dependency or weakness. Same-Sex Bonds and Well-Being p. 142 D. Practical Implications of the Study 1. Implications for Counselling The results of the study have several implications for mental health professionals who counsel adults. The first implication is one of perspective. Counsellors should be apprised of the importance of social relationships as a potentially vital resource contributing to adaptive functioning and well-being in daily life. Moreover, counsellors should convey to their clients that social relationships are important for the fulfillment of normal affiliative needs and well-being during the course of daily life. Moreover, the results of the study suggest the importance of assessing and linking client strengths and difficulties within the context of his/her social network. Counsellors should assess the fulfillment of Weiss's relational provisions, identify areas of deficit, and design counselling interventions to assist clients in strengthening their social relationships. With respect to counselling women, therapists and clients should be apprised of the natural, therapeutic, instrumental and emotionally sustaining potential available within women's same-sex bonds, explore the barriers to their formation (Caplan, 1981), and encourage their development. Moreover, counsellors should help women to recontextualize their pull towards affiliation and emotional expressiveness as a basic strength which can enhance their adaptive functioning and well-being (Bernard, 1976; Miller, 1976) and not as an indication of their "neediness" or dependency. Same-Sex Bonds and Well-Being p. 143 With respect to counselling men, the results of the study suggest that a previously underestimated source of men's well-being is provided through their personal relationships. Counsellors should validate men's affiliative needs, explore their fears of vulnerability, intimacy, and homophobia as barriers that may prevent them from developing strong same-sex bonds (Lewis, 1978). Finally, with respect to counselling couples, therapists can link blocks in patterns of communication between spouses and areas of deprivation to differences in relational styles. Moreover, the findings suggest the importance of recognizing the supportive functions same-sex and cross-sex bonds provide same-sex and cross-sex bonds may supplement relational needs not supplied within the marital bond thereby reducing the pressure placed on the spousal relationship as the main vehicle for need fulfillment. 2. Implications for Future Research The results of the study indicate several exciting avenues for future research. In the section to follow, directions for future research in same-sex bonding, cross-sex bonding and social support are discussed. With respect to Bernard's theory of homosociality, future researchers might consider designing a study that would compare the relational provisions supplied through the marital relationship with the relational provisions supplied through women's same-sex and cross-sex (non-marital) bonds. Such a study would test Bernard's application of Weiss's (1974) "relational deficit" thesis to the Same-Sex Bonds and Well-Being p. 144 marital relationship. Such a study would indicate whether women's same-sex bonds serve to compensate for the relational deficits women experience in their heterosexual bonds. Future researchers might consider comparing the relational provisions obtained through the same-sex and cross-sex relationships of depressed and non-depressed women. Such a study would test Bernard's (1976) theory of homosociality and female depression. The results of the study suggest that a measure of sex-role attitudes, such as those developed by Bern (1974) or by Spence and Helmreich (1978) may further account for the level of social support and amount of well-being that men and women obtain through their same-sex and cross-sex bonds. For Bernard's theory, a measure of sex-role attitudes may clarify the effect that women's attitudes have on the level of support and well-being that women derive through their same-sex and cross-sex bonds. Such a study would assess whether women who espouse more positive attitudes towards other women benefit more strongly in terms of social support and well-being, than women who espouse more negative attitudes towards other women. Such a study would test Bernard's (1976) and Caplan's (1981) theories on the effect that the devaluation of women has on women's same-sex and cross-sex relationships. Similarly, the results of the study also suggest the relevance of the variable sex-role attitudes to future investigations of Lipman-Blumen's (1976) theory. Lipman-Blumen's theory may have greater relevance and applicability to Same-Sex Bonds and Well-Being p. 145 men who are power-oriented in their sex-role identity or to men who occupy a male dominated work context. In formulating such a study researchers should consider extending Weiss's dimensions of social relationships to include more instrumentally-oriented variables than were used in this study, such as status, power/dominance, utility value, and intellectual stimulation. Such a study would further enhance an understanding of the specific dimensions of male homosociality, its limitations as well as its applicability. With respect to cross-sex bonding, the results of the study suggest that cross-sex relationships have been an underestimated component of well-being to adults during married life. The results of this study also suggest that different aspects of same-sex and cross-sex bonds may contribute to women's and men's global well-being. Future researchers might consider examining the specific dimensions of same-sex and cross-sex social support that enhance well-being. With respect to the relationship between social support and well-being, the results of the study suggest that there may be both positive and negative aspects to intimacy. Future researchers might consider assessing the positive as well as the stressful aspects of intimacy to well-being. Such a study would serve to clarify the theoretical boundaries of social support as well as its application as a mental health enhancing concept. Same-Sex Bonds and Well-Being p. 146 E. Conclusions In conclusion, the results of this study suggest that the factors influencing the attainment of social support differ from the factors that influence the overall relationship between social support and global well-being. For the women, the results of this study draw attention to the relevance of relationship gender composition as an important variable influencing the attainment of social support. Women's same-sex bonds provide higher levels of social support, overall, than do women's cross-sex bonds. In addition, women's same-sex bonds provide higher levels of the attachment, social integration, reassurance of worth, help and guidance and opportunity for nurturance provisions than do women's cross-sex bonds. As well, the women in this study also report receiving supplies of these relational provisions in addition to the reliable alliance provision from more people within their same-sex networks than from people within their cross-sex networks. For the men, the results of this study indicate that relationship gender composition does not influence the attainment of social support as strongly as it does for women. Overall, men's same-sex bonds appear to provide higher levels of the social integration and reliable alliance provisions than do their cross-sex relationships. As well, the men in this study report receiving only three support dimensions (social integration, reliable alliance and help and guidance) from more people within their same-sex networks than from people within their cross-sex networks. In all, the social support data indicate fairly strong support for Bernard's (1976) theory of female homosociality and partial support for Lipman-Same-Sex Bonds and Well-Being p. 147 Blumen's theory of male homosociality and Weiss's (1974) theory of social relationships. With respect to the overall relationship between social support and global well-being, the results of this study indicate strong support for the importance of socially supportive relationships as a mental health enhancing resource during the course of everyday life, supporting Weiss's (1974) theoretical formulations. This finding is particularly noteworthy because while previous studies have shown that social support buffers stress, the results of this study demonstrate that social support is also strongly related to global well-being. However, the data do not appear to indicate the presence of sex differences in the overall relationship between social support and global well-being, again supporting propositions that seem to arise from Weiss's (1974) theory, and failing to support predictions derived from Bernard's 1976) and Miller's (1976) theories. The results of the study do suggest sex differences in the network characteristics predicting global well-being. Although the data do not indicate differences in the overall relationship between same-sex bonds and cross-sex bonds with global well-being, the data do suggest that different aspects of same-sex and cross-sex bonds predict global well-being. These findings indicate mixed support for propositions that seem to arise from Weiss's (1974) theory and fail to support predictions derived from Bernard's (1976), Lipman-Blumen's (1976) and Safilios-Rothschild's (1981) theories. Same-Sex Bonds and Well-Being p. 148 Thus, the results of this study suggest two propositions: First, that the factors affecting the attainment of social support differ from the factors affecting the overall relationship between social support and global well-being; second, that these factors tend to operate differently within men's and women's relationships. It is hypothesized that the relationship between social support and global well-being may be more complex than previously shown and may be further affected by such variables as the devaluation of women and stresses associated with intimacy. Future research investigators should continue to identify the factors that enhance and the factors that limit the attainment of social support as well as the relationship between social support and well-being. Same-Sex Bonds and Well-Being p. 149 REFERENCES Adler, A. (1927). Understanding human nature. New York: Greenburg. Allan-Kee, D. (1980). The relationship of psychological androgyny to self-esteem and life satisfaction in career-oriented and home-oriented women. (Doctoral dissertation. University of Southern California, 1980). Dissertation Abstracts International, 41, 1479 A. Anderson, M.R. (1977). A study of the relationship between life satisfaction and self-concept, locus of control, satisfaction with primary relationships and work satisfaction (Doctoral dissertation, Michigan State University, (1977). Dissertation Abstracts International, 38, 2638 9A. (University Microfilms No. 77-25.214). Andrews, F . M . & Withey, S.B. (1976). Social indicators of well-being: America's perception of life quality. New York: Plenum Press. Baker, E.K. (1977). Relationships of retirement and satisfaction with life events to locus-of-control (Doctoral dissertation, University of Wisconsin-Madison, (1976). Dissertation Abstracts International, 37, 4748B. (University Microfilms No. 76-28, 900). Bandura, A. , Ross, D. & Ross, S.A. (1963). A comparative test of the status, envy, social power and secondary reinforcement theories of identificatory learning. Journal of Abnormal and Social Psychology, 67, 527-534. Bardwick, J .M. (1971). Psychology of women: A study of biocultural conflicts. New York: Harper & Row. Same-Sex Bonds and Well-Being p. 150 Barrera, M. (1981). Social support in the adjustment of pregnant adolescents: Assessment issues. In B.H. Gottlieb (Ed.), Social networks and social support (pp. 69-96). London: Sage Publications. Barrett, C . J . (1978). Effectiveness of widow's groups in facilitating change. Journal of Consulting and Clinical Psychology, 46, 20-31. Bell, R.R. (1981). Friendships of women and of men. Psychology of Women Quarterly, 5, 402-417. Bell, R.R. (1981). Worlds of friendship. London: Sage Publications. Bern, S.L. (1974). The measurement of psychological androgyny. Journal of Consulting and Clinical Psychology, 42, 155-162 Bernard J . (1976). Homosociality and female depression. Journal of Social Issues, 32, 213-238. Bernard, J . (1981). The female world. New York: Free Press. Bloom, B.L. (1979). Prevention of mental disorders: Recent advances in theory and practise. Community Mental Health Journal, 15, 179-191. Booth, A. (1972). Sex and social participation. American Sociological Review, 37, 183-192. Booth, A. , «5c Hess, E . (1974). Cross-sex friendships. Journal of Marriage and the Family, 36, 38-47. Bradburn, N.M. (1969). The structure of psychological well-being. Chicago, Aldine. Brandt, A.S. (1980). Relationship of locus of control, environmental constraint, length of time in the institution and twenty-one other variables to morale and life satisfaction in the institutionalized elderly Same-Sex Bonds and Well-Being p. 151 (Doctoral Dissertation Abstracts International, 40, 5802B (University Microfilms No. 80-12, 153). Broverman, I.K., Broverman, D.M. , Clarkson, F . E , Rosenkrantz, P.S. & Vogel, S.R. (1970). Sex-role stereotypes and clinical judgements of mental health. Journal Consulting and Clinical Psychology, 34(1), 1-7. Brown, G.W., Bhrolchain, M.N., & Harris, T. (1975). Social class and psychiatric disturbance among women in an urban population. Sociology, 9, 225-254. Brown, G.W. & Harris, T. (1978). Social origins of depression. London, Tavistock. Buss, A . H . & Plamin, R. (1975). A temperament theory of personality development. John Wiley & Sons, New York. Campbell, A. (1976). Subjective measures of well-being. American Psychologist, 31, 117-124. Candy, S.G., Troll, L . G . , Ac Levy, S.G. (1981). A developmental exploration of friendship functions in women. Psychology of Women Quarterly, 5_> *56-471. Caplan, G. (1974). Support systems and community mental health. New York: Behavioural Publications. Caplan, P.J. (1981). Between women: Lowering the barriers. Personal Library, Toronto. Carveth, B.W., & Gottlieb, B.H. (1979). The measurement of social support and its' relation to stress. Canadian Journal of Behavioural Science, 11, 179-187. Same-Sex Bonds and Well-Being p. 152 Chodorow, N. (1978). The reproduction of mothering; Psychoanalysis & the sociology of gender. Los Angeles, CA: University of California Press. Cobb, S. Social support as a moderator of life stress. Psychosomatic Medicine, 38(5), 300-314. Cohen, S. & Syme, L.S. (1985). Issues in the study and application of social support. In S. Cohen & L.S. Syme (Eds.), Social support and health (pp. 3-22). Academic Press: New York. Cohen, S. 6c Wills, T.A. (1985). Stress, social support and the buffering hypothesis. Psychological Bulletin, Jj5(a), 310 - 357. Coopersmith, S. (1967). The antecedents of self-esteem. San Francisco: Freeman. Cott, N. (1977). The bonds of womanhood. New Haven: Yale University Press. Crandall, R. (1973). The measurement of self-esteem and related constructs. In J.P. Robinson 6c P.R. Shaver (Eds.), Measures of social psychological  attitudes; Revised edition (pp. 45-168). Ann Arbor, Michigan: Institute for research. Crowne, D.P. 6c Marlowe, D. (1964). The approval motive. New York: Wiley. Cutrona, C . E . (1982). Transition to college: Loneliness and the process of social adjustment. In L . A . Peplau 6c D. Perlman (Eds.), Loneliness: A sourcebook of current theory, research and therapy (pp. 291-309). New York: Wiley-Interscience. Cutrona, C . E . 6c Russell, D.W. (1986). The provisions of social relationships and adaptation to stress. In W.H. Jones and D. Perlman (Eds.), Same-Sex Bonds and Well-Being p. 153 Perspectives on interpersonal behaviour and relationships. Greenwich, Conn.: JAI Press. Cutrona, C . E . , Russell, D. & Rose, 3. (1984). Psychosocial factors and physical and mental health among the elderly. Manuscript in preparation. Czaja, D. (1975). Life satisfaction and activity theory: A multidimensional approach. Journal of Gerontology, 35, 935-941. Davidson, S. (1983, March). The felt connection: Friendship between women. Paper presented at the meeting of the Association of Women in Psychology, Seattle, Washington. Davidson, S., & Packard, T. (1981). The therapeutic value of friendship between women. Psychology of Women Quarterly, 5, 495-510. Derogatis, L.R. , Lipman, R.S., Rickels, J . 6c Covi, L . (1974). The Hopkins Symptom Checklist: A measure of primary symptom dimensions. In P. Pichot (Ed.), Psychological measurements in psychopharmacology: Modern problems in pharmacopsychiatry. Basal, Switzerland: S. Krager. Dickens, W.J., & Perlman, D. (1981). Friendship over the life cycle. In S. Duck & R. Gilmour (Eds.), Personal relationships. Vol. 2: Developing personal relationships. London: Academic Press. Diener, E . (1984). Subjective well-being. Psychological Bulletin, 95(3), 542-575. Diener, E . , Emmons, R.J . , Griffin, S. (1984). The satisfaction with life scale. Personality Assessment, in press. Di Matteo, R .M. & Hays, R. (1981). Social Support and serious illness. In B.H. Gottlieb (Ed.), Social networks and social support (pp. 117-148). London: Sage Publications. Same-Sex Bonds and Weil-Being p. 154 Dooley, D. (1985). Causal inference in the study of social support. In S. Cohen & L.S. Syme (Eds.), Social support and health (pp. 109-124). Academic Press: New York. Eisenberg, D.M. (1981). Autonomy, health and life satisfaction among older persons in a life care community (Doctoral dissertation, Bryn Mawr College, 1980). Dissertation Abstracts International, 41, 3724A. (University Microfilms No. 81-03, 906). Ellison, C.W. & Paloutzian, R. (May, 1979). Developing an abbreviated loneliness scale. Paper presented at the UCLA Research conference on Loneliness. Los Angeles, California. Eysenck, H.J. , & Eysenck, S.B.G. (1964). Manual of the Eysenck personality inventory. San Diego, C A : Educational and Industrial Testing Service. Gibbs, M. , Auerbach, D., & Fox, M. (1980). A comparison of male and female same-sex friendships. International Journal of Women's Studies, 3, 261-272. Gilligan, C . (1977). In a different voice: Women's conceptions of self and morality. Harvard Educational Review, 47(4), 481-517. Goodchilds, J . (1979). Power: A matter of mechanics? SASP Newletter, 5(3), 3. Goodenow, C. (1985, August). Women's friendships and their association with  psychological well-being. Paper presented at the annual meeting of the American Psychological Association, Los Angeles, California. Gorman, B.S. (1972). A multivariate study of the relationships of cognitive control and cognitive style principles to reported daily mood experiences. Same-Sex Bonds and Well-Being p. 155 (doctoral dissertation, City University of New York, 1971). Dissertation Abstracts International , 32, 421 IB. (University Microfilms, No. 72-5071). Gottlieb, B.H. (Ed.). (1981). Social networks and social support. Beverly Hills: Sage. Gottlieb, B.H. (1983). Social support as a focus for integrative research in psychology. American Psychologist, 38, 278-289. Gump, 3.P. (1972). Sex-role attitudes and psychological well-being. Journal of Social Issues, 28(2) 79-92. Harris, R.J . (1975). A primer of multivariate statistics. New York: Academic Press. Hatfield, E . (1984). The dangers of intimacy. In V.J . Derlega (Ed.), Communication, intimacy and close relationships, pp. 207-220. Academic Press, Inc. Henderson, P. (1986). Social relationships of lesbian and heterosexual feminist women. Unpublished master's thesis, University of British Columbia, Vancouver. Henderson, S. (1982). The significance of social relationships in the etiology of neurosis. In C M . Parkes <5c J . Stevenson-Hinde (Eds.), The place of attachment in adult life (pp. 200-230). London: Tavistock Publications. Hirsch, B.J. (1980). Natural support & systems and coping with major life change. American Journal of Community Psychology, 8, 59-172. Hirsch, B.J. (1981). Social networks and the coping process: Creating personal communities. In B.H. Gottlieb (Eds.), Social networks and social support (pp. 149-170). London: Sage Publications. Same-Sex Bonds and Well-Being p. 156 Hjelle, L . A . & Butterfield, R.A. (1974). Self-actualization and woman's attitudes toward their roles in contemporary society. The Journal of  Psychology, 87, 225-230. House, J.S. (1981). Work stress and social support. Reading, Mass: Addison-Wesley. Jahoda, M. (1958). Current concepts of positive mental health. New York: Basic Books. Johnson, P. (1976). Women and power: Toward a theory of effectiveness. Journal of Social Issues, 32, 99-110. Jones, W.H. (August, 1978). The persistence of loneliness. In D. Perlman (Chair). Toward a psychology of loneliness. Symposium presented at the annual meeting of the American Psychological Association, Toronto. Jones, W.H., Freeman, J .A. & Goswick, R.A. (1981). The persistence of loneliness: Self and other determinants. Journal of Personality, 49, 27-48. Kahn, R.L. & Antonucci, T. (1980). Conveys over time life cycle: Attachment roles and social support. In P.B. Baltes and C. Brim (Eds.). Lifespan development and behaviour, 3_, Boston, Lexington. Leavy, R.L. (1983). Social support and psychological disorder: A review. Journal of Community Psychology, 11, 3-21. Lewis, R.A. (1978). Emotional intimacy among men. Journal of social issues, 34, 108-121 Lipman-Blumen, J . (1975). Toward a homosocial theory of sex roles: An explanation of the sex segregation of social institutions. In M. Blaxall & B. Reagan (Eds.), Women and the work place: The implications of Same-Sex Bonds and Well-Being p. 157 occupational segregation (pp.15-31). Chicago: University of Chicago Press. Lips, H.M. (1981). Women, men, and the psychology of power. Englewood Cliffs N.J.: Prentice Hall. Logan, D.D. & Kaschak, E. (1980). The relationship of sex, sex role, and mental health. Psychology of Women Quarterly, 4(4), 573-580. McClelland, D.C. (1975). Power: The inner experience, New York: Irvington. Merton, R.K. (1949). Social theory and social structure. Glencoe, 111.: Free Press. Miller, J.B. (1976). Toward a new psychology of women. Boston: Beacon Press. Miller, P. <5c Ingham, J .G. , (1976). Friends, confidents and symptoms. Social Psychiatry, 11, 51-58. Paulhus, D. (1983). Sphere-specific measures of perceived control. Journal of Personality and Social Psychology, 44(6), 1253-1265. Paulhus, D.L. <3c Christie, R. (1981). Spheres of control: An interactionist approach to assessment of perceived control. In H.M. Lefcourt (Ed.), Research with the locus of control construct, New York: Adademic Press. Paulhus, D.L. Molin, J . Ac Schuchts, R. (1979). Control profiles of football players, tennis players and nonathletes. Journal of Social Psychology, 108, 199-205. Paykel, E.S., Emms, E .M. , Fletcher, J . , & Rossaby, E.S. (1980) Life events and social support in puerperal depression. British Journal of Psychiatry, 136, 339-346. Same-Sex Bonds and Well-Being p. 158 Pearlin, L . (1975). Sex roles and depression. In N. Datan & L. Ginsberg (Eds.), Life-span developmental psychology: Normative life crises, pp. 191-206. New York: Academic Press. Peplau, L . , Sc Perlman, D. (Eds.). (1982). Loneliness: A source book of current theory research and therapy. New York: John Wiley. Pleck, J .H. (1976). The male sex role: Definitions, problems and sources of change. Journal of Social Issues, 32(3), 155-165. Radloff, L . (1975). Sex differences in depression: The effects of occupation and marital status. Sex Roles, J_(3), 249-266. Rawlings, E.I., Sc Carter, D.K. (1977). Values and value change in psychotherapy. In E.I. Rawlings Sc D.K. Carter (Eds.), Psychotherapy for women. Springfield, 111.: Charles C . Thomas. Reis, H.T. (1984). Social interaction and well-being. In S. Duck (Ed.), Personal relationships. 5: Repairing Personal Relationships (pp. 21-45). New York: Academic Press. Reisman, J .M. (1981). Adult friendships. In S. Duck Sc R. Gilmour (Eds.), Personal relationships. 2: Developing personal relationships (pp. 205-230). New York: Academic Press. Rose, S.M. (1985). Same and cross-sex friendship and the psychology of homosociality. Sex Roles, 12 (1), 63-74. Rosenberg, M. (1979). Conceiving the self. New York: Basic Books. Russell, D. Altmaier, E . Sc Van Velzen, D. (1984). Job-related stress, social support and burn-out among teachers. Manuscript submitted for publication. Same-Sex Bonds and Well-Being p. 159 Russell, D., Cutrona, C . (1984, August). The provisions of social relationships and adaptation to stress. Paper presented at the American Psychological Association convention in Toronto, Ontario, Canada. Russell, D., Cutrona, C . E . , Rose, J . , Yurko, K. (1984). Social and emotional loneliness: An examination of Weiss' typology of loneliness. Journal of Personality and Social Psychology, 46(6), 1313-1321. Russell, D., Peplau, L . A . , & Cutrona, C . (1980). The revised UCLA Loneliness Scale: Concurrent and discriminant validity evidence. Journal of Personality and Social Psychology, 39(3), 472-480. Safilios-Rothschild, C . (1981). Toward a social psychology of relationships. Psychology of Women Quarterly, 5(3), 377-384. Sandler, I.N. <5c Barrera, M. (1984). Toward a multimethod approach to assessing the effects of social support. American Journal of Community Psychology, _12(D, 37-51. Seiden, A . M . & Bart, P.B. (1975). Women to women: Is sisterhood powerful? In N. Glazer-Malbin (Ed.), Old family/New family (pp. 189-228). New York: Van Nostrand. Seligman, M.E. (1974). Depression and learned helplessness. In R.J. Friedman, & M.M. Katz (Eds.), The psychology of depression: Contemporary theory and research (pp. 15-30). Washington, D .C. Serafica, F . C . (1982). Conceptions of friendship and interaction between friends: An organismic-developmental perspective. In F. Serafica (Ed.), Social studies development in context (pp. 100-132). New York: Guildford Press. Same-Sex Bonds and Weil-Being p. 160 Shaver, P. 4c Buhrmester, D. (1983). Loneliness, sex-role orientation and group life: A social needs perspective. In P.B. Paulhus, (Ed.), Basic group processes (pp. 259-288). New York: Springer-Verlag. Shipley, T. & Veroff, J . (1952). A projective measure of the need for affiliation. Journal of Experimental Psychology, 43, 349-56. Shulman, N. (1975). Life cycle variations in patterns of close friendship. Journal of Marriage and the Family, 37, 813-821. Silber E . & Tippitt. J.S. (1965). Self-esteem clinical assessment and measurement validation. Psychological Reports, 16, 1017-1071. Sisenwein, R.J . (1982). Loneliness in the individual as veiwed by himself and others. Unpublished doctoral dissertation. Columbia University. Swenson, E.V. & Ragucci, R. (1984). Effects of sex-role stereotypes and androgynous alternatives on mental health judgements of psychotherapists. Psychological Reports, 54, 475-481. Smith-Rosenberg. C . (1975) The female world of love and ritual: Relations between women in nineteenth century America. Journal of Women in Culture and Society, 1,(11), 1-29. Solano, C .H. (1980). Two measures of loneliness: A comparison. Psychological Reports, 46, 23-28. Spence, J.T. & Helmreich, R.L. (1978). Masculinity and feminity: Their psychological dimensions, correlates and antecedents. Austin: University of Texas Press. Steiger, J .H. & Browne, M.W. (1984). The comparison of independent correlations between optimal linear composites. Psychometrika, 49, 11-24. Same-Sex Bonds and Well-Being p. 161 Strupp, H.H. & Hadley, S.W. (1977). A tripartite model of mental health and therapeutic outcomes. American Psychologist, 187-196. Tiger, L . (1969). Men in groups. New York: Vintage Books. Tolor, A. (1978). Personality correlates of the joy of life. Journal of Clinical Psychology, 34> 671-676. Tolsdorf, C C . (1976). Social networks, support and coping: An exploratory study. Family Process, 15, 407-417. Weiss, R.S. (1973). The fund of sociability. In H. Lopata (Ed.), Marriages and families (pp. 197-202). New York: Van Nostrand. Weiss, R.S. (Ed.). (1973). Loneliness: The experience of emotional and social isolation. Cambridge, Mass.: M.I.T. Press. Weiss, R.S., (1968). Materials for a theory of social relationships. In W.G. Bennis, E .H. Schain, F . E . Steele, & D.E. Berlew, (Eds.), Interpersonal dynamics (pp. 154 - 163). Homewood, Illinois: Dorsey, 1968. Weiss, R.S. (1974). The provisions of social relationships. In Z. Rubin (Ed.), Doing unto others (pp. 17-26). Englewood Cliffs, N.J.: Prentice hall. Weiss, R. (1982). Attachment in adult life. In C M . Parkes <5c J . Stevenson-Hinde (Eds.), The place of attachment in human behaviour (pp. 171-184). New York: Tavistock. Weiss, L . & Lowenthal, M.F. (1975). Life course perspectives on friendship. In M.F. Lowenthal, M. Thurnher <5c D. Chiriboga (Eds.), Four stages of life (pp. 48-61). San Francisco: Jossey-Bass. Wheeler, L . , Reis, H. , Nezlek, J . (1984). Loneliness, social interaction and sex roles. Journal of Personality and Social Psychology, 45(4), 943-953. Same-Sex Bonds and Well-Being p. 162 Wilcox, B.L. (1981). Social support, life stress and psychological adjustment: A test of the buffering hypothesis. American Journal of Community  Psychology, 9, 371-386. Winter, D.G. (1973). The power motive. New York: Free Press. Wilson, W. (1967). Correlates of avowed happiness. Psychological Bulletin, 67, 294-306. Woolsey, L .K. (1977). Psychology and the reconciliation of women's double-bind: To be feminine or to be fully human. Canadian Psychological  Review, JjJ, 66-78. Woolsey, L . K . (1983 June). Bonds between women: A theoretical review. Paper presented at the Canadian Psychological Association, Winnipeg, Manitoba. Wright, P.H. (1982). Men's friendships, women's friendships and the alleged inferiority of the latter. Sex Roles, 8, 1-20. Yoon, G.H. (1978). The natural history of friendship: Sex differences in best-friendship patterns. Dissertation Abstracts, 39, (3-B), 1553. Same-Sex Bonds and Well-Being p. 163 A P P E N D I C E S Same-Sex Bonds and Well-Being p. 164 A P P E N D I X A Supplementary Statistical Analyses Table 12 Differences in the Correlations between Women's Same-Sex and Cross-Sex Bonds with Indices of Well-Being Index of Well-Being Same-Sex Cross-Sex z £ Interpersonal Control £ .42 .40 .170 > .05 Personal Efficacy r_ .06 -.07 -.921 > .05 Self-Esteem £ .40 .37 .212 > .05 Life Satisfaction r .18 .22 .297 > .05 Depression £ -.33 -.29 -.283 > .05 Loneliness r -.49 -.56 -.730 > .05 Note: Magnitude of r_ necessary for significance at .05 = .18. Magnitude of £ necessary for significance at .01 = .33. Same-Sex Bonds and Well-Being p. 165 Table 13 Differences in the Correlations between Men's Same-Sex and Cross-Sex Bonds with Indices of Well-Being Index of Well-Being Same-Sex Cross-Sex z £ Interpersonal Control r_ .36 .33 .212 > .05 Personal Efficacy _r .04 .19 1.07 > .05 Self-Esteem j_ .35 .28 .936 > .05 Life Satisfaction j_ .25 .17 .590 > .05 Depression j_ -.15 -.14 -.070 > .05 Loneliness jr -.46 -.49 -.276 > .05 Note; Magnitude of r_ necessary for significance at .05 = .17. Magnitude of j_ necessary for significance at .01 = .25. Same-Sex Bonds and Well-Being p. 166 Table 14 Global Well-Being Regressed onto the Individual Relational Provisions: Men Step Variable Entered R 2 F_ £ 1 cross-sex reassurance of worth .30 42.32 <.001 2 cross-sex opportunity for .35 26.28 <.001 nurturance 3 same-sex social integration .39 20.70 <.001 Table 15 Global Well-Being Regressed onto the Individual Relational Provisions: Women Step Variable Entered II 2 F £ 1 same-sex reassurance of worth .20 24.84 <.001 2 cross-sex help and guidance .28 18.99 <.001 Same-Sex Bonds and Well-Being p. 167 Table 16 Social Provisions Scale: Intercorrelation Matrix for Men's and Women's Same-Sex Bonds Att Socint Rw Ra Hg On Att — .50 .37 .43 .48 .48 Socint .60 .50 .52 .53 .41 Rw .60 .65 .39 .47 .39 Ra .63 .55 .61 — .69 .25 Hg .68 .59 .61 .73 .32 On .40 .35 .40 .26 .27 — Note: Intercorrelation matrix for the men is above the diagonal Intercorrelation matrix for the women is below the diagonal Same-Sex Bonds and Well-Being p. 168 Table 17 Social Provisions Scale: Intercorrelation Matr ix for Women's and Men's Cross-Sex Bonds A t t Socint Rw Ra Hg On A t t .59 .44 .67 .74 .43 Socint .64 .53 .53 .54 .35 Rw .40 .44 .53 .50 .21 Ra .50 .53 .30 — .73 .29 Hg .61 .53 .36 .60 .30 On .38 .40 .22 .27 .34 — Note: Intercorrelation matrix for the women is above the diagonal Intercorrelation matr ix for the men is below the diagonal Same-Sex Bonds and Well-Being p. 169 Table 18 Full Regression Analyses: Comparison of Dependent Multiple Rs Gender Composition £(D £ Sex Same-Sex Cross-Sex Male Multiple R .5507 .6012 .4889 >.4844 Female Multiple R .4932 .4801 .0230 >.8794 Table 19 Full Regression Analyses: Comparison of Independent Multiple Rs Sex z £ Gender Composition Female Male Same-Sex Multiple R .4932 .5507 .581 > .28 Cross-Sex MuitipleR .4801 .6012 1.20 > .11 Same-Sex Bonds and Well-Being p. 171 Background Information Please circle or fill in your response on the line provided. (Omit items you feel uncomfortable with) 1. Sex: M F 2. Age: 3. Marital Status: currently married or living with someone formerly married never married 4. Number of Children: 5. What are the age(s) of your children: 11. What ethnic group do you identify with? 6. Are your children living with you? Full Time Part Time 7. What is your religious affiliation? 8. What is your political affiliation? 9. Are you employed? Full Time Part Time 10. What is your job/profession? 12. To what extent do you feel your ethnic origin affects you? (Please indicate by circling one of the numbers below.) 7 6 5 4 3 2 1 A Great Deal Not At All 13. What is your student status? Graduate Undergraduate Same-Sex Bonds and Well-Being p. 172 14. What year of university studies are you currently in? 1 2 3 4 5 6 7 8 9 15. What programme are you enrolled in at UBC? 16. How long have you lived in the Vancouver area? just moved to Vancouver temporarily for spring/summer session less that one year 1- 2 years 2- 5 years 5 + years 17. If you are just here for spring/summer session, do you live in a community of less than 10,000 people? Yes No Same-Sex Bonds and Well-Being p. 173 Please list the first name and last initial of all the women (not including your spouse or mate, if applicable) who are IMPORTANT in your life at this moment, whether you like them or not. These women may be family members, relatives, friends, neighbours, co-workers, supervisors, community members, and so on. Use your own definition of who is important and use the following to guide yourself. After listing each woman, indicate how long you have known her; her age; relationship to you; and how often you have contact with her (face to face, by phone or by letter). Indicate your frequency of contact by filling in the appropriate number. 5 = usually daily 4 = usually at least once a week 3 = usually at least once a month 2 = usually at least once every 6 months 1 = usually at least once a year Name and Last Initial Years Known Age Relationship Frequency of Contact example Janet F. Edith W. Laura K. Sara D. 3 30 3 I 33 58 45 1 friend mother supervisor niece 5 1 4 3 Same-Sex Bonds and Well-Being p. 174 Now, thinking about your relationships with the list of women you have outlined, please indicate the first name and last initial of the woman or women you get the following things from and add up the total you have listed for each category. Example First name and Total last initial // Which woman (women) do you get a sense of emotional security and well-being from when your are with her (them)? With which woman (women) do you share similar attitudes and beliefs? Who gives you the feeling that your competencies and skills are recognized? Whom do you feel you could count on in an emergency? Whom could you talk to about important decisions in your life? Whom do you feel a sense of responsibility for in terms of their well-being? Same-Sex Bonds and Well-Being p. 175 Social Provisions Scale In answering the next set of questions asked, please think about your current relationships with women friends, family members (not including your spouse), co-workers, community members, and so on. Please tell me to what extent you agree that each statement describes your current relationships with women. Use the scale 1 to 4 (see below) to give me your opinion. So, for example, if you feel a statement is very true of your current relationships with women, you would circle #4 "strongly agree". If you feel a statement clearly does not describe your relationships with women, you would circle / / l "strongly disagree". 1 = Strongly Disagree; 2 = Disagree; 3 = Agree; 4 = Strongly Agree. Disagree Agree 1. There are women I can depend on to help me if I really need it. 1 2 3 4 2. I feel that I do not have close personal relationships with women. 1 2 3 4 3. There is no woman I can turn to for guidance in times of stress. 1 2 3 4 4. There are women who depend on me for help. 1 2 3 4 5. There are women who enjoy the same social activities I do. 1 2 3 4 6. Women do not view me as competent. 1 2 3 4 7. I feel personally responsible for the well-being of a woman. 1 2 3 4 8. I feel part of a group of women who share my attitudes and beliefs. 1 2 3 4 9. I do not think that women respect my skills and abilities. 1 2 3 4 10. If something went wrong, there is no woman who would come to my assistance. 1 2 3 4 Same-Sex Bonds and Well-Being p. 176 11. I have close personal relationships with women that provide me with a sense of emotional security and well-being. 12. There are women I talk to about important decisions in my life. 13. I have relationships with women where my competence and skill are recognized. 14. There are no women who share my interests and concerns. 15. There is no woman who relies on me for her well-being. 16. There are trustworthy women in my life I could turn to for advice if I were having problems. 17. I feel a strong emotional bond with at least one other woman. IS. There is no woman I can depend on for aid if I really need it. 19. There is no woman I feel comfortable talking about my problems with. 20. There are women who admire my talents and abilities. 21. There is no woman who likes to do the things I do. 22. I lack a feeling of intimacy with another woman. 23. There are women I can count on in an emergency. 24. No woman needs me to care for her. Disagree Agree 2 3 4 2 3 4 2 3 4 2 3 4 2 3 4 2 3 4 2 3 4 2 3 4 2 3 4 2 3 4 2 3 4 2 3 4 2 3 4 2 3 4 Same-Sex Bonds and Well-Being p. 177 Social Provisions Scale Scoring Key A score for each provision is derived such that a high score indicates the individual is receiving that provision. Items that are asterisked should be reversed before scoring (i.e., 4=1, 3=2, 2=3, 1=4). 1. Attachment: 2*, 11, 17, 22* 2. Social Integration: 5, 8, 14*, 21* 3. Reassurance of Worth: 6*, 9*, 13, 20 4. Reliable Alliance: 1, 10*, 18*, 23 5. Help and Guidance: 3*, 12, 16, 19* 6. Opportunity for Nurturance: 4, 7, 15*, 24* Same-Sex Bonds and Well-Being p. 178 Please list the first name and last initial of all the men (not including your spouse or mate, if applicable) who are IMPORTANT in your life at this moment, whether you like them or not. These men may be family members, relatives, friends, neighbours, co-workers, supervisors, community members, and so on. Use your own definition of who is important and use the following to guide yourself. After listing each man, indicate how long you have known him; his age; relationship to you; and how often you have contact with him (face to face, by phone or by letter). Indicate your frequency of contact by filling in the appropriate number. 5 = usually daily 4 = usually at least once a week 3 = usually at least once a month 2 = usually at least once every 6 months 1 = usually at least once a year Name and Last Initial Years Known Age Relationship Frequency of Contact Gordon W. 30 60 father 5 . Paul D. 5 35 cousin 3 example r Mac. D. 4 30 friend 2 Edwin G. 1 45 supervisor 4 Same-Sex Bonds and Well-Being p. 179 Now, thinking about your relationships with the list of men you have outlined, please indicate the first name and last initial of the man or men you get the following things from and add up the total you have listed for each category. Example First name and Total last initial # Whom could you talk to about important decisions Mac. D in your life? Gordon W. 3 Edwin G. Which man (men) do you get a sense of emotional security and well-being from when your are with him (them)? With which man (men) do you share similar attitudes and beliefs? Who gives you the feeling that your competencies and skills are recognized? Whom do you feel you could count on in an emergency? Whom could you talk to about important decisions in your life? Whom do you feel a sense of responsibility for in terms of their well-being? Same-Sex Bonds and Well-Being p. 180 Social Provisions Scale In answering the next set of questions, please think about your current relationships with men friends, family members (not including your spouse), co-workers, community members, and so on. Please tell me to what extent you agree that each statement describes your current relationships with men. Use the scale 1 to 4 (see below) to give me your opinion. So, for example, if you feel a statement is very true of your current relationships with men, you would circle //4 "strongly agree". If you feel a statement clearly does not describe your relationships with men, you would circle #1 "strongly disagree". 1 = Strongly Disagree; 2 = Disagree; 3 = Agree; 4 = Strongly Agree. Disagree Agree 1. There are men I can depend on to help me if I really need it. 1 2 3 4 2. I feel that I do not have close personal relationships with men. 1 2 3 4 3. There is no man I can turn to for guidance in times of stress. 1 2 3 4 4. There are men who depend on me for help. 1 2 3 4 5. There are men who enjoy the same social activities I do. 1 2 3 4 6. Men do not view me as competent. 1 2 3 4 7. I feel personally responsible for the well-being of man. 1 2 3 4 8. I feel part of a group of men who share my attitudes and beliefs. 1 2 3 4 9. I do not think that men respect my skills and abilities. 1 2 3 4 10. If something went wrong, there is no man who would come to my assistance. 1 2 3 4 Same-Sex Bonds and Well-Being p. 181 11. I have close personal relationships with men that provide me with a sense of emotional security and well-being. 12. There are men I talk to about important decisions in my life. 13. I have relationships with men where my competence and skill are recognized. 14. There are no men who share my interests and concerns. 15. There is no man who relies on me for his well-being. 16. There are trustworthy men in my life I could turn to for advice if I were having problems. 17. I feel a strong emotional bond with at least one other man. 18. There is no man I can depend on for aid if I really need it. 19. There is no man I feel comfortable talking about my problems with. 20. There are men who admire my talents and abilities. 21. There is no man who likes to do the things I do. 22. I lack a feeling of intimacy with man. 23. There are men I can count on in an emergency. 24. No man needs me to care for him. D i s a g r e e A g r e e 2 3 4 2 3 4 2 3 4 2 3 4 2 3 4 2 3 4 2 3 4 2 3 4 2 3 4 2 3 4 2 3 4 2 3 4 2 3 4 2 3 4 Same-Sex Bonds and Well-Being p. 182 Social Provisions Scale Scoring Key A score for each provision is derived such that a high score indicates the individual is receiving that provision. Items that are asterisked should be reversed before scoring (i.e., 4=1, 3=2, 2=3, 1=4). 1. Attachment; 2*, 11, 17, 22* 2. Social Integration: 5, 8, 14*, 21* 3. Reassurance of Worth: 6*, 9*, 13, 20 4. Reliable Alliance: 1, 10*, 18*, 23 5. Help and Guidance: 3*, 12, 16, 19* 6. Opportunity for Nurturance: 4, 7, 15*, 24* Same-Sex Bonds and Well-Being p. 183 Rosenberg Self-Esteem Scale Please read the following list of sentences and decide whether you agree or disagree with them. Use the scale from 1 to 4 (see below) to describe how you feel about each sentence and indicate by circling the number beside the statement which describes how you feel. 1 = Strongly Disagree; 2 = Disagree; 3 = Agree; 4 = Strongly Agree. Disagree Agree 1. On the whole I am satisfied with myself. 1 2 3 4 2. At times I think I am no good at all. 1 2 3 4 3. I feel I have a number of good qualities. 1 2 3 4 4. I am able to do things as well as most other people. 1 2 3 4 5. I feel I do not have much to be proud of. 1 2 3 4 6. I certainly feel useless at times. 1 2 3 4 7. I feel that I am a person of worth at least on an equal plane with others. 1 2 3 4 8. I wish I could have more respect for myself. 1 2 3 4 9. All in all, I am inclined to feel I am a failure. 1 2 3 4 10. I take a positive attitude towards myself. 1 2 3 4 Scoring Key Reverse items that are asterisked before scoring (i.e., 4=1, 3=2, 2=3, 3=4). Then add the responses to the ten items. Same-Sex Bonds and Well-Being p. 184 UCLA Loneliness Scale Directions: Indicate how often you feel the way described in each of the following statements. Circle one number for each. 1 = Never; 2 = Rarely; 3 = Sometimes; 4 = Often. Never Often 1. I feel in tune with the people around me. 1 2 3 2. No one really knows me well. 3 2 3. I can find companionship when I want it. 1 2 3 4. People are around me but not with me. 3 2 Scoring Sum all four items. Same-Sex Bonds and Well-Being p. 185 HAKSTIAN-McCLEAN DEPRESSION SCALE How relaxed have you been in the last two days compared to how you normally are? (please circle the appropriate number) 10 9 8 7 6 5 4 3 2 1 Extremely Calm and Tense Relaxed How satisfied are you with your ability to perform household duties? 10 9 8 7 6 5 4 3 2 1 Very " Very Dissatisfied Satisfied To what extent have you had difficulty starting and following through an ordinary job or task to completion during the last week compared to when you feel things have been going well? 10 9 8 7 6 5 4 3 2 1 Putting things off Start and finish jobs Starting and not finishing as well as most other for a long time if at all people How many times in the last 2 days have you been preoccupied by thoughts of hopelessness, helplessness, pessimism, intense worry, unhappiness etc. Please put a check ( ) on one of the lines below. 1. Not at all. 2. Rarely 3. Frequently 4. Most of the time 5. All the time Same-Sex Bonds and Well-Being p. 186 Hakstian-McClean Depression Scale Scoring Key Sum items 1, 2 and 3. Multiply the response indicated on the fourth item by 4. Add this to the previous sum. Same-Sex Bonds and Well-Being p. 187 THE SATISFACTION WITH LIFE SCALE Below are five statements with which you may agree or disagree. Using the 1-7 scale below, indicate your agreement with each item by circling the appropriate number on the line following that item. 7 = Strongly Agree; 6 = Agree Very Much; 5 = Agree Slightly; 4 = Neutral; 3 = Slightly Disagree; 2 = Disagree Very Much; 1 = Strongly Disagree. Strongly Strongly Agree Neutral Disagree 1. In most ways my life is close to my ideal. 7 6 5 4 3 2 1 2. The conditions of my life are excellent. 7 6 5 4 3 2 1 3. I am satisfied with my life. , 7 6 5 4 3 2 1 4. So far I have gotten the important things I want in life. 7 6 5 4 3 2 1 5. If I could live my life over, I would change almost nothing. 7 6 5 4 3 2 1 Scoring Key Sum the responses to the five items r Same-Sex Bonds and Well-Being p. 188 SPHERES OF CONTROL SCALE Below are 20 statements with which you may agree or disagree. Using the 1-7 scale below, indicate your agreement with each item by circling the appropriate number on the line following that item. Please be open and honest in your responding. 7 = Strongly Agree; 6 = Agree Very Much; 5 = Agree Slightly; 4 = Neutral; 3 = Slightly Disagree; 2 = Disagree Very Much; 1 = Strongly Disagree. Strongly Strongly Agree Disagree Personal Efficacy 1. When I get what I want it's usually because I worked hard for it. 7 6 5 4 3 2 2. When I make plans I am almost certain to make them work. 7 6 5 4 3 2 3. I prefer games involving some luck over games requiring pure skill. 7 6 5 4 3 2 4. I can learn almost anything if I set my mind to it. 7 6 5 4 3~~ 2 5. My accomplishments are entirely due to my hard work and ability. 7 6 5 4 3 2 6. I usually don't set goals because I have a hard time following through on them. 7 6 5 4 3 2 7. Competition discourages excellence. 8. Often people get ahead just by being lucky. 7 6 5 4 3 2 7 6 5 4 3 2 9. On any sort of exam or competition I like to know how well I do relative to everyone else. 7 6 5 4 3 2 10. It's pointless to keep on working on something that's too difficult for me. 7 6 5 4 3 2 Same-Sex Bonds and Well-Being p. 189 Strongly Strongly Agree Disagree Interpersonal Control 11. Even when I'm feeling self-confident about most things, I still seem to lack the ability to control social situations. 12. I have no trouble making and keeping friends. 7 6 5 4 3 2 13. I'm not good at guiding the course of a conversation with several others. 7 6 5 4 3 2 14. I can usually establish a close personal relationship with someone I find attractive. 7 6 5 4 3 2 15. When being interviewed I can usually steer the interviewer toward the topics I want to talk about and away from those I wish to avoid. 7 6 5 4 3 2 16. If I need help in carrying off a plan of mine, it's usually difficult to get others to help. 7 6 5 4 3 2 17. If there is someone I want to meet I can usually arrange it. ' 7 6 5 4 3 2 18. I often find it hard to get my point of view across to others. 7 6 5 4 3 2 19. In attempting to smooth over a disagreement, I usually make it worse. 7 6 5 4 3 2 20. I find it easy to play an important part in most group situations. 7 6 5 4 3 2 Spheres of Control Scale Scoring Key Items that are asterisked should be reversed before scoring (i.e., 7=1, 6=2, 5=3, 4=4, 3=5, 2=6, 1=7). Responses for each item are then added together. Three scores are derived for each respondent, a personal efficacy score, an interpersonal control score and a total score combining the scores from the two scales. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0056024/manifest

Comment

Related Items