RELATIONSHIPS OF ATTACHMENT STATUS AND GENDER TO PERSONAL MEANING, DEPRESSIVENESS AND TRAIT ANXIETY AMONG COLLEGE STUDENTS By IBRAHIM KEKLIK A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFIL LMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2004
Copyright 2004 By IBRAHIM KEKLIK
This dissertation is dedicated to the loving memory of my mother, Fateâ€™ Alsoreâ€™, 1931 April 16, 2002.
iv ACKNOWLEDGMENTS There have been many persons whose empowerment gave me the strength and inspiration while working on this project. Without their presence and support such a challenging process would not turn into a rewarding one. I would like to first an d foremost thank Dr. James Archer, whose work ethic, compassion, support, understanding, and guidance inspired me through this project. Second, I would like to thank my dearest friends, Burak Taskonak, Deniz Rende, and Kate Gallagher whose presence, love and support was no less than that of a close knit family. I owe special thanks to my mother, Fateâ€™ Alsoreâ€™, who left us as this project was ongoing, and who by her brilliance, vision and courage set a miraculous example for my personal and professional g rowth. I thank my father, Uram i Useâ€™, whose hard work, generosity, humility, and simplicity taught me resilience. I am forever indebted to my colleagues and friends Drs. Michael Murphy, Cassie Hessler Smith and Nancy Coleman, who exemplified divinity i n humanness, and whose conversations gave me tools for many rites of passages of life. I owe special thanks to my dearest friend Ana Ghoreishian who taught me the mystical nature of human existence. I thank my lovely friends and co workers at Meridian Be havioral Healthcare CSU whose support and encouragement showed me the universality of friendship. Finally, I am thankful to my committee members Drs. Dave Miller, Harry Daniels, Peter Sherrard, and Silvia Doan for their great support, guidance and scholars hip.
v TABLE OF CONTENTS Page ACKNOWLEDGMENTS ................................ ................................ ................................ .. iv LIST OF TABLES ................................ ................................ ................................ ........... viii ABSTRACT ................................ ................................ ................................ ........................ x CHAPTER 1 INTRODUCTION ................................ ................................ ................................ ...... 1 Scope of the Problem ................................ ................................ ................................ .. 3 Attachment Theory ................................ ................................ ................................ ..... 3 Attachment and Well Being ................................ ................................ ....................... 9 Personal Meaning. ................................ ................................ ................................ 11 Attachment and Psychopathology ................................ ................................ ............ 13 Attachment and Anxiety ................................ ................................ ................. 15 Attachm ent and Depressiveness ................................ ................................ ..... 17 Gender. ................................ ................................ ................................ ......... 18 Need for the Study ................................ ................................ ................................ .... 20 Purpose of the Study ................................ ................................ ................................ . 23 Research Questions ................................ ................................ ................................ .. 23 Definitions of Terms ................................ ................................ ................................ . 24 Organization of the Study ................................ ................................ ......................... 26 2 REIVEW OF THE LITERATURE ................................ ................................ .......... 27 Attachment ................................ ................................ ................................ .. 27 Origins of Attachment ................................ ................................ .................... 27 Stability of Attachme nt ................................ ................................ ................... 33 Criticism of Attachment Theory ................................ ................................ ..... 35 Classification of Attachment Orientation ................................ ....................... 37 Research on Attachment ................................ ................................ ........................... 42 Attachment in Non Humans ................................ ................................ ..................... 45 Attachment In Non Technological Communities ................................ .................... 46 Attachment and College Student Mental Health ................................ ...................... 46 Attachment Research and College Student Mental Health ................................ ...... 50 Attachment and Adjustment to College ................................ ................................ ... 52 Personal Meaning ................................ ................................ ................................ ..... 55 Attachment and Psychopathology ................................ ................................ ............ 58
vi Attachment Status and Depressiveness ................................ ................................ .... 63 Attachment Status and Trait Anxiety ................................ ................................ ....... 68 Attachment and Gender ................................ ................................ ............................ 71 Summary. ................................ ................................ ................................ .......... 74 3 METHODO LOGY ................................ ................................ ................................ ... 76 Statement of Purpose ................................ ................................ ................................ 76 Hypotheses ................................ ................................ ........................... 76 Definitions of Relevant Variables ................................ ................................ ............ 77 Population. ................................ ................................ ................................ ........ 79 Sampling Procedure. ................................ ................................ ......................... 79 Data Collection Procedures. ................................ ................................ ............. 80 Instruments. ................................ ................................ ................................ ...... 80 Experiences in Close Relationships Questionnaire Revised. ................ 81 Life Regard Index Revised. ................................ ................................ .. 86 Depressive Experiences Questionnaire. ................................ .................. 90 State Trait Anxiety Form Y. ................................ ................................ . 94 Data Analytic Procedures. ................................ ................................ ................ 97 4 RESULTS ................................ ................................ ................................ ................. 98 Summary and Chapter Overview ................................ ................................ ............. 98 Descriptive Statistics ................................ ................................ ................................ 98 Results. ................................ ................................ ................................ ....... 102 Summary. ................................ ................................ ................................ ........ 123 5 DISCUSSION ................................ ................................ ................................ ......... 127 Summa ry. ................................ ................................ ................................ ........ 127 Discussions of Results. ................................ ................................ ................... 129 Attachment and Personal Meaning ................................ ............................... 129 Attachment and Depressiveness ................................ ................................ ... 135 Attachment and Trait Anxiety ................................ ................................ ...... 137 Gender.. ................................ ................................ ..................... 139 Personal Meaning and Depressiveness ................................ ......................... 141 Personal Meaning and Trait Anxiety ................................ ............................ 142 Depressiveness a nd Trait Anxiety ................................ ................................ 142 Correlations among the Variables ................................ ................................ 143 Theoretical Implications ................................ ................................ ......................... 145 Clinical Implications ................................ ................................ .............................. 148 Limitations of the Study ................................ ................................ ......................... 149 Implications for Future Research ................................ ................................ ........... 152
vii APPENDIX A INFORMED CONSENT ................................ ................................ ........................ 155 B EXPERIENCES IN CLOSE RELATIONSHIPS REVISED ................................ 157 C LIFE REGARD INDEX REVISED ................................ ................................ ....... 160 REFERENCES ................................ ................................ ................................ ................ 161 BIOGRAPHICAL SKETCH ................................ ................................ ........................... 178
viii LIST OF TABLES Table page 4. 1. Frequencies of Gender ................................ ................................ ............................... 99 4.2. Decriptive Statistics for Attachment Related Anxiety, Attachment Related Avoidance, Personal Meaning, Dependency, Self Criticism, and Trait Anxiety .... 99 4.3. Descriptive Statistics for Attach ment Related Anxiety and Attachment Related Avoidance in Comparison to those of Fraley (2003) ................................ ............ 100 4.4. Distribution of the Participants into the Attachment Categories .............................. 101 4.5. Descriptive Statistics for the Life Regard Inventory Revised in Comparison to Those by Harris and Standard (2001) and Nickels and Stewart (2000) ................ 101 4.6. Descriptive Statistics for Dependency and Self Criticism in Comparison to a McGill University Sample ................................ ................................ ..................... 102 4.7. Descriptive Statistics for Trait Anxiety in Comparison to Norms by Spielberger (1983) ................................ ................................ ................................ .................. 102 4.8. Simultaneous Multiple Regression Output Using Personal Meaning as the Dependent Variable ................................ ................................ ............................... 103 4.9. Multinomial Logistic Regression Analyses Output for Attachment Style as the Outcome Variable with Four Levels Main Effects O nly ................................ ..... 105 4.10. Multinomial Logistic Regression Analyses Output for Attachment Security as the Outcome Variable with Two Levels Main Effects Only ................................ . 107 4.11. Multinomial Logistic Regression Analyses Output for Attachment Style as the Outcome Variable with Four Levels Gender Interaction Effects Included ........... 107 4.12. Multinomial Logistic Regression Analyses Output for Attachment Security as the Outcome Variable with Two Levels Main Effects Only ................................ . 109 4.13. Simultaneous Multiple Regression Output Using Dependenc y Factor of Depressiveness as the Dependent Variable ................................ ........................... 111
ix 4.14. Simultaneous Multiple Regression Output Using Self Criticism Factor of Depressiveness as the Dependent Variable ................................ ........................... 112 4.15. Simultaneous Regression Output Using Trait Anxiety as the Dependent Variable ................................ ................................ ................................ ................ 114 4.16. Independent Sample t Tests Output for Attachment Related Anxiety, Attachment Related Avoidance, Personal Meaning, Dependency, Self Criticism and Trait Anxiety According to Gender ................................ ................................ ............... 118 4.17. Simultaneous Multiple Regression Analysis Out put For Personal Meaning And Trait Anxiety ................................ ................................ ................................ ......... 120 4.18. Correlation Matrix for Attachment Related Anxiety, Attachment Related Avoidance, Personal Meaning, Dependency, Self Criticism, and Trait Anxiety ................................ ................................ ................................ ......... 122
x Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy RELATIONSHIPS OF ATTACHMENT STATUS AND GENDER TO PERSONAL MEANING, DEPRESSIVENESS AND TRAIT ANXIETY AMONG COLLEGE STUDENTS By Ibrahim Keklik May 2004 Chair: James Archer, Jr. Major Department: Counselor Education At a time of increased need for efficient mental health services for college students, empirical studies informing these serv ices are vital. Therefore, the purpose of this study was to investigate the interrelationships between some major personality constructs of college students. Studies with attachment theory propose that how individuals experience close relationships has r elevance to other areas of their functioning. Finding empirical evidence linking attachment to measures of wellbeing and psychopathology will help psychotherapists attain further clarity in interrelationships between major areas of human functioning. Suc h clarity will, in turn, strengthen efficacy of therapeutic interventions with college students. Therefore, this study investigated if college studentsâ€™ attachment styles and their gender were related to measures of wellbeing and vulnerability to psychopa thology. Personal meaning was elected as the measure of wellbeing, whereas depressiveness and trait anxiety were indicators of vulnerability to psychopathology. In this study, attachment consisted of dimensions of anxiety and avoidance.
xi It was viewed as individualsâ€™ tendencies and preferences with respect to anxiety and avoidance in close romantic relationships. Personal meaning was conceptualized as the degree to which one perceives his or her life as purposeful and fulfilling. Depressi veness consisted of factors of self criticism and dependency. Self criticism refers to tendencies of perfectionism, whereas dependency has to do with excessive need and concerns about approval and availability of others for oneâ€™s sense of identity and wel lbeing. Trait anxiety was viewed as oneâ€™s vulnerability to conceive a variety of life situations as threatening and to react to such situations with elevated degrees of fear, tension and anxiety. A convenience sample of college students in personal growt h classes in the College of Education at the University of Florida was used. Surveys of the study were given to 166 voluntary participants. After attrition, responses of 155 students were included in the analyses. One hundred seven of these students wer e female and 48 were male. Data obtained from these students were analyzed using the simultaneous multiple regression analysis, multinomial and binary logistic regression analyses, independent t tests, and correlation matrix. There were results mixed resu lts regarding the relationship between attachment and personal meaning. Factors of attachment and factors of depressiveness were for the most part significantly related. There was no significant relationship between attachment and trait anxiety. Males a nd females differed only on attachment related avoidance, dependency factor of depressiveness and trait anxiety. Personal meaning was not significantly related to depressiveness. Implications of these results, limitations of the study and directions for future research were discussed.
1 CHAPTER 1 INTRODUCTION College experience constitutes a developmentally crucial period for traditional age students (17 23 years of age). The importance of this period is twofold. First, this time is marked with essential developmental tasks such as ident ity formation and career decision making. Second, in addition to such normative challenges, the college experience itself provides an array of challenges and tasks. College students â€œare not merely adolescents who happen to have entered college, nor are t hey adults who happen to have not yet finished school they are adolescents who are in the process of becoming young adultsâ€ (GAP; Group for the Advanceme nt of Psychiatry, 1990, p. 5). For many students, college is the first time they leave home and encoun ter major challenges of adjusting to a new environment and succeeding both personally and academically on their own. The period of 17 23 years of age is often referred to as late adolescence and is characterized by remarkable interpersonal and intraperso nal activity. Many individuals form their first long term romantic relationships, take on considerable educational challenges, and leave home to pursue education.
2 With all these activities college students are â€œengaged in a process of self definition and are attempting to forge a sense of identity in the context of their previous experiences, current relationships, and future expectationsâ€ (Kenny & Rice, 1995, p. 43 3). In last few decades the nature of college experience has been of great interest to various professionals including those from the fields of education and mental health. Sanfordâ€™s (1962, 1966) work has had great impact on research with college student d evelopment (Chickering & Reisser, 1993). He strongly emphasized the importance of balancing the support and challenges college students encounter in order to foster optimum growth. This concept stems from the idea that if learners encounter challenges th at exceed their perceived resources, their performance is not likely to be optimal. Similarly, optimal learning cannot be attained if the resources are too plentiful and there are not enough challenges. This notion of balance has guided research and serv ices pertaining to college students. Given the unique challenges of contemporary college students, it is essential to attain greater understanding of their handling of such challenges. Since the attachment theory is a theory of survival and adaptation, it can provide insight into college studentsâ€™ efforts toward adjusting to and surviving through such a developmentally significant experience. Similar to attachment, the concept of personal meaning is also closely associated with adaptation and survival. T his study examines the relationships of attachment status to indicators of well being and proneness to psychopathology. More specifically, personal meaning will be used as an indicator of psychological well being, while depressiveness and trait anxiety wi ll be the two measures of psychopathology.
3 The reason for the selection of these two issues is twofold. The first reason has to do with the prevalence of depression and anxiety among college students (Fulkerson, 1986; Archer & Cooper, 1998). The second r eason for this choice is the fact that these two constructs (depressiveness and trait anxiety) are personality traits. Since the literature more commonly involves studies with states, an examination of these two traitsâ€™ relations to attachment and persona l meaning can make unique contributions to research. Scope of the Problem Despite the increased attention attachment theory has received during the last decade, attachment research is yet to grow beyond basic research inquiries. In addition, in a time of limited resources for mental health services (Friedman, 1997), the development of time effective interventions is only possible with sufficient empirical knowledge. Similarly, attachment theory can contribute to clinical practice only when sufficient pro gress is made in attachment research. Part of such progress involves studies of various populations regarding an array of mental health issues. Hence, this study will contribute to this advancement in attachment research and college student mental health . Attachment Theory Attachment theory originated from research on the infant caregiver relationship. Robertson and Bowlby (1952) identified infantsâ€™ behavioral patterns followed by separations from their mothers. These observations led Bowlby to pay clo se attention to the importance of the bond between the infant and care giver. In addition to these observations, Bowlby (1960, 1969) benefited from multiple disciplines in constructing the attachment theory. Lopez (1995) indicates that Bowlby drew from th e developmental views of Jean Piaget (1950, 1972) and Erik Erikson (1968), the ethological research of
4 Konrad Lorenz (1952) and Harry Harlow and Zimmerman (1959), and the general systems theory of Ludwig von Bertalanffy (1968). Bowlby concluded that behav iors by which infants sought and maintained proximity to caregivers were keys to evolutionary survival of the human species. He viewed such behaviors as vital strategies that help survival in species with developmentally immature and dependent offspring ( Simpson, 1999). He used the term attachment behavioral system to refer to a species specific and organized set of behaviors that enable human infants seek proximity to their caregiver. Attachment theorists have paid particular attention to three behavior al systems, attachment, fear/wariness, and exploration (Cassidy, 1999). Along with John Bowlby, Mary Ainsworth (1964, 1967, & 1990) is considered to be another founder of the attachment theory. Ainsworth furthered attachment theory with her studies in l aboratory conditions (the Strange Situation; Ainsworth et al., 1978) and with first multicultural comparisons of infantsâ€™ attachment behaviors. Along with her colleagues, Ainsworth was the first researcher to identify attachment orientations, which she de lineated as secure, avoidant and anxious ambivalent (Ainsworth et al., 1978). These researchers used laboratory observations through a standardized procedure, also known as the Strange Situation (Ainsworth et al., 1978). The procedure was consisted of tw o separations and reunions between infants and their caregivers. Infantsâ€™ behaviors following separations and reunions were carefully recorded and significant individual differences found were classified into three distinct styles of attachment; secure, a nxious ambivalent, and avoidant. Moreover, Ainsworth and colleagues (1978) confirmed these laboratory findings with their home observations of infants and caregivers (Goldberg, Muir, & Kerr, 1995).
5 The secure infants were more likely to use the mother as a secure base for exploration. They showed signs of missing the caregiver upon reunion and actively sought contact and comfort from the caregiver. Avoidant infants were characterized with readily having exploratory behavior independent of the caregiversâ€™ presence. They also seemed to have little display of affect or secure base behavior. These infants did not show signs of distress when caregiver departed nor did they show much interest in the caregiverâ€™s return. Finally, the anxious ambivalent infants did not display much exploratory behavior, and appeared passive. Upon separation from the parent they displayed distress. However, when the parent returned these infants often displayed signs of anger and tantrums, and seemed to have difficulty finding c omfort in the parent. Following this initial focus on infant caregiver relationship, attachment theory has evolved to address various periods of the lifespan development. Furthermore, this remarkable blooming in attachment research has embraced a variety of groups and mental health issues. The relationship of attachment to adult romantic love (Hazan & Shaver, 1987), adultsâ€™ relational representations (Main, 1995), developmental psychopathology (Petersen, Kennedy, & Sullivan, 1991), severe psychopathology (Dozier, 1990), family dysfunction and parental alcoholism (Mothersead, Kivlghan, & Wynkoop, 1998), adjustment to college (Kenny, 1987; Kenny & Perez, 1996; Kenny & Rice, 1995), and same sex romantic relationships (Mohr, 1999) are only a few examples of t he growth in attachment research. Such studies have examined the relationships of adult attachment to development of adult psychopathology and to adult functioning and well being. Similarly, the present work focused on two major issues of psychopathology (depression and anxiety) and their possible relations to college studentsâ€™ attachment orientations.
6 In addition, it looked into college studentsâ€™ attachment orientation as it relates to their degree of personal meaning. Attachment theory has been appli ed to college experience in a variety of ways. For instance, some researchers have examined adjustment to college and perceived parental availability (Hazan & Hutt, 1993). Attachment theory looks at this availability based on the notion of secure base (A insworth, 1989). The term secure base is one of the key concepts in attachment theory. It was originally defined as an infantâ€™s perceived availability of the caregiver as a safe haven to return to in the face of distress or danger. Perceiving family and friends as a secure base to return to for reassurance and support might be of significant importance in fostering young adultsâ€™ normative risk taking behavior and exploration in various areas of their lives, such as academic, social, personal and the like. Attachment theorists view a perceived secure base to also have indirect contributions to oneâ€™s development and experience in college. Such contributions are thought to be through internal working models of the self and other. Attachment theory (Bowlby, 1980) argued that when the attachment figure is available, responsive and reliable the child will form an image/representation of self as good, worthy, and lovable. Attachment theory also maintains that such a person will have an internal model of other a s trustworthy, responsive and reliable. Thus it is assumed that individuals with such representations will be more likely to establish and utilize supportive interpersonal relations.
7 For traditional age college students, starting college is a significa nt experience of separation (Lapsley & Edgerton, 2002; Rice, FitzGerald, Whaley, & Gibbs, 1995). According to traditional developmental theories (i.e. Erikson, 1968) separation is the prerequisite to individuation, which refers to the process of becoming oneâ€™s own person. From an attachment theory stand point, college is an experience of exploring the world and, thus is a modern rite of passage toward seeking oneâ€™s identity and place in the adult world. Attachment theory claims that in this process, the young college studentâ€™s relational orientation will impact his or her adaptation and functioning as a result of his or her perception of self and others which are presumed to have been formed in early close relationships. For instance, it is hypothesized that compared with individuals with insecure attachment, securely attached ones would be more likely to form new secure bases in their new environment. The connection between internal working models and psychological well being is established with the ass umption that internal working models influence ways in which one perceives subsequent life events. Thus it is thought that a person with a negative internal working model of self would have a predisposition to perceive negative events as his or her failur e, which would in turn perpetuate distress. Conversely, based on these assumptions Kenny and Rice (1995) postulate that college students who posses positive internal working models will be less vulnerable to experience inevitable situational stressors, su ch as rejection by their peers or receiving undesirable grades in ways that would be disempowering to the self. The authors add that those who anticipate others to be trustworthy may be better equipped to form supportive relationships with peers and facul ty.
8 Considering such assumptions and Sanfordâ€™s (1962, 1966) notion of balance between challenges and support, it is likely that students with positive internal working models would be better able to recognize existing support or obtain further social sup port as needed. In addition, college students use internal working models of self and other â€œas filters through which self perceptions are developed and level of social support from others is judgedâ€ (Rice & Cummings, 1996, p. 51). Such perceptions of se lf and others are crucial since college experience is a period of identity formation (Chickering & Reisser, 1993). That is to say, the quality of such schemas will impact oneâ€™s sense of self and how college is experienced. Consequently, the quality of co llege years could greatly affect how young adults construct a sense of identity. Psychopathology is often developed or exacerbated during transitional times such as adolescence and young adulthood. One of the important factors influencing the development of psychopathological symptomatology is the extent to which individuals resist ongoing stressors by utilizing internal and external resources (Kenny & Rice, 1995). Attachment orientation can directly or indirectly influence the quality of such resources. Attachment theory proposes that parental attachment may â€œindirectly influence coping resources through their contributions to internal working models of self and otherâ€ (Kenny & Rice, 1995, p. 436). In their literature review Mikulincer, Horesh, Eilati, and Kotler (1999) document studies finding associations between attachment organization and management of distress. Securely attached persons have been found to utilize constructive stress management strategies and seek help from others when needed
9 (Miku lincer & Florian, 1998); have positive expectations about manageability of the distress experienced and strong sense of self efficacy (Mikulincer & Florian, 1995); and hold positive perceptions about othersâ€™ intentions (Bartholomew & Horowitz, 1991). On t he other hand, Mikulincer et al. (1999) report that avoidant persons are likely to deactivate attachment needs in dealing with stress, deny negative feelings, inhibit expression of perceived distress, and utilize dissociative coping strategies (Mikulincer & Florian, 1998). Finally, anxious ambivalent persons deal with stress in hypervigilant ways, hyperactivate negative thoughts and memories, and utilize less active ways of coping. In short, the management of distress seems to be mediated by intrapersonal factors such as a strong sense of self efficacy and interpersonal factors such as positive perceptions of others. These two sets of factors appear to foster utilization of constructive strategies in dealing with distress. Attachment and Well being A grea t number of studies have confirmed relationships between attachment status and various measures of well being. Such efforts have focused on attachmentâ€™s relationship to social and relational competence (Engels, Finkenauer, Meeus, & Dekovic, 2001); self es teem (Rice & Cummings, 1996); self efficacy (Rice & Cummings, 1996); autonomy, competence and relatedness (La Guardia, Ryan, Couchman, & Deci, 2000); health and illness (Feeney, 2000); psychological health of partners in long term marriage or cohabiting re lationships (Kotler & Omodei, 1988); lack of symptomatology (Kenny, Lomax, Brabeck, & Fife, 1998; Raja, McGee, & Stanton, 1992); and self worth (Kenny et al., 1998). These studies found moderate to strong associations between secure attachment and measur es of well being.
10 Only a few of such studies however focused on attachment and well being during late adolescence. This study examined the concept of personal meaning as a measure of well being in college students for several reasons. First, a sense of meaning is essential in transitional and difficult times (Klinger, 1998). Personal meaning does not only motivate goal oriented behavior but also buffers against stress, thus promoting a sense of well being. Second, along with its other roots, personal meaning emerged as a component of the concept of sense of coherence (Antonovsky, 1979). Sense of coherence has been linked to attachment security and adjustment in college students (Rood, 2000). Moreover, personal meaning has been found as the most predi ctive component of sense of coherence that is associated with well being and adjustment (Klinger, 1998). Third, similar to attachment, personal meaning seems to be a global and organizational construct. Stated differently, personal meaning has to do with purposefulness. As such, it has the potential to be an integrating personality construct. Consequently, a likely association between attachment and personal meaning could imply a link between attachment and major constructs of personality. Therefore, a ssociations among attachment status, personal meaning, anxiety and depression could inform clinical practice and future research. It appears likely that oneâ€™s perceived level of personal meaning might be essential in motivating oneâ€™s involvement in variou s domains of life, thus creating a circular causation that maintains the chain of meaning involvement meaning. Furthermore, such a chain might be fostered in a desirable direction if the person feels belonging to circles of others.
11 Personal Meaning Alon g with the research tradition focusing on development of psychopathological symptomatology there has been a line of research emphasizing the development of adaptive mechanisms. This research was in part inspired by traumatic events such as the experience of German concentration camps during World War II. Researchers intended to find out how some persons managed to survive through such tragic experiences. Dr. Victor Frankl (1963), a survivor of these camps, was a major figure who influenced this school of thought. However, such inquiries date before Dr. Frankl. For instance, the German philosopher Friedrich Nietzscheâ€™s writings impacted scholars like Frankl in their quest to understanding what might enable individuals to endure seemingly unbearable condi tions of living. The decades following the war witnessed a great deal of interest in finding out such buffering effects against high stress situations. Frankl (1963) saw meaning as one of these buffers. In his book Manâ€™s Search for Meaning he summarizes this with a quote from Nietzsche: â€œhe who has a why to live for can bear with almost any howâ€ (p. 121). Klinger (1998) provides linguistic etymological, philosophical and psychological analyses of the word â€œmeaningâ€. These analyses indicate that in many languages the words corresponding to the term meaning refer to aim, goal, and intentionality. It is considered a high cognitive entity that has goal directed, attributional, motivational, and integrative functions. In other words, meaning is closely rel ated to the goal directed nature of human behavior and, thus it is motivational. In addition, it is viewed as connecting and integrating related mental representations, as such having an integrative function. Klinger (1998) notes that without such a func tion human behavior would be
12 merely guided by instincts or impulses. In short, meaning making is an evolutionary gain of the human species and it is about certain understandings of things that are specific to humans. Klinger (1998) furthermore claims tha t the human brain cannot sustain purposeless life. In other words, he believes that meaningfulness is an inevitable need that is essential for human functioning and survival. Then the question becomes, what happens when individuals lack a sense of such an essential element of living? According Klinger (1998) such deficiency is analogous with not having goals and is associated with psychopathology. Moreover, such outcomes are thought to be plausible in the case of having excessively aversive goals as well . Klinger (1998) proposes that â€œa set of goals steers attention, perception, cognition, recall, thought content, and emotional responses, hence, exercises pervasive influence over consciousnessâ€ (p. 44). In addition, it is claimed that such a state inter feres with the major ingredient of psychological organization. There could be direct and or indirect connection between lack of meaning and behavioral patterns that are self destructive. In the face of not having goals and purposes one would resort to ot her means to improve a sense of well being (Klinger, 1998). One of the issues often associated with lack of goals and purpose is depression, which is often characterized with excessive self criticism. Similarly, the lack of meaning or predominantly avers ive goals could lead to prolonged negative affective states that may lead to stress and anxiety, thus endangering oneâ€™s physical well being. Klinger (1998) reports empirical studies that have found associations between the level of personal meaning and var ious indicators of well being and psychopathology. A global sense of well being, substance abuse related issues, and depression are some of
13 these indicators. There is still a great need for studies focusing on personal meaning and various measures of men tal and physical health. No studies dealing specifically with attachment and personal meaning were located. The only work found is a dissertation study (Rood, 2000) examining the relationship between a sense of coherence and attachment, which involves mea ningfulness or personal meaning as a component of coherence. Similarly, relationships between depression or anxiety and personal meaning have only been studied as a part of the concept of sense of coherence (Rood, 2000; Sandell, Blomberg, & Lazaar, 1998). This study examined the relationships of personal meaning, attachment status, depressiveness and trait anxiety. By so doing, along with its additional objectives, it intends to extend research on personal meaning. Attachment and Psychopathology Despite the recent interest in adult attachment (Allen, Hauser, & Borman Spurrell, 1996; Bartholomew & Horowitz, 1991; Cummings & Davies, 1996; Feeney & Noller, 1990; Hazan & Shaver, 1987; Kobak & Hazan, 1991; Magai, Hunziker, Mesias, & Culver, 2000; Main, 1996; R osenstein & Horowitz, 1996; Pianta, Egeland, & Adam, 1996) there is a need for further exploration on the possible associations between adult attachment and psychopathology (Roberts, Gotlib, & Kassel, 1996; Mothersead, Kivlghan, & Wynkoop, 1998);). The ex isting body of research acknowledges links between insecure attachment and psychopathology. It also ties secure attachment to various indicators of well being (del Carmen & Huffman, 1996; Rosenstein & Horowitz, 1996). On the other hand, as pointed out by Roberts et al. (1996) the underlying mechanisms of these associations are yet to be examined.
14 Whether looked at from a linear traditional perspective or from a pathway developmental perspective, attachment theory suggests that attachment quality plays a crucial role in leading to vulnerability to psychopathology (Rosenstein & Horowitz, 1996). From a theoretical standpoint, internal working models could at least play a partial role in the link between attachment and psychopathology. As mentioned earlier, internal working models of attachment are representations of the self, and other, which evolve through significant relationships starting from the onset of life. Accordingly, such schemata are thought to also constitute representations about ways in whic h one thinks or feels about the world in general. Hence, the connection between attachment status and psychopathology might partially be mediated by cognitions that are parts of internal working models (Roberts et al., 1996). For instance, an individual who holds negative notions about self and others would be likely to experience depression and or anxiety. As acknowledged by Roberts et al. (1996) such reasoning is inline with widely endorsed views of cognitive theorists (i.e. Beck, 1963). The unique con tribution attachment might have in the development of psychopathology could be due to its early formation, which in the case of insecure attachment might lead to deviations from adaptive developmental pathways (Rosenstein & Horowitz, 1996). Hence, such de viations in early years of life might constitute vulnerability to risk factors whether they be genetic or environmental. According to Main (1996) along with genetic (i.e. temperament) and other factors (i.e. psychopathology in the caretaker), early trauma or loss, also leads to development of disorganized attachment.
15 One way of examining whether there are convincing associations between psychiatric classification and attachment organization, would be to find consistency in specific classifications regardi ng attachment style. Such efforts have yielded to contradictory outcomes (Rosenstein & Horowitz, 1996). Therefore, such a link is yet to be empirically established. Issues such as depression and anxiety are thought to be due to mechanisms with which dis tress is directed inward as opposed to outward, to the external world as it is in conduct disorders. Existing literature often links issues resulting from internalization, such as depression, to preoccupied attachment organization (Allen et al., 1996; Cow an, Cohn, Cowan, & Pearson, 1996). Attachment and Anxiety Since the emergence of attachment theory, researchers have pointed out the link between attachment and affect regulation. Kobak and Sceery (1988) view attachment theory as a model of affect regulat ion. They reason that attachment styles are differentiated based upon individualsâ€™ responses to stressful situations. The formation of attachment in the infant caregiver relationship is indeed a process of regulating affect. Attachment literature often associates preoccupied (anxious ambivalent) attachment pattern with the experience of anxiety. Unlike those with a dismissing style (avoidant), preoccupied persons are thought to continually seek support and acknowledgment from the attachment figure with whom a secure bond has not been established. Furthermore, these individuals do not dissociate memories of negative events and tend to have clear memories of such events. Preoccupied persons hold negative models of self and positive models of others.
16 Lop ez (2001) found relationships between preoccupied attachment style and low self other differentiation, high levels of emotional reactivity, strong expressed needs for social approval and self concealment. Such findings are in line with research suggesting these individuals being highly concerned about rejection and approval of others. Lopez recommends that such persons are likely to project their own self traits onto others in order to feel self other similarity. The presumption is that along with directl y dealing with anxiety and distressing affect, individualsâ€™ defenses are set to either dismissing such affect or becoming preoccupied with it. Lopez (2001) found these two tendencies to be analogous to avoidance versus anxiousness respectively. Some refe rred to these two mechanisms as means by which attachment system is deactivated (dismissing/avoidant) or hyperactivated (anxious/preoccupied) (i.e. Kobak, Cole, Ferenz Gillies, Fleming, & Gamble, 1993; Main, 1990). Many researchers point out the relationsh ip between anxiety issues (i.e. phobias, obsessive compulsive disorder) and the tendency to have excessive concern about control of emotions (Mantynen, Happonen, & Toskala, 2000). Such self consciousness and tendency to control feelings is also seen in in dividuals with preoccupied attachment orientation. Thus there might be a significant link between insecure attachment particularly preoccupied type, and trait anxiety. Establishing such association could inform clinicians about the likely presence of one of these factors when the other is reported.
17 Attachment and Depressiveness There is a convincing body of research linking insecure attachment and depression (i.e. Carnelley, Pietromonaco, & Jaffe, 1994; Feeney & Noller, 1990; Roberts et al., 1996). Rob erts et al. (1996) recommend a model describing association between insecure attachment and depression. The researchers assert â€œattachment insecurity is associated with dysfunctional attitudes, which in turn decreases self esteemâ€ (p. 312) thus leading to depressive symptoms. To test this model they conducted a study with a sample of college students. The results indicated that the relationship between attachment and depression was â€œmediated almost entirely by maladaptive contingencies of worth and self esteemâ€ (p. 317). Finally, the findings also illustrated that individuals with negative working models of self and positive working models of others were more likely to report clinical depression whereas those with negative working models of both self and others were likely to report nonclinical depression. Susceptibility to depression is often linked to two personality traits: dependency and self criticism/perfectionism. Blatt (1974) was the first researcher to formulate this model (Whiffen, Aube, Thomson , & Campbell, 2000). Dependency is referred to as an excessive need for love and approval. Whereas self criticism or perfectionism is viewed oneâ€™s excessive concerns about achievement and success. However, what seems to be problematic is not the mere pr esence of such needs and tendencies. How these needs lead to vulnerability to depression has to do with the extent to which they dominate oneâ€™s personality. The connection between insecure attachment and depression is assumed to be mediated by internal w orking models. Cummings and Cicchetti (1990) note that
18 â€œdisturbances in attachment may play a role in the development, maintenance, and intergenerational transmission of depressionâ€ (p. 342). These researchers provide a lengthy review of the history of t he work linking attachment and depression. They note that the psychoanalytical tradition has long acknowledged a connection between early childhood experiences with attachment figures and development of depression. Furthermore, this link is more visible when depression is viewed as resulting from actual or perceived loss. Hence, as acknowledged by the object relations tradition such loss is also seen in those who did not internalize positive attributes of significant individuals. Finally, similar to ma ny other scholars Cummings and Cicchetti (1990) state that such sense of loss, which is referred to as object loss, is essential in facilitating vulnerability or resilience to the experience of depression. Considering the prevalence of depression among co llege students (Archer & Cooper, 1998), examining associations between attachment styles and depression will strengthen the previous linkages reported in the literature. Furthermore, by focusing on depressiveness as a personality trait rather than a state of depression, the possible links between attachment orientation and depressiveness could imply the necessity of interventions with more psychodynamic and interpersonal focuses along with the commonly used cognitive behavioral strategies. Gender Starting with the work of Gilligan (1982), researchers have been sensitive to gender issues in psychology. Before such work, traditional developmental models of psychology (i.e. Erikson, 1968) put strong emphasis on individuation at the cost of neglecting interper sonal relatedness. Males are typically thought to be concerned with
19 separation, whereas females with relationality and connectedness. Thus these early models are criticized for having a significant gender bias in favor of males. Current literature in v arious areas of psychology includes a great deal of work addressing gender as a highly important factor in various domains of inquiry. Attachment theory is unique in acknowledging interrelations of individuation and interpersonal connectedness. Researcher s have consistently found differences between male and female behaviors in the context of interpersonal relationships (i.e. Belenky, Clinchy, Goldberger, & Tarule, 1986; Gilligan, 1982). Despite these findings, attachment research has reported mixed resul ts concerning gender differences in the attachment behavioral system. Kenny and Rice (1995) report studies showing the quality of attachment being more â€œimportant to the psychological well being (Kenny & Donaldson, 1991) and identity achievement (Benson, Harris, & Rogers, 1991) of women than menâ€ (p. 443). On the other hand, there are studies, which did not detect any gender differences in attachment related behaviors. There is evidence to conclude that even if there are gender differences they might be manifested differently for each gender. For instance, Rice and Whaley (1994) found that in general, attachment to both parents was more predictive of adjustment to college for female students than it was for males. On the other hand, attachment to the fa ther was a predictor of adjustment only during highly stressful times (i.e. exam period). Other studies have found certain differences between genders. However these differences don not seem to be of a magnitude that would be expected by feminist research ers. Using the IPPA (Inventory of Parent and Peer Attachment; Armsden & Greenberg, 1987) some researchers have found that females in both adolescent (Raja, McGee, & Stanton, 1992) and college samples (Lapsley, Rice, &
20 Fitzgerald, 1990) scored higher on at tachment to peers; trust and communication scales. However, males and females did not differ on several other scales such as attachment to parents. Raja et al. (1992) concluded that perhaps gender differences found in developmental processes of males and females may not be generalized to every area of functioning. For example, these researchers cite studies (i.e. Steinberg & Silverberg, 1986; Williams & McGee, 1991) finding that girls viewed themselves as more autonomous and independent than boys. In su m, there are mixed findings about the claim that female identity is more narrowly based upon attachment to others. Given the extensive empirical work in gender differences and similarities, and the contrasting research findings pertaining to attachment and gender, further work is needed. The current study examined the relationship of gender to attachment orientation, personal meaning, trait anxiety, and depressiveness. Need for the Study The last century has witnessed a great increase of knowledge in the field of psychology. This period, which is also known as the modernist era, has witnessed work by great scholars such as Sigmund Freud, B.F. Skinner, Jean Piaget, Carl Rogers and many others. The inquiry into human functioning gave way to various fields of psychology and related disciplines. In the mean time, great developments in areas such as anthropology, neurology, genetics, etc. have enriched the understanding of human behavior. Given the presence of such accumulated scientific and clinical work, the inquiry of human behavior is faced with the challenge of utilizing multidisciplinary knowledge in understanding specific domains of human functioning. In other words, the existing depth and breath of knowledge in various interrelated fields make it po ssible to
21 establish descriptive, causal or predictive models explaining human behavior. Accordingly, there is need â€œfor a metaperspective capable of integrating important personality and developmental constructs in ways that both inform and stimulate our research and practiceâ€ (Lopez, 1995, p. 305). The search for such a metaperspective is in fact an effort to identify personality constructs that have the potential to explain and simplify complex dynamics of human functioning, development and well being. Self esteem, self concept, attachment, learned helplessness, locus of control, coherent sense of self, personal meaning and the like are some of such concepts. The current work focused on two of these constructs, namely, attachment orientation and person al meaning as they relate to college student mental health. Mental health services are essential parts of college life (Talley & Rockwell, 1985). Counseling with college students constitutes a supportive service that aids college students in their effor ts to deal with developmental and academic tasks. In many countries, the US in particular, college campuses are increasingly becoming diverse. Given the number of college students, and their importance for the future of any given society, providing them with quality educational experiences is crucial. Part of this quality has to do with supportive services offered on campuses. Considering studentsâ€™ degree of utilization of counseling services on university campuses, it seems clear that these services ar e integral parts of college education (Archer & Cooper, 1998). There are various reasons for focusing on college student mental health. First, ages 18 through 23 is a transitional developmental period. The transitional nature of this period is essential in the formation of psychopathological as well as adaptive behavioral patterns. Second, in any given society the quality of college experience will have
22 essential outcomes for the societyâ€™s future since these individuals are valuable assets of the society . Normative difficulties of transitional late adolescence are further exasperated by contemporary societal factors. Societies, markets and higher educational systems are in greater changing paces then ever. Such rapid change itself is also a contributin g factor to challenges faced by college students (Archer & Cooper, 1998). Third, any improvement in the well being of this group, or any preventive effort of the development of enduring psychopathology for that matter, could have longstanding positive con tributions to these personsâ€™ adult life. With the increasing diversification of college student bodies, mental health services have been in a process of seeking new ways to effectively respond to the needs of this population. An important way of doing so is to utilize scientific inquiries in order to gain greater understanding of various aspects of this population. Given the high numbers of college students and the great demand for mental health services on contemporary college campuses, mental health pro fessionals are in greater need than ever to attain a level of efficacy to meet these emerging needs (Bishop & Trembley, 1987; Stone & Archer, 1990). Empirical support for certain metaconstructs capable of integrating complex dynamics of young adult colleg e studentsâ€™ functioning is thus vital. For instance, such data could contribute to professionalsâ€™ arriving at some theoretical clarity that can greatly improve the efficacy with which they serve to increasing numbers of college students. Hence, examining two major issues depression and anxiety that are prevalent among college students in light of constructs as attachment status and personal meaning, could improve professionalsâ€™ understanding of college studentsâ€™ experience with these
23 two issues. Accordin gly, such improvement will have important implications for both research and practice. Purpose of the Study The purpose of this study was to investigate the relationships among attachment orientation, gender, personal meaning, depressiveness, and trait anx iety in college students. More specifically, it examined the extent to which college studentsâ€™ attachment orientation, and gender account for variations in individual differences on measures of personal meaning, depressiveness, and trait anxiety. Research Questions This study sought to answer the following questions: 1. What is the relationship between college studentsâ€™ attachment status and their level of personal meaning? 2. Do gender, personal meaning, depressiveness, and trait anxiety significantly predict college studentsâ€™ likelihood of being categories of attachment? 3. What is the relationship between college studentsâ€™ attachment status and their degree of depressiveness? 4. What is the relationship between college studentsâ€™ attachment status and their degre e of trait anxiety? 5. What is the relationship between college studentsâ€™ gender, their attachment status, their levels of personal meaning, depressiveness, and trait anxiety? 6. What is the relationship between college students' degree of personal meaning and their degree of depressiveness? 7. What is the relationship between college studentsâ€™ level of personal meaning and their degree of trait anxiety? 8. What is the relationship between college studentsâ€™ degree of depressiveness and their degree of trait anxiety ? 9. What are interrelationships among college studentsâ€™ attachment status, degree of personal meaning, degree of depressiveness, and their degree of trait anxiety?
24 Definitions of Terms College Students For the purpose of this study the term college studen t refers to young adult undergraduate students who are of ages 18 through 23. Attachment The terms attachment, attachment style, attachment orientation, and attachment status are used interchangeably. They refer to young adult college studentsâ€™ perceived orientations in interpersonal contexts, particularly in romantic relationships. This definition is inline with the research tradition of Hazan and Shaver (1987) who emphasize romantic adult relationships as reflective of early attachment relations. It m ore specifically utilizes the perspective Brennan and colleagues (Brennan, Clark,& Shaver, 1998; Fraley, Waller, & Brennan, 2000). Accordingly, attachment classification or orientation is based upon individualsâ€™ self reported degrees of anxiety and avoida nce in the context of intimate relationships. According to Brennan et al. (1998) scores on the ECR correspond to the four attachment categories of Bartholomew and Horowitz (1991) in the following manner: (a) Low scores on both avoidance and anxiety dimens ions correspond to the secure category. (b) High scores on the avoidance and low scores on anxiety dimensions correspond to avoidant (dismissing) category. (c) High scores on the anxiety and low scores on the avoidance dimension correspond to the anxious ambivalent (preoccupied) category of attachment. (d) High scores on both anxiety and avoidance dimensions correspond to the fearful classification of attachment. While referring to this particular conceptualization of attachment the term attachment class ification will be used.
25 Internal Working Models In this study the term internal working models refers to individualsâ€™ schemas about themselves and others, which are not merely of a cognitive nature but also affective. Furthermore, such perceptions are thought to be formed in early childhood and be relatively enduring throughout later development. Personal Meaning The term personal meaning is used to refer to oneâ€™s general evaluations of his or her behaviors and life style in comparison to personal goals and values. These goals and values refer to a personâ€™s life long aspirations, which might also be motivated in part by the personâ€™s beliefs about after life. Stated differently, personal meaning has to do with an individualâ€™s perceptions of his or her c urrent life depending upon its perceived compliance with what the person views as meaningful. Depressiveness For the purpose of this study depressiveness refers to college studentsâ€™ self reported vulnerability and experiences with depressive symptomatology . This vulnerability is based upon the individualâ€™s responses to items on scales of self criticism and perfectionism. High scores on these two dimensions are thought to indicate high degree of vulnerability to experiencing depression symptomatology. It is important to note that depressiveness is not analogous to clinical depression. High scores on self criticism and perfectionism may or may not correspond to clinical depression.
26 Trait Anxiety In this study trait anxiety refers to college studentsâ€™ se lf reported proneness to experiencing anxiety. In other words, it refers to predisposition to experiencing a variety of life situations as threatening and responding to these situations with excessive states of anxiety. Organization of the Study This diss ertation consists of five chapters. Chapter 1 introduces the scope of the problem, theoretical framework for the study, need for the study, purpose of the study, the research questions, and definitions of the terms. Chapter 2 introduces a review of the r elevant literature. Chapter 3 describes the research methodology, which consists of the following: population and sample; instrumentation; and procedures of sampling; data collection and analysis. Chapter 4 introduces results of the study. Finally, chap ter 5 presents conclusions of the study, its theoretical and clinical implications, recommendations for future research, and its limitations.
27 CHAPTER 2 REVIEW OF THE LITERATURE The purpose of this study was to investigate the relationships among attachment, status, gender, personal meaning, depressiveness, and trait anxiety. In this chapter, literature related to these variables is provided to develop a framework for the research hypotheses. It is organized in the following ten sections: (a) Origins of attachment; (b) stability of attachment; (c) criticism of attachment theory; (d) classification of attachment orientation; (e) research on attac hment; (f) attachment and college student mental health; (g) attachment and well being; (h) attachment and psychopathology; (i) attachment and gender; and (j) summary. Attachment Origins of Attachment John Bowlby and Mary Ainsworth are considered the found ers of attachment theory. Through their individual and collaborative work, both researchers sought to find the role of enduring emotional bonds in adjustment and functioning. Bowlby was trained in classical psychoanalysis and in the object relations mode l of Klein (1932). However, he was dissatisfied with these approaches. Bowlby rejected seeing â€œall relationship seeking motivations in terms of the direct or indirect gratification of libidinal needâ€ (Lopez, 1995, p. 397).
28 Through his work with the delinquent youth and hospitalized orphans, Bowlby concluded that â€œthere were qualities of the parent child emotional bond, independent of the gratification of libidinal needs or drives, that regulated important aspects of child be havior and emotional experienceâ€ (Lopez, 1995, p. 398). Bowlby developed his model using an interdisciplinary perspective. Lopez (1995) indicates that Bowlby drew from the developmental views of Jean Piaget (1950, 1972) and Erik Erikson (1968); the etho logical research of Konrad Lorenz (1952) and Harlow and Zimmerman (1959); and the general systems theory of Ludwig von Bertallanffy (1968). Attachment theory emerged from research with infant caregiver relationship. Moreover, it inquired as to how these relationships impact personality development. Subsequent studies extended these inquiries across various developmental periods throughout the life span. Bowlby argued that human attachment behavior is a product of the evolutionary process and that it ori ginated from a necessity for protection from predators (Harwood, Miller, & Irizarry, 1995). Bowlby did not conceptualize attachment merely on the basis of such protection. He â€œdid emphasize that because attachment behaviors help to maintain the infantâ€™s proximity to the parent, they thus served to promote the infantâ€™s physical safety and survival in the environment in which humans originally evolvedâ€ (Harwood et al., 1995, p. 5). Bowlby (1982) saw striving for relatedness as vital to survival through the evolutionary process. Complementing this evolutionary point of view, Lyddon (1995) conceptualizes attachment as â€œa hardwitted, neurologically based behavioral system that has evolved to promote proximity to a caregiverâ€ (p. 479).
29 Bowlby focused on two k ey factors in conceptualizing the development of internal working models: (a) the amount of stress on the attachment system; and (b) the availability of the attachment figure. Prolonged high levels of stress and/or the lack of availability of the attachme nt figure could lead to an internal representation of the environment as dangerous and of the self and the others as ineffective in moderating these dangers; such a representation may leave the child fearful of exploration, uncertain of the availability o f safety, doubtful of his or her ability to master environmental demands, and/or distrustful of significant others. (Harwood et al., 1995, p. 6) Lopez (1995) reasons that Bowlby considered these schema as working models â€œbecause they (a) organized intern al appraisals and interpersonal behaviors along pathways that were adaptive in the personâ€™s earlier development, and (b) thereby shaped the personâ€™s later social experiences in schema consistent waysâ€ (p. 399). In short, an internal working model â€œis a se t of conscious and/or unconscious rules and expectationsâ€ regarding oneâ€™s perceptions of self and others, which stems from early care giving relationships and functions as a model or schema that â€œis used as a template in future relationshipsâ€ (Flanagan, 19 99, p. 103). Concepts similar to internal working models were used by other theorists such as Fairbairn (1952) and Klein (1932) of the object relations tradition. Object relations focused on relationships with others and how the internalizations of these individuals impact psychological functioning. Similar to other psychodynamic traditions, object relations very much highlighted the important implications of early relationships on later development. Like internal working models, the internalized represe ntations conceptualized by object relations theory had both cognitive and affective mental representations (Cummings & Cicchetti, 1990).
30 Crittenden (1990) proposes a unique approach to conceptualization of internal working models. This approach views int ernal representations as open versus closed and working versus nonworking models. Closed or nonworking models do not permit new interpretations, thus they constitute a defensive posture in interpreting interpersonal experiences. On the other hand, the op en working models are flexible and open to new and alternative interpretation of experience. Thus these two classifications correspond to the classical secure versus insecure attachment categories. Alternatively, Magai et al. (2000) offer the concept of em otional organizations. They define emotional organizations as biases and traits included in personality constructs that guide expression, attention, perception, and information processing strategies. Emotional organizations are thought to be formed as par ts of internal working models and to constitute a â€œpredisposition to express or inhibit certain kinds of emotion, strategies for interpreting social stimuli, and styles of adult relatednessâ€ (p. 301). Kobak et al. (1993) postulate that â€œattachment theory p rovides an account of how patterns of parent infant interaction become a self regulatory feature of the childâ€™s personalityâ€ (p. 231). Through these interactions the child forms internal working models of self and the other. â€œDespite the enormous develop mental change, attachment remains a goal corrected system that continually monitors the availability of the caregiverâ€ (p. 231). When the caregiverâ€™s availability is questionable the child employs a variety of strategies to reestablish the caregiver avail ability. Attachment researchers claim that when these primary strategies (Main, 1990) help succeed in reestablishing caregiverâ€™s availability, the child diverts attention to other activities or needs such as playing with peers or toys. In other words, wh en the proximity seeking behaviors
31 (crying, calling the caregiver etc.) manage to attain caregiverâ€™s availability then the comfort of having this availability is used as a secure base to explore the environment. On the other hand, when these strategies do not yield to availability and responsiveness, then â€œthe child must not only continue to monitor the attachment figureâ€™s availability but must also develop alternative strategies for regulating their own attachment behaviorâ€ (Kobak et al., 1993, p. 232). These alternative behaviors are referred to as secondary strategies (Main, 1990). Kobak et al. (1993) state that depending on the childâ€™s internal working models of the caregiver, these strategies consist of either deactivation or hyperactivation of the a ttachment system. These authors postulate that if the caregiver is perceived as rejecting, infants use deactivation strategies to â€œminimize conflict with the caregiverâ€ whereas adults use them â€œto divert attention from attachment topics by restricting acc ess to attachment memories, idealizing parents, or devaluing attachment relationshipsâ€ (p. 232). These strategies have to do with diverting oneâ€™s attention from attachment figures such as being involved with exploratory behavior. On the other hand, hyper activation involves limiting exploratory behavior and seeking the caregiver in disappointment. Kobak et al. indicate that the equivalent of these behaviors in adults as measured by the Adult Attachment Interview (AAI, George, Kaplan, & Main, 1985) consist s of â€œdiscourse incoherence characterized by excessive and irrelevant information, difficulty maintaining a clear sense of the discourse context, and anger that interferes with achieving a balanced perspective on self and parentsâ€ (p. 232). It is therefor e inferred that there are significant differences between affect regulation strategies of persons with mainly primary or secondary strategies. For instance, persons with deactivating strategies minimize access to affective information and experience
32 part icularly in the context of interpersonal intimacy. On the other hand, persons with hyperactivating strategies have excessive experiencing and expression of emotionality. As mentioned above, by providing templates for interpretation of experience, internal working models are adaptive tools enabling a sense of continuity in oneâ€™s perception of self, other and social relationships. Main (1990) asserts that some children with insensitive parents might develop conditional or secondary attachment strategies to maintain this sense of continuity in proximity and self organization. She further notes that such strategies are defensive in nature. Main (1990) recommends that coupled with childrenâ€™s relative limitations in cognitive capacities, the unconscious nature of these strategies makes it difficult to be easily accessible for revision. Whether insecure or secure, internal working models are unconscious and resistant to change. Rosenstein and Horowitz (1996) note that since internal working models guide oneâ€™s expectations and interpretations, individuals form relationships that are consistent with their existing internal working models. The authors state that such â€œdefensively biased multiple models form the initial defensive structures that can lead to such d efensive distortions in personality and psychopathology (p. 245). Morelli and Tronick (1991) recommend a strategic model as an alternative to attachment models developed by Bowlby, Ainsworth and others. They view normal human development â€œas occurring th rough an interaction of mutually regulatory behavioral strategies flexibly developed by children and care givers in the service of achieving short and long term goals (p. 41). The authors refer to childrenâ€™s strategies as resource acquisition strategies a nd to those of caregivers as investment strategies.
33 Morelli and Tronick (1991) use the term resource in both material and psychological senses. According to this perspective caregivers strategies have three main purposes, which are claimed to be univer sal; â€œinfant survival and eventual reproduction; economic self reliance; and enculturationâ€ (p. 41). Stability of Attachment Attachment theorists claim that attachment is formed during the first years of life and is resistant to change in later years. Th ere has been strong interest in testing this hypothesis. Finding empirically answers to this claim is important for two main reasons. On the one hand, as pointed out by Shaver and Norman (1995) if such stability did not exist there would not be much res earch interest in attachment theory. That is to say, if oneâ€™s attachment style was easily changeable over time, then it wouldnâ€™t be a stable enough concept to be captured in empirical inquiries, nor would it have any substantial implications. On the other hand, if oneâ€™s attachment style were not changeable, then the theory would not have much clinical or practical value, for there would not be any sense in attempting to modify some unchangeable construct. There is evidence supporting both the stability an d malleability of attachment. Reviewing studies on attachment continuity Shaver and Norman (1995) concluded that oneâ€™s attachment style â€œis quite stable but not precisely measuredâ€ (p. 494). Furthermore they recommend that assessing adult attachment migh t have a great deal of relevance to the individualsâ€™ present relational circumstances. Consequently, the relevance of current relational circumstances might pose doubt on the reliability of self report measures of adult attachment. Research on stability of attachment has confirmed Bowlbyâ€™s proposition that attachment is relatively stable yet modifiable (Lopez & Gormley, 2002). Moreover, such
34 stability is found regardless of the assessment method used. Indeed, Lopez, and Gormley (2002) report studies (i. e. Baldwin & Fehr, 1995) finding only 30% of individuals who displayed change in attachment orientation over various time intervals. This change was not due to lack of reliability of the assessment methods used. These findings are important in that if at tachment orientations were unchangeable attachment theory would not have much practical value. That is to say, such relative flexibility in attachment orientation over time does give room for clinical interventions. Although the more reliable ways of look ing into attachment stability should involve longitudinal studies, inquiries with cross sectional methodologies also reveal valuable information. One of the frequently reported ways of testing attachment stability involves comparing percentages of infants and adults in each attachment category. Such a method is based on the premise that if the percentages are similar to statistically significant degrees, then such similarity implies continuity in individualsâ€™ attachment orientations. Hazan and Shaver (198 7) developed a questionnaire to measure adult attachment styles. They used the three classifications used by Ainsworth et al. (1978); secure, avoidant and anxious ambivalent. The breakdown of their participantsâ€™ attachment styles was as follows; 55 % se cure, 25 % avoidant, and 20 % anxious ambivalent. This distribution was similar with those of infant studies. Furthermore, their findings were supported by other studies with adults both in the United States and in several industrialized countries (Shave r & Clark, 1996). Such studies did not only show similar percentages of attachment styles but also supported some essential tenets of attachment theory.
35 For instance, the participants who were classified as secure reported positive views of themselves a nd of their partners. Criticism of Attachment Theory Morelli and Tronick (1991) criticize attachment theory for attempting to identify a universal optimum of human development. They furthermore, propose that it is not feasible to identify a universal pro cess for social and emotional development. Infant caregiver relationship cannot be seen as the sole determining factor in human development. For instance, the mutuality embedded in infant caregiver interactions must be taken into account. Also, due to t he complex nature of the contexts in which the development of human babies takes place, there are various other factors to be considered. For example, individuals other than the caretaker might also influence development. Indeed, there are communities in which caretaking is done by multiple individuals. Morelli & Tronick (1991) also raise some concerns about the Strange Situation procedure (Ainsworth et al., 1978). They point out that there have been various crosscultural studies with non US populations that found differing proportions of infants falling into the original three attachment categories. Considering culture specific child rearing practices, differing values about independence, self reliance, and meanings attributed to childrenâ€™s interactions with strangers, these findings are not surprising. Perhaps it would better encompass the complexity of the conditions in which human babies are being raised if developmental models such as attachment theory were to assume an ecological contextual view in identifying the safety, desirability and growth adaptability fostering of a given developmental environment.
36 This would not necessarily undermine the crucial role of those who are in close contact with infants, including the primary caregiver(s). As men tioned before, among other issues the scope of attachment has generated a great deal of discussion. For instance, although they acknowledge that attachment patterns might impact various features of social behaviors, Bartholomew and Thomson (1995) specific ally emphasize that attachment is â€œonly a key aspect of relational behavior in a narrow subset of relationships, notably parent child relationships and long term sexual relationshipsâ€ (p. 485). Thus the authors caution against â€œthe indiscriminate extensio nâ€ of attachment concepts to all social relationships. Some researchers have criticized attachment theory for proposing a deterministic perspective on human development (Morelli & Tronick, 1991). Such analysis views attachment theoryâ€™s emphasis on the imp act of early development as faulty. However, attachment theorists such as Bowlby (1988) view impacts of early experiences from a developmental pathway model. A pathway model does give an essential role to the relative stability of internal working models once they are developed. In the mean time, such a perspective provides sufficient room for modification and alteration of internal working models. Accordingly, such a model acknowledges that through greater self understanding, interpersonal relationship s, and many other factors some individuals with insecure internal working models could develop more adaptive functioning, whereas some with secure internal working models could demonstrate significant adjustment difficulties (Kenny & Rice, 1995). A number of researchers have studied the degree to which attachment theoryâ€™s basic tenets apply to other cultures. Cultural groups vary in the extent to which they
37 emphasize certain emotional expressions and behaviors as favorable and others to be avoided (Harwood et al., 1995). Such cultural meaning systems function as conceptual frame of references with which individuals interpret their emotional experiences. For instance, in Turkish culture being nice and well mannered is a highlighted construct. On the other hand, in American culture self sufficiency is highly valued. Therefore, individuals from these two cultures would attribute different degrees of importance to personal adequacy versus their own public images. It is essential to note that such frameworks are not merely guides for individuals to interpret and attribute meaning to their experiences and behaviors but they also function in the same manner in perceiving others. Classification of Attachment Orientation The original categorization of attachment patterns originated from laboratory observations of infants and their caregivers by Mary Ainsworth and her colleagues (1978). This observational arrangement, also known as the Strange Situation (Ainsworth et al., 1978), was a standardized procedure consis ted of the following stages; 1. Mother and infant enter an unfamiliar room 2. Mother sits, letting the baby explore (3 minutes) 3. An adult stranger enters the room, sits down, talks to the mother and then tries to engage the infant in play (3 minutes) 4. Mother e xits the room, leaving the baby with the stranger (up to 3 minutes) 5. First reunion; mother returns, stranger leaves, tries to settle the baby if necessary then withdraw to her chair (3 minutes) 6. Mother exits the room, leaving the infant alone (3 minutes) 7. The stranger returns to the room, tries to settle the baby if necessary, then withdraws to his/her chair (up to 3 minutes)
38 8. Second reunion, mother returns and stranger leaves. Mother settles the infant and tries to withdraw to her chair (3 minutes). Ba sed on the infantsâ€™ behaviors in this procedure these researchers identified three patters on attachment to the caregiver. They confirmed their laboratory findings with observations at these participantsâ€™ home settings (Goldberg, Muir, & Kerr, 1995). The y found that there was a great degree of parallel between infantsâ€™ behaviors in these two settings. From the Strange Situation procedure, Ainsworth et al. (1978) identified three categories of attachment, namely, secure (type B); anxious ambivalent (type C ); and avoidant (type A). They differentiated these types based on babiesâ€™ behaviors indicative of discomfort while caregivers departed and on their attitudes toward the caregivers upon reunion. The researchers observed that the caregiversâ€™ behaviors wer e complimentary to those of the infants. Secure infants showed moderate degrees of seeking closeness with their caregivers. They were upset when the caregivers left the room and greeted them positively upon their return. The anxious ambivalent infants s ought and rejected closeness. They were strongly upset when their caregivers left the room and were not easily consoled. These babiesâ€™ caregivers also displayed inconsistent behaviors. Their attitudes ranged from being overly responsive to rejecting at different times of the observations. Finally, Ainsworth et al. (1978) observed that the avoidant infants showed indifference when their caregivers left the room. Similarly, they avoided their caregivers when they returned to the room. Similar lack of re sponsiveness was found in the caregivers behaviors toward the infant during the play, which was a part of the Strange Situation.
39 Through their work Main and Solomon (1986) identified some infants who did not fit in any of the original three categories. T hey labeled this attachment category as disorganized type (type D) (Flanagan, 1999). These infants were characterized with the lack of any organized pattern of behavior upon the departure or reunion with their mothers. Flanagan (1999) asserts that such i nfant behaviors are associated with abuse or chronic depression in mothers. As mentioned earlier, Bowlby (1980) suggested that attachment patterns are formed as a result of infantsâ€™ efforts to seek and maintain parental access. Main (1990) referred to the se efforts as primary strategies, which are analogous to secure attachment orientation. She proposed that when such strategies donâ€™t yield to favorable outcomes, infants form secondary strategies. There are two secondary strategies: deactivating and hype ractivating strategies. These strategies coincide with dismissing and preoccupied attachment styles respectively. There have been attachment models developed specifically for adult attachment. Some of these models have had significant impact on adult att achment research (Bartholomew, & Horowitz, 1991; Brennan, Clark, & Shaver, 1998; Fraley et al., 2000; Hazan and Shaver, 1987). Relying on continuity of internal working models Hazan and Shaver (1987) adapted the three infant attachment organizations (Ains worth et al., 1978) to adult romantic love. They summarized each relational orientation in a paragraph. Participants were to pick the most fitting statement to their relational patterns. Each of these statements corresponded to the original three catego ry model of Ainsworth and colleagues (secure, avoidant, and anxious/ambivalent).
40 Later, Bartholomew and Horowitz (1991) proposed a four category adult attachment model. This categorization is based upon individualsâ€™ internal model of self and other. The se authors dichotomized personsâ€™ models of self and other and came up with four combinations: secure, preoccupied, dismissing, and fearful. Secure persons perceive themselves worthy and lovable and others as generally accepting. Preoccupied persons feel themselves as unworthy but have a highly positive image of others. Thus such persons are thought to be preoccupied with relationships. Dismissing individuals have an overall positive perception of self and a negative perception of others. Accordingly, such individuals try to maintain a certain distance in their relationships in order to protect their perceived sense of independence and invulnerability. Finally, fearful individuals hold a perception of self as unworthy and others as rejecting and untrus tworthy. Hence, these persons would make sure not to get close to others in order to avoid the presumed inevitable rejection. MODEL OF SELF (Dependence) Positive (Low) Negative (High) MODEL OF OTHER Positive (Low) (Avoidance) Cell I SECURE Comfortable with intimacy & autonomy Cell II PREOCCUPIED Preoccupied with relationships Negative (High) (Anxiety) Cell IV DISMISSING Dismissing of intimacy, & counter dependent Cell III FEARFUL Fearful of intimacy & socially avoidant Table re printed with the permission of Bartholomew & Horowitz (1991, p. 227)
41 During the last two decades various models of attachment and its measurement have emerged as a result of the immense growth in attachment research. This diversity has been highly fruitful. On t he one hand, it has caused confusion. New researchers in the area of attachment were faced with many options in terms of conceptualization and measurement of attachment. Brennan, et al. (1998) developed a new model and a corresponding measurement instrum ent in an effort to overcome this confusion. This effort was to promote theoretical clarity and measurement precision. Part of the need for such clarity and precision was that some authors had identified individuals who didnâ€™t quite fit any of the catego ries. Such persons had presented a rather mixed orientation that was a combination of two categories (Bartholomew & Horowitz, 1991; Crittenden, 1988; Main & Solomon, 1986). Thus examining the original data by Ainsworth et al.â€™s (1978), Brennan et al. (19 98) identified two discriminant functions; anxiety and avoidance. It must be noted however that this was not the first attempt in identifying underlying factors or dimensions of attachment orientations. As mentioned above the model by Bartholomew and Hor owitz (1991) also used a two dimensional model with which they came up with their four categories (model of self; model of other). Indeed, Brennan et al. (1998) acknowledge that a negative model of self is associated with anxiety, while a negative model o f other is related to avoidance. These two models are relevant but not identical. One of the important differences is that the model by Bartholomew, and Horowitz (1991) identifies attachment categories, whereas the model by Brennan et al. (1998) identifi ed scores on continues subscales of anxiety and avoidance. This model was improved by Fraley et al. (2000) who utilized the Item Response Theory in formulating and validating their model.
42 Similar with Brennan et al. (1998), these researchers offer model of attachment which measures attachment orientation on continues dimensions of anxiety and avoidance in close romantic relationships. Research on Attachment Attachment research originated from inquiries on infant caretaker relationships. After mid 198 0s two different lines of research emerged: (a) studies focusing on attachment stylesâ€™ of adults in predicting their infantsâ€™ attachment behaviors in laboratory conditions (i.e. the Strange Situation), and (b) studies focusing on attachment styles and thei r relevance to adult behavior in the context of romantic relationships (Shaver & Clark, 1996). A significant body of research has illustrated that attachment patterns observed in infancy significantly predict a childâ€™s later social behaviors with peers, fa mily members, teachers and so on. Shaver and Clark (1996) provide a detailed account of the research done with the three category model of Hazan, and Shaver (1987). Adults with avoidant attachment described their parents as rejecting and nonaffectionate, reported having poor relationships with their parents during college years, (Hazan & Hutt, 1993; Rothbard & Shaver, 1994). Compared to individuals with other categories of attachment, these persons were more likely to have a parent with an alcohol proble m (Brennan, Shaver, & Tobey, 1991). Furthermore, individuals with avoidant style expressed a lack of interest in developing intimate relationships (Shaver & Brennan, 1992); were pessimistic about having long term relationships (Hazan & Shaver, 1987); comp ared to the secure ones they were more likely to have break ups; (Hazan & Shaver, 1987; Shaver & Brennan, 1992) and were less likely to grieve after a break up; (Simpson, 1990). In their work environments, avoidant individuals preferred to work alone and used work to avoid close
43 relationships (Hazan & Shaver, 1990). Finally, these persons were more likely than the other two groups to be agnostic or to view God as controlling and distant (Kirkpatrick & Shaver, 1992). Shaver and Clark (1996) summarize curren t research findings on adults who were classified as anxious ambivalent as the following: These individuals referred to their parents as intrusive and unfair (Hazan & Shaver, 1987); they expressed longing for romantic relationships (Hazan & Hutt, 1993); s howed more obsessive behavior toward their partners and suffered from extreme jealousy, were argumentative, intrusive and overly controlling (Hazan & Shaver, 1987). These persons were also more likely than the other two groups to have break ups and to get back together with the same partner (Kirkpatrick & Hazan, 1994). They preferred working with others but felt unappreciated and misunderstood (Hazan & Shaver, 1987). Interestingly, religious anxious ambivalent individuals reported having experiences of s peaking in tongues than the other two groups (Kirkpatrick & Shaver, 1992). Regarding research findings of the secure group, Shaver and Clark (1996) report the following: Such individuals describe their parents in favorable ways and maintain positive relati onships with them during their college years (Hazan & Shaver, 1987). In addition, these persons are highly invested in their relationships (Collins & Read, 1990); they tend to choose conflict resolution strategies that are satisfactory to them and to othe rs (Pistole, 1989a). They were more likely to have long lasting stable and mutually satisfying relationships (Hazan & Shaver, 1987; Kirkpatrick & Shaver, 1994). In their work environment they feel that they are liked by coworkers (Hazan & Shaver, 1987).
44 Finally, they appear to adopt their parentsâ€™ religious belief systems and conceive God in positive terms. Some researchers have examined intergenerational transmission of attachment. Main (1995) looked into the relationship between parentsâ€™ early attac hment experiences and their infantsâ€™ attachment behavior in the Strange Situation. She predicted that the more secure the parent the more secure the child would be. However, this assumption was not supported by her data. The crucial element was not whet her the parent was nurtured or deprived but rather â€œthe degree of coherence versus incoherence in the parentâ€™s subsequent memory of her childhoodâ€ (Mitchell, 2000, p. 85). Hence, the content of early experience was not as important in impacting the future parenting behavior as was the parentâ€™s narrative organization of the actual content. Slade (1997) postulates that such findings signify the importance of â€œlistening for coherence, for the capacity to reflect upon and make sense of internal experience, an d to listen for incoherence, of meaninglessness, of disorganization, and of disintegration, for moments when meaning cannot be madeâ€ (In Mitchell, 2000, p. 86). It is essential to investigate factors beyond the relationship between the primary caregiver an d the infant. There is a need for inclusion of various contextual factors in attachment research. In such a study with Israeli college students, Mikulincer and Florian (1999) looked at the correspondence between parents and their offspringsâ€™ attachment s tyles. The researchers found that parentsâ€™ and childrenâ€™s attachment styles did in fact match. However, this match was more significant with gender matching. In other words, such a correspondence was more visible between same sex parents children dyads. Such finding is partially challenging to assumptions of traditional attachment theory.
45 Attachment in Non Human Primates Since the development of attachment theory, many researchers have seen attachment as a product of evolutionary survival and as univers al. Studies with different cultures as well as with species closer to humans could test validity of this proposition. Rosenblum and Paully (1991) looked at nonhuman primates attachment behaviors. They specifically focused on the primatesâ€™ utilization of environment in relation to the nature of their dyadic relationships with their mothers. The findings were consistent with the Yerkes Dodson Law of arousal and learning (Yerkes & Dodson, 1908). In other words, the researchers found that when infants were forced into breaking contact with their mothers, they demonstrated diminished capacity to recognize certain characteristics of their environments and appropriately respond and adapt. The infantsâ€™ poor performance is attributed to the high level of anxiet y and distress felt due to forced separation from their mothers. Moreover, when they had time on their own and were having to deal with their environment upon maternal rejection, the infants again displayed difficulty in coping. One the other hand, the p rimate infants demonstrated high degrees of adaptive behavior when they monitored their separations based upon their own affective state and moved away from an available and contingently responsive mother. Accordingly, Rosenblum and Paully (1991) conclude d that it was permissive but responsive mothers who fostered such separation behavior that enabled the infants to effectively cope with their environments on their own.
46 Attachment in Non Technological/Indigenous Communities In their work, Morelli & Tron ick (1991) studied social emotional development and attachment process in 4 to 12 month old children from a technologically simple culture the Efe (Pygmies) of Ituri forest of Northeastern Zaire. They chose this specific culture because of the way of life of Efe who were hunters and gatherers â€œa way of life argued to strongly resemble the niche of human adaptednessâ€ (p. 45). The observations of infants of this age group confirmed the claim of attachment theorists that at these early times infants tend to prefer one caregiver as the basis for attachment formation. Hence the researchers noted that sufficient understanding of attachment process could be acquired when attachment is conceptualized as a process that â€œis supported by and grows out of the interp lay of factors that are infant, mother and culture basedâ€ (p. 49). As illustrated by the above studies there has been a great deal of effort to diversify attachment research. However, the existing literature has various measurement issues as well as metho dological ones. Attachment research continues to utilize correlational methodologies as well as more complex ones such as path analysis. There appears to be an increase in the number of studies dealing with cause effect relationships. This increase has become more visible particularly in last several years. A great portion of adult attachment research focuses on college students. Attachment Theory and College Student Mental health There are various aspects of attachment theory that can be applied to c ollege student development and mental health. According to attachment theory, close interpersonal relationships are important in various phases through the life span. Such bonds are thought to be even more crucial in transitional times, such as the adole scence
47 and young adulthood. Attachment theory is a theory of adaptation and adjustment, which are also major themes of college experience. Furthermore, attachment is relevant to college student development due to its emphasis on relationality and its rel evance to identity formation. Along with these characteristics of attachment theory below are some ways in which college studentsâ€™ experience has been viewed in lights of attachment concepts. College experience is often seen as a vital time of separation individuation. From an attachment theory stand point, leaving home for college very much resembles the Strange Situation (Ainsworth et al., 1978). Consequently, it is a crucible for separation individuation and, thus for identity formation. That is to s ay, leaving home is an act toward increasing separation and, thus individuation. Group for the Advancement of Psychiatry (1990) views college experience as a modern rite of passage. College students go through a process of self definition and strive to â€œ forge a sense of identity in the context of their previous experiences, current relationships and future expectationsâ€ (Kenny & Rice, 1995, p. 433). Such conceptualization is inline with traditional developmental theories (i.e. Eriksonian theory). Such mo dels have been known with their emphasis on separation and individuation. On the other hand, attachment theory â€œhighlights the adaptive value of supportive and interdependent relationships throughout the life span and especially during periods of stress, such as the transition to collegeâ€ (Kenny & Rice, 1995, p. 435). Different developmental and personality theories have had varying emphasis on identity formation or relatedness. There have been various theoretical orientations that acknowledge these two processes as central to personality development (Blatt & Blass,
48 1996). For instance, Bowen (1978) used the concept of differentiation; Klein (1976) used the term â€œwe nessâ€; Gilligan (1982) puts it as â€œself in relationâ€; to Sampson (1988) recommended â€œense mbled individualismâ€ to convey the dialectical unity of these two vital processes. Those supportive of attachment theory find it capable of capturing these two processes. Various concepts of attachment theory have been applied to young adultsâ€™ college exp erience. Parental availability has been one of these concepts. Attachment theory looks at this availability based on the notion of secure base, which can foster young adultsâ€™ risk taking behavior and exploration in various areas of their lives, such as a cademic, social, and personal. Attachment theorists also see this secure base to have indirect contributions to oneâ€™s development and experience in college. Such contributions are thought to be through internal working models of self and other. Attachme nt theory (Bowlby, 1988) proposes that when attachment figure has been available, responsive and reliable the individual will form an image/representation of self as good, worthy, and lovable. It also claims that such a person will have an internal model of other as trustworthy, responsive and reliable. Thus it is assumed that such individuals will be more likely to establish and utilize supportive interpersonal relations. One of the key factors in applying concepts of attachment theory in understanding adult behavior or more specifically the behaviors of young adult college students would be the construct of secure base. The beginning of college experience of traditional college students is one of the significant experiences of separation. From an atta chment theory stand point this experience could be seen in light of whether the young college student perceives the home left behind as a secure base, which he or she could utilize in
49 order to master a new and challenging environment. In addition, the ext ent to which such perception is in existence could make it more likely for the young person to form new relationships that can also function as secure base in their new environments. The connection between internal working models and psychological well bei ng is established with the assumption that internal working models influence ways in which one perceives subsequent life events. Thus it is thought that a person with negative internal working models would have a predisposition to perceive negative events as his or her failure, which would in turn perpetuate distress. Conversely, based on these assumptions Kenny and Rice (1995) postulate that college students who posses positive internal working models will be less vulnerable to experiencing possible situ ational stressors (i. e. rejection by peers or receiving undesirable grades) in ways that would be disempowering to the self. The authors go further to claim that those who anticipate others as trustworthy may be better equipped to form supportive relatio nships with peers and faculty. Therefore, it could be inferred that students with positive internal working models would be advantageous in obtaining, and maintaining social support, which is an important resource that aids to coping strategies. In additi on, college students use internal working models of self and others â€œas filters through which self perceptions are developed and level of social support from others is judgedâ€ (Rice & Cummings, 1996, p. 51). These perceptions of self and other are crucial since the college experience is a period of identity formation (Chickering & Reisser, 1993). That is to say, the quality of such schemas will impact oneâ€™s sense of self. How college students perceive themselves has a great deal of impact on how they fun ction as well as on how they build their futures.
50 Petersen et al. (1991) articulated a model of mental health trajectories that resembles Bowlbyâ€™s developmental pathway model. This model proposes that ways in which college students adjust to stressful eve nts whether they be situational or developmental â€œis determined by internal and external resourcesâ€ (Kenny & Rice, 1995, p. 437). Examples of internal resources would be self efficacy, intelligence, coping skills, and so on. External resources refer to interpersonal resources, including attachment relationships. According to this model, internal and external resources moderate or buffer the distress experienced in challenging situations. In transitional periods such as late adolescence (college years) how a person copes with such challenges is important for constructive adaptation. Moreover, how such times are experienced will greatly impact the internal and external resources that will be available to the individual in subsequent periods of life. In s ummary, attachment concepts have been utilized in understanding various aspects of the optimum balance between support and challenge college students seek in dealing with various developmental issues. Some examples of these issues are; identity formation, interpersonal competence, adjustment to a new environment, acquiring academic competence, affect regulation, and the like. Attachment Research and College Student Mental Health Since the beginning of the last decade there has been increasing interest in attachment theory for understanding of late adolescents (Rice & Cummings, 1996). This attention has resulted in a great number of studies. Along with the interest in attachment theory, the emphasis on college students might also be due to their availabil ity as research participants. The following is an extensive review of some of these studies.
51 It is a curious question as to whether attachment or its artifacts lead to healthy adjustment. Empirically supported answers to this question could improve the c linical utility of attachment theory. In their study with college students, Engels et al. (2001) hypothesized that the relationship between parental attachment and emotional adjustment is mediated by social skills and relational competence. The findings revealed a moderate association between parental attachment and social skills. On the other hand, parental attachment and relational competence were found to predict adolescentsâ€™ emotional adjustments. Further studies are needed to explain as to what spe cific components of attachment lead to adaptive competencies. College students often seek counseling services for complaints about their romantic relationships (Creasey, Kershaw, & Boston, 1999). Studies have found associations between late adolescent att achment difficulties with a range of issues in romantic relationships (Bartholomew & Horowitz, 1991; Brennan & Shaver, 1995; Creasey & Hesson McInnis, 2001; Feeney & Noller, 1990; Hazan & Shaver, 1987). More specifically, studies have confirmed that insec ure attachment is associated with a variety of issues in interpersonal relationships, including romantic ones. In their work, Creasey and Hesson McInnis (2001) used path analysis to identify associations between attachment orientation and conflict managem ent strategies in steady relationships. They found that students with insecure attachment orientations were more likely to report greater negative affect during disagreements, and less confidence in coping with such disputes. However, similarities betwee n anxious and avoidant groupsâ€™ conflict management strategies were questionable. In other words, these authors concluded that â€œavoidant and anxious adolescents may select conflict management tactics by different
52 affective cognitive routes or trajectoriesâ€ (p. 92). In addition, contrary to their expectations, the authors did not find significant gender differences in negative or positive conflict tactics. This study supported the hypothesis that secure attachment has favorable impacts on conflict manageme nt strategies in interpersonal relationships. However, it also confirmed the need for further investigation of the precise cognitive affective components of differences in conflict management strategies as they relate to attachment orientation. In their e xtensive review of literature on college students, Rice and Cummings (1996) conclude that secure attachment is associated with personal and interpersonal well being, whereas, insecure (avoidant and anxious/ambivalent) attachment is correlated with issues s uch as depression, eating disorders, lower self esteem, and lack of assertiveness. However, the researchers note that studies with such results had only moderate effect size. Thus they hypothesized that looking into the degree of correspondence between p arents and childrenâ€™s perceptions of their bond could yield to better understanding of the unaccounted variance in the pervious studies. Attachment and Adjustment to College Hazan and Hutt (1993) studied new college studentsâ€™ transition from home to colleg e. They found significant differences between secure and insecure attachment groups in seeking and utilizing social support. Secure students reported negative affect for having left home. They had frequent contact with home. These contacts were initiat ed by the students, their families and friends. These students found such efforts of family and friends as supportive. Compared to others, these students were most likely to be visited by parents. Finally, after the first semester they were functioning at their pre -
53 college levels with few visits, if any, to the student health clinic. On the other hand, students with insecure attachment indicated lower levels of functioning prior to college. They reported feeling better upon leaving home. The persons i n anxious subgroup felt less lonely and depressed, whereas, the avoidant participants felt lonely and depressed but with less anxiety and fewer physical symptoms. During the first semester individuals with anxious style had frequent contacts with home. T hey initiated the majority of these phone calls and the phone conversations were usually ended by friends and family members. Students with avoidant style had the fewest number of contacts with home. They mostly initiated and ended these contacts. Compa red to the secure group anxious and avoidant students were less likely to be visited by their family members. In their second semesters anxious students reported feeling worse than they did during the first semester. These students were lonelier than the secure and avoidant ones and more likely to have visited the student health clinic. Finally, Hazan and Hutt (1993) found the avoidant group functioning better than anxious ones and worse than secure ones in their second semesters. The findings of these researchers confirmed that although all the groups sought support, only the secure group benefited from it. Shaver and Norman (1995) indicate that such findings also illustrate the coexistence of secure internal working models and a security sustaining so cial support network. This co occurrence is consistent with attachment theoryâ€™s hypothesis that secure attachment can be instrumental in developing and maintaining healthy interpersonal relationships. Findings of Davis, Morris, and Kraus (1998) also supp orted this hypothesis.
54 Kenny and Perez (1996) propose that students of ethnic and racial minorities would have additional challenges in adjusting to college because they are more likely to experience the environment as alienating and lacking support. Th ese authors indicate that due to their collectivist cultural backgrounds students of color are more likely to rely on support from their extended families as they go through stressful experiences, such as adjusting to college. They add that combined with their minority status, these students are more likely to come from economically deprived working class backgrounds, which might further complicate their adaptation. In their work Kenny and Perez (1996) examined the relationship between attachment status a nd psychological well being of a multiethnic and economically disadvantaged sample of first year college students. They found that attachment security was associated with psychological well being and college adjustment. One of the unique findings of this study was that along with security of attachment, whether the students had a perceived encouragement of their families toward autonomy or not was also found to be a significant factor in their psychological well being and adjustment to college. In conclus ion, there is increasing empirical interest in attachment theory and its application to young adult college students. Although still barely reaching at applied level, the current body of research has in fact revealed a great deal of knowledge in understan ding college experience from an attachment point of view. Perhaps more important than whether attachment theory could be utilized as a metaperspective (Lopez, 1995) or not is the degree of research it has stimulated in the area of college student mental h ealth.
55 Personal Meaning Various researchers have examined the link between attachment status and a range of indicators of well being. These studies have focused on attachmentâ€™s relation to social and relational competence (Engels, Finkenauer, Dekovic, & Meeus, 2001); self esteem (Rice & Cummings, 1996); self efficacy (Rice & Cummings, 1996); autonomy, competence and relatedness (La Guardia et al., 2000); health and illness (Feeney, 2000); psychological health of partners in long term marriage or cohabitin g relationships (Kotler & Omodei, 1988); dynamics of client therapist relationship (Delvey, 1985; Pistole, 1989b, 1999); lack of symptomatology (Kenny, Brabeck, & Fife, 1998; Raja, McGee, & Stanton, 1992); career development (Bluestein, Prezioso, & Schulth esis, 1995); and self worth (Kenny et al., 1998). These studies found moderate to strong associations between secure attachment and measures of well being. Along with the research tradition investigating development of psychopathological symptomatology, there has been a line of research examining development of adaptive mechanisms. This research was in part inspired by traumatic events such as the experience of German concentration camps during the World War II. Researchers intended to find out as to ho w some persons manage to survive through such tragic experiences. Victor Frankl who was a survivor of these camps was a major figure impacting this school of thought. However, such inquiries date before Frankl. For instance, the German philosopher Fried rich Nietzscheâ€™s writings influenced scholars like Frankl in their quest toward understanding what enables individuals to endure seemingly unbearable conditions of living. The decades following the war witnessed great interest in finding out the buffering effects against high stress situations.
56 Frankl (1963) saw meaning as one of these buffers. In his book Manâ€™s Search for Meaning he summarizes this with a quote from Nietzsche: â€œhe who has a why to live for can bear with almost any howâ€ (p. 121). For pra ctical purposes the term meaning is often used to refer to purposefulness. It appears that what many scholars meant (i.e. Friedrich Nietzsche and Victor Frankl) with the term meaning rather had an ontological emphasis. That is to say the concept did not merely address the extent to which one feels or act in goal oriented manners. The term appears to be used to refer to oneâ€™s ongoing quest for feeling a global sense of worthiness in being in this world, which exceeds the sum of his or her acts, aspiration s and so on. This point is summarized by Wong (1998) as follows; â€œmeaning seeking depends not only on what a person thinks or does, but also on who the person isâ€ (p. 113). This view of the term meaning appears to be an essential feature of the self actu alizing person Maslow (1968). A close look at how Maslow conceptualized the concept of self actualization does not only reveal the presence of a deep sense of meaning in oneâ€™s life but also an openness to experiencing life. Such openness might be closely related to attachment theoryâ€™s concept of exploration, which involves seeking new experiences in the presence of perceived availability of a secure base to return to in the face danger or distress. Empirical observations of Jean Piaget (1972) and evoluti onary scholars have illustrated that similar with other species, humans are inclined to actualize or put in practice their potentialities. Perhaps factors such as attachment security are important so as to allow or facilitate such actualization of oneâ€™s c apacity. Klinger (1998) provides linguistic etymological, philosophical and psychological analyses of the term meaning. These analyses indicate that in many languages the words
57 corresponding to the term meaning refer to aim, goal, and intentionality. It is referred to as a high cognitive entity with goal directed, attributional, motivational, and integrative qualities. In other words, meaning is closely associated with the goal directed nature of human behavior, thus it is motivational. As such, it is t hought of as a construct that connects related mental representations therefore, having an integrative function. Klinger (1998) notes that without such function human behavior would be merely guided by instincts or impulses. In short, meaning making is a n evolutionary gain of the human species. Klinger (1998) furthermore claims that the human brain cannot sustain purposeless life. He postulates that meaningfulness is an inevitable need that is vital for human functioning and survival. What happens when there is a lack of meaningfulness? According to Klinger (1998) prolonged lack of meaning and goals would inevitably lead to psychopathology. Klinger further claims that having aversive goals would similarly lead to pathology. There could be direct and or indirect connection between lack of meaning and behavioral patterns that are self destructive. In the absence of goals and purposes one would resort to other means to improve a sense of well being (Klinger, 1998). One of the issues often associated wi th a lack of goals and purpose is depression, which is often characterized with excessive self criticism. Similarly, absence of meaning or presence of predominantly aversive goals could lead to prolonged negative affective states such as stress and anxiet y that can constitute risk to oneâ€™s physical well being. Klinger (1998) introduces empirical studies reporting associations between perceived level of personal meaning and various indicators of well being and psychopathological symptoms. Some of these ind icators are as the following: global
58 sense of well being, substance abuse related issues, and depression. However, these reports indicate a great need for studies focusing on personal meaning and various mental health or physical health related variables. There are no empirical studies dealing specifically with attachment and personal meaning. The only study available is a dissertation study (Rood, 2000) examining the relationship between a sense of coherence and attachment, which involves meaningfulness or personal meaning as a component of coherence. Similarly, relations between depression or anxiety and personal meaning have only been studied as a part of the concept of sense of coherence (Rood, 2000; Sandell, Blomberg, & Lazaar, 1998). The current re search is aimed at looking into interrelations among personal meaning, attachment status, gender, depressiveness, and trait anxiety. Attachment and Psychopathology There have been many studies investigating adult attachment and psychopathology (Allen et al ., 1996; Bartholomew & Horowitz, 1991; Cummings & Davies, 1996; Dozier, 1990; Feeney & Noller, 1990; Hazan & Shaver, 1987; Kobak & Hazan, 1991; Magai et al., 2000; Main, 1996; Rosenstein & Horowitz, 1996; Pianta et al., 1996). These studies acknowledge as sociations between insecure attachment and psychopathology as well as secure attachment and various indicators of well being (del Carmen & Huffman, 1996; Rosenstein & Horowitz, 1996). On the other hand, as pointed out by Roberts et al. (1996) the underlyi ng mechanisms of these associations are yet to be examined. According to attachment theory, the quality of attachment plays a crucial role in leading to vulnerability to psychopathology (Rosenstein & Horowitz, 1996).
59 Early attachment relationships and th ose that follow in later years play crucial roles in fostering resilience versus vulnerability. It is thought to be so because human life span development involves ongoing tasks and challenges. The outcomes of such tasks are greatly dependent upon oneâ€™s internal and external resources. Such outcomes are important in their own right as well as in impacting later development Insecure attachment has often been associated with psychopathology. Nevertheless, such information might not have much practical val ue unless it reaches some degree of precision. For instance, there is ample reason to question what types of insecure attachment are related to specific psychopathological symptoms. As mentioned before, by providing templates for interpretation of experi ence internal working models are adaptive tools enabling a sense of continuity in oneâ€™s perception of self, the other and social relationships. Main (1990) asserts that some children with insensitive parents might develop conditional or secondary attachme nt strategies to maintain a sense of continuity in proximity to attachment figures and self organization. She further notes that such strategies are defensive in nature. Main (1990) recommends that coupled with childrenâ€™s relative limitations in cognitiv e capacities the unconscious nature of these strategies make it difficult to be easily accessible for revision. Whether insecure or secure, internal working models are unconscious and resistant to change. Rosenstein and Horowitz (1996) believe that since internal working models guide expectations and interpretations, individuals form relationships that are consistent with their existing internal working models. The authors state that such â€œdefensively biased multiple models form the initial defensive str uctures that can lead to such defensive distortions in personality and psychopathology (p. 245).
60 Attachment is often mentioned along with exploratory behavior. Attachment theorists propose the human infant should feel the availability of a responsive and available caregiver in order to safely explore the world around. That is to say, the attachment bond is an important component and prerequisite of exploratory behavior. On the other hand, such behavior is in fact antithetical to attachment (Guidano & Li otti, 1983). Exploration refers to infantâ€™s efforts in searching and learning away from the caregiver. Hence, â€œit implies detachment or active separationâ€ (Guidano & Liotti, 1983, p. 102). Such an interrelation brings out the importance of attachment wi thin the context of human experience and development. Guidano and Liotti (1983) point out that as individuals learn survival skills, any problems in establishing a balance between attachment and exploration could lead to distortions in either or both area s. As a result, in an effort to adopt, the infant may compromise either attachment or exploration. For instance, the infant might be anxious about the availability of the attachment figure that he or she could put great effort in maintaining proximity ra ther than freely exploring the environment. More importantly this early formations are thought to establish patterns that could impair development and lead to pathways to psychopathological symptoms (Guidano & Liotti, 1983). The link between attachment an d maltreatment might appear obvious. However, as Toth and Cicchetti (1996) point out, not all maltreated children have insecure attachment. This might be due to many factors including the presence of more positive relationships during or after maltreatme nt. The authors noted that there is a great deal of empirical evidence suggesting that compared to non maltreated children, maltreated ones are more likely to develop depressive symptoms. Toth and Cicchetti (1996) found that
61 the likelihood of forming ps ychopathological symptoms was most significant among survivors of sexual abuse. Moreover, among sexually abused children those with confused patterns of relatedness reported the highest levels of depressive symptoms. The researchers used relatedness as a n attachment like construct. Patterns of relatedness were significant contributing factors to depressive symptomatology as well as impaired sense of social acceptance. Furthermore, the children with confused patterns of relatedness showed highest levels of depressive symptomatology whereas, children with impaired social acceptance reported disengaged deprived patterns of relatedness. Experiences of maltreatment have significant impacts on internal working models of self and other. Cole and Putman (1992) maintain that such experiences could lead to (a) deviations in the processes of defining, regulating and integrating components of the self; and (b) deviations in perceptions of others as trusting and reliable. Cole and Putman (1992) also note that these deviations interfere with perceiving self as a coherent entity and forming a sense of identity. In addition, issues with affect regulation, impulse control, and self destructive behaviors are commonly encountered issues by childhood sexual abuse survivor s. Finally, such deviations also seem to often lead these individuals to experiencing relationships in ways that maintain the existing perceptions of the self and others. Maltreatment or severely traumatic experience in sensitive developmental times are t hought to often result in dissociation. Liotti (1999) summarizes studies associating disorganized attachment with adult experience of dissociation. Although the findings in this area of research are tentative and mixed, Liotti reported a body of research supporting some developmental continuity between childhood experience of disorganized
62 attachment and adult tendencies of dissociation. He adds that along with the presence or absence of disorganized attachment, the experiences leading to them are also im portant to be taken into account. This notion corresponds to the idea that â€œsuccess in the personal synthesis as a matter of degree within a continuous process of meaning making rather than as an all or none phenomenonâ€ (Liotti, 1999, p. 298). Some attach ment researchers have criticized the existing diagnostic system. Zeanah (1996) asserts that attachment disorders should be conceptualized in relational terms. He criticizes Diagnostic and Statistical Manual of Mental Disorders (4 th ed.; DSM IV; American Psychological Association, 1994) and International Classification of Diseases (10 th ed.; ICD 10; World Health Organization, 1992) for conceptualizing attachment disorders in merely intrapsychic terms. Zeanah also finds the procedure of the Strange Situati on (Ainsworth et al., 1978) problematic by basing attachment on separation and reunion, which he claims to be only two aspects of attachment relationships. Thus Zeanah (1996) cautions that the behaviors in the Strange Situation are not indicative of psych opathology but are rather risk for psychopathology. Furthermore, he contends that â€œattachment problems become psychiatric disorders when emotions and behaviors displayed in attachment relationships are so disturbed as to indicate, or substantially to incre ase the risk of, persistent distress or disability in the infantâ€ (p. 47). In conclusion, Zeanah (1996) endorses the conceptualization of Lieberman and Pawl (1988), who defined attachment issues as disorders of secure base. Hence, Zeanah recommends the fo llowing three patterns of secure base distortions formulated by
63 Lieberman and Pawl (1988): The first, recklessness and accident proneness, described infants who failed to check back with their caregivers at times when their infantsâ€™ attachment system ou ght have been aroused. In the second type, inhibition of exploration, infants seemed unwilling to venture away from the secure base that their caregivers were to provide. Finally, in the third pattern, precocious competence in self protection, infants see med to have inverted the secure base so that they were excessively self reliant and providing care and protection to the parent. (Zeanah, 1996, p. 48) Attachment Status and Depressiveness Susceptibility to depression is often linked to two personality tr aits: dependency and self criticism or perfectionism. Blatt (1974) was the first researcher to formulate this model. Dependency is referred to as a need for love and approval. Whereas self criticism or perfectionism is viewed oneâ€™s concern about achieve ment and success. However, what seems to be problematic is not the mere presence of such needs and tendencies. How these needs lead to vulnerability to depression has rather to do with the extent to which they dominate oneâ€™s personality. Although there i s a great deal of empirical work supporting this model, there have been considerable questions raised about it, particularly from attachment theorists (Whiffen et al., 2000). The underlying motivation for both traits seems to be approval of others. The d ifference between them has to do with the means through which this approval is sought. Whiffen et al. (2000) asserted that the above mentioned model operationalizes dependency and self criticism in ways that conflict with attachment theoryâ€™s emphasis on t he necessity of interpersonal approval seeking as a natural and healthy need. Furthermore, these researchers hypothesized that items of dependency measures assessing neediness would be associated with depressive symptoms more so than connectedness items. Indeed, their findings confirmed that neediness was correlated
64 with problematic interpersonal and intrapersonal functioning more so than connectedness. In addition, the findings indicated that self criticism was more strongly associated with interpersona l functioning than was neediness. Whiffen et al. (2000) reasoned that this is so because self critics are characterized by emotional distance and hostility. Finally, the findings showed that neediness was associated with depressive symptoms for women but not for men. In a similar study Enns, Cox, and Larson (2000) looked into self criticism and dependency in relation to perceived parental practices. They found that overprotection by father was related to self criticism in males, lack of maternal and or p aternal care associated with self criticism in females. Dependence was only related to a lack of paternal care in females. Carnelley et al. (1994) utilized an attachment framework in examining depressive experiences. They noted that attachment theory cou ld facilitate valuable insight into the experience of depression for it integrates individual and interpersonal factors. Such integration is achieved through the concept of internal working models. As mentioned before, these models are cognitive and affe ctive representations of the self and other that result from close relationships at crucial early developmental periods. Carnelley et al. (1994) found depression to be related to insecure internal working models. They also found that compared with nonde pressed ones, female college students who were mildly depressed were more likely to have an insecure attachment style; preoccupied, fearful or avoidant. On the other hand, recovering clinically depressed women were more likely to have a fearful avoidant s tyle but not a preoccupied one when compared with nondepressed ones. Similarly, working with a sample of children, Muris, Mayer, and
65 Meesters (2000) found that insecure attachment was associated with depression. The researchers did not find any significa nt difference between avoidant and ambivalent styles. Adam et al. (1996) examined attachment organization with a clinical sample. Attachment organization was classified based on the following formulation: An autonomous or secure state of mind (F classifi cation) is characterized by good access to memories of early attachment history and ability to describe this experience and its effects coherently. A dismissing state of mind (Ds classification) is characterized by idealization or derogation of the parents in which the individual claims of memory for childhood and emphasizes personal strength or normalcy. A preoccupied state of mind (E classification) is characterized by a focus on relationships with parents in either an angry unobjective manner (E2 subcate gory) or passive, helpless manner (E1 subcategory). A third subcategory (E3) applies to individuals who appear to overwhelmingly preoccupied with trauma (p. 366) The researchers also used an additional category (Ud classification) for those who indicated unresolved unorganized responses to attachment related trauma. The findings of this study showed that unresolved disorganized, and preoccupied classifications were significant predictors of suicidal history in a clinical sample. On the other hand, dismi ssing category was negatively associated with having a history of suicidal behavior. History of suicidal behavior was associated with symptoms of both externalizing (delinquency) and internalizing (depression). In order to examine associations between em otional biases and attachment Magai et al. (2000) used a four category (Bartholomew & Horowitz, 1991) attachment classification system. Their results indicated the preoccupied category of attachment was â€œpredicated by parental love withdrawal, facial disg ust, fantasies of closeness and affiliation, and trait anger and depressionâ€ (p. 301).
66 Rosenstein and Horowitz (1996) hypothesized that â€œcontinuity with the mental organization of attachment throughout the life span and between generations is of central importance in the psychopathology of adolescenceâ€ (p. 245). They expected high correspondence between adolescent and maternal attachment styles. Moreover, the authors hypothesized that there would be associations between certain psychiatric disorders and attachment related mental organizations that lead to ways in which distress is dealt with. Furthermore, these authors envisaged that issues such as conduct disorder, narcissistic personality disorder in which distress in not acknowledged, affect is conta ined, and expression is externally directed would co occur with dismissing attachment style. Conversely, issues such as depressive and anxiety disorders â€œin which distress is acknowledged, modulated, and symptom expression is directed toward the selfâ€ (p. 246) would co occur with preoccupied with preoccupied attachment style. Rosenstein and Horowitz (1996) confirmed both of these hypotheses in their study with a clinical sample. Comparing a clinical group to a nonclinical one, Fagony et al. (1996) found s trong association between psychiatric disorders and unresolved difficulties of early relationships. Similarly, the authors found the narratives of the clinical group to include experiences distinctively beyond those in non clinical group, which they view as suggestive of a causal relationship. Narratives of the psychiatric patientsâ€™ on the Adult Attachment Interview (AAI, George et al., 1985) placed them in preoccupied and unresolved categories. Anxiety was associated with the unresolved classification o f attachment. Furthermore, AAI scales illustrated that individuals with eating disorders held idealized views of parents while depressed individuals held lower idealization of their
67 parents. The researches also found that depressed individuals had high s cores on the anger scale of AAI. They interpret this finding as supportive of the psychoanalytic notion that depression stems from anger directed toward the self. Perfectionism and codependence are often linked to depressiveness. It has always been a par t of human experience to strive for improvement. Some of these efforts can be perfectionistic. Rice and Mirzadeh (2000) worked with a sample of college students. They compared students who used predominantly adaptive to those with maladaptive perfection ism. The researchers investigated if attachment predicted the type of perfectionism among college students. Their results showed that adaptive perfectionists were more likely to report secure attachment. On the other hand maladaptive perfectionists were more likely to experience depression to a clinically significant degree. Accordingly, it could be inferred that secure attachment might facilitate realistic self perception, expectations and aspirations. Furthermore, the above findings confirm that secu re attachment is associated with hope and optimism. Researchers have paid a great deal of attention to the concept of codependence. The issue is frequently mentioned in attachment literature (i.e. Springer, Britt & Schlenker, 1998; Bartholomew & Horowitz, 1991). Springer et al. (1998) studied associations between codependency, relationship quality, and personality characteristics. Their findings demonstrated that codependency was related to lower self esteem and lower perception of interpersonal control. In addition, it was associated with greater self consciousness, social anxiety, and insecure attachment. A unique finding of this study was that codependency was associated with greater ability for empathy and emotional disclosure in relationships. How ever, such ability did not seem to foster increased sense
68 of connectedness, and supportiveness in close relationships. Contrary to common notions about codependence, college students with codependency were found no more caring and supportive to their part ners. These participants in fact reported greater competitiveness with partners. Studies involving attachment and depression have often focused on currently experienced depression. Not all individuals who experience depression at some given times in thei r lives could be seen as having a general vulnerability to depression. Accordingly, clinical work with individuals who do have such vulnerability and interventions with those experiencing depression at a given time would naturally differ. Such difference might not be only in the kinds of therapeutic interventions but also perhaps in their duration. In other words, those with depressiveness as a trait like construct might in fact need longer term psychotherapy. In short, the current study intended to foc us on depressiveness as opposed to currently experienced depression among college students in an attempt to make a unique contribution to attachment literature. Attachment Status and Trait Anxiety The literature on trait anxiety includes a great deal of de bate as to whether it is analogous to the concept of sensitivity to anxiety. Some view the two concepts as similar while others consider them as two related yet distinct constructs. Although a discussion on these two views exceeds the scope of the curren t study, due to the interpersonal focus of attachment theory, it seems reasonable to assume these concepts as similar. One of the basic premises of attachment theory is that internal working models are the means by which attachment behavior and affect are regulated from the beginning of life on. Anxiety has been the most commonly studied emotion related to attachment.
69 In fact, in the original attachment classification formulated by Ainsworth et al. (1978) one attachment style was referred to as anxious a mbivalent. This specific orientation corresponds to the recent preoccupied attachment category, which has been found to be closely associated with the experience of anxiety. Torquati and Vazsonyi (1999) indicate that: relationship specific models of atta chment include expectations about the behavior of the self in the context of a specific relationship and of specific significant others [and that also] internal working models are activated in the context of perceived threat, and therefore, function as an organizational construct for affect, appraisal, coping behavior, and conflict resolution styles. (p. 546) Furthermore, when the attachment system is activated, the secure individuals tend to seek proximity, while insecure ones develop secondary strategie s (Main, 1991) to â€œregulate negative affect and or availability of the attachment figureâ€ (Torquati and Vazsonyi, 1999, p. 547). As mentioned earlier, the terms â€˜secondary strategiesâ€™ and â€˜insecure attachmentâ€™ have been used interchangeably. Of insecure individuals, preoccupied ones are characterized by tendencies to be (a) overly involved in close relationships; (b) heavily dependent upon othersâ€™ acceptance for a sense of well being; and (c) incoherent in disclosing their stories about close relationship s (Kemp & Neimeyer, 1999). More importantly, as pointed out by Kobak and Sceery (1988), preoccupied individuals tend to have mechanisms of hyperactivation of distress, which might be the linking factor between this attachment orientation and the experienc e of anxiety. There is convincing empirical evidence linking preoccupied attachment with anxiety. Kobak and Sceery (1988) worked with a sample of first year college students to examine associations between attachment orientations and affect regulation. T hey found that preoccupied individualsâ€™ self and peer reports indicated higher levels of anxiety than
70 did the secure and dismissing ones. Similarly, Meyers (1998) looked into attachment styles of college students and their relationships with experiences o f affect management and anxiety. He found significant correlations between insecure attachment style and difficulty in affect management and higher reports of anxiety. As noted by Kobak and Sceery (1988) despite the empirical support linking preoccupied attachment and anxiety, little is known about how these individuals cope with such high levels of anxiety. Hence, the authors recommend that perhaps due to (a) preoccupation with close relationships; (b) the urgency with which they seek and experience rel ationships; and (c) their tendencies to be clinging and dependent might interfere with successfully alleviating high anxiety. In other words, it could be inferred from this hypothesis that individuals who have higher levels of trait anxiety and who are pr eoccupied would perhaps enter relationships hoping to alleviate their anxiety since they tend to draw comfort and self acceptance from such relationships. However, this very higher level of anxiety might interfere with such relationships occurring in stab le and satisfying manners. Consequently, such experiences might confirm their early working models and representations of the self and other. Thus the experiential cycle of high anxiety relationship seeking and anxious ambivalent orientation to relations hips would be maintained. Despite the existing body of research reporting a strong link between preoccupied attachment and anxiety, there arenâ€™t many studies specifically addressing trait anxiety and its relationship to attachment status. In one of the fe w methodologically sound studies in the literature Evans and Wertheim (1996) found a significant relationship between preoccupied attachment and trait anxiety. They assert that individuals with this attachment orientation have a high level of reporting ne gative affect, which might explain
71 their commonly experienced dissatisfaction in close relationships. Similarly, Dozier, Stevenson, Lee, and Velligan (1992) found that greater preoccupation was related to increased report of psychiatric problems. It could be hypothesized that compared to those with dismissing attachment, individuals with preoccupied style might be more likely to benefit from psychotherapy since they tend to have greater oriented toward relationships. However, in their work with a clinical sample Fagony et al. (1996) found that dismissing persons were more likely to show improvement in therapy. There havenâ€™t been other studies focusing on associations between attachment orientation and responses to psychotherapy. There is a great need for additional studies. Such need is not only vital from an empirical point of view, more importantly such work could inform clinical practice with individuals have higher vulnerability to anxiety. If there were such a link to be confirmed by scientific studi es, then it might be useful to consider utilizing interventions that involve a strong focus on early experiences and interpersonal aspect of such personsâ€™ lives. This study intended to trait anxiety as it relates to attachment orientation, gender, persona l meaning, and depressiveness. Attachment and Gender A great number of researchers (i.e. Belenky et al., 1986; Gilligan, 1982) have recommended that compared to males, females tend to have a more relational orientation. The works of such authors have stim ulated a great deal of studies research in various areas. Kenny et al. (1998) report that cross sectional studies have not found any significant gender differences in adolescentsâ€™ parental attachment. These researchers found that adolescent boys perceive d less stability in their parental attachment than did girls. Furthermore, changes in the level of perceived parental attachment was more
72 influential on boysâ€™ well being. Therefore, Kenny et al. (1998) concluded that boys were more vulnerable to psycholo gical distress while experiencing trouble in close relationships because their relationships are less stable or secure. Gender differences might exist in measures of well being. Kotler and Omodie (1988) report that women prefer to talk with intimates whil e men prefer companionship and shared activities. Thus they infer that such difference might differentially function as a buffer against stress. Kobak et al. (1991) assert that along with insecure attachment, gender could be one of the variables that cou ld add to risk factors for adolescent depressiveness. Their findings did indeed confirm that females were in greater risk for depressive symptomatology than males. Cole Detke and Kobak (1996) studied attachment, eating disorders and depression in women. They found that when eating disorders and depression were reported together women tended to have hyperactivating strategies (preoccupied attachment style). On the other hand, when they presented with eating disorders deactivating strategies (dismissing attachment style) were more prevalent. When women reported eating disorders and depression, they had the most extreme levels of symptomatology. As noted by the authors these findings might be due to the attachment classification system they used. They i ndicate that if they had utilized a four category classification perhaps these women would have unresolved attachment classification as opposed to the preoccupied one. As mentioned earlier some authors link codependence with depression (i.e. Blatt, 1974). In their work with college students Springer et al. (1998) found no significant differences between males and femalesâ€™ scores on dependence. Volling, Notaro, and Larsen (1998) found that in general women were more likely to report depressive affect
73 than their husbands. Avoidant husbands married to secure wives were more depressed than secure husbands regardless of whether the secure husbands were married to secure or avoidant wives. Kobak et al. (1991) found that female adolescents were at greater risk to experience depression than their male peers. The researchers propose that this might be due to femalesâ€™ tendency to adopt preoccupied attachment orientation. Perhaps males and females respond differently to normative frustrations and distress at give n developmental stages. In other words, females might be more likely to respond to distress with strategies of internalization while males are likely to use externalization. Consequently, high levels of distress might predispose females for depression a nd males for conduct related problems (Kobak, Sudler, & Gamble, 1991). Kirkpatrick and Davis (1994) found significantly high occurrence of pairing between anxious (preoccupied) and avoidant (dismissing) individuals. Furthermore, this frequency was more pr onounced when the female partner was avoidant and male partner anxious. On the other hand, non avoidant males indicated a preference for non anxious females and non anxious females reported a preference for non avoidant males. The authors observed pairin g of secure males and avoidant females to rank second in frequency. These findings could be viewed as supportive of stereotypical male tendency for independence and female tendency for relatedness. Partnersâ€™ relationship ratings showed that personsâ€™ rati ngs were more influenced by their own attachment status than their partnersâ€™. Kirkpatrick and Davis (1994) also looked at relationship stability based upon attachment style. Females did not show any significant differences in stability based on
74 their atta chment status. On the contrary, anxious males showed the lowest degree of relationship stability. Furthermore, the researchers reason that relationships involving avoidant women, who are not sufficiently skilled in relationships, would show the highest b reakup rates. Relationships involving anxious women, for whom relationships are very important, would show greatest stability. The above studies seem to point to differences in attachment and affect regulation strategies. Therefore, examining how gender relates to attachment status, affect regulation (trait anxiety and vulnerability to depression), and personal meaning among college students would further understanding of college student mental health. Summary There has been a debate among scholars as t o whether attachment could be a utilized as a metaconstruct (Lopez, 1995) that could guide conceptualizing normal human development as well as development of psychopathology. Those who support the idea seem to have a very strong emphasis on the importance of close interpersonal relationships. On the other hand, the authors who oppose to the idea tend to view attachment as an important construct that is informative only in early close relationships and later in romantic relationships. Even without making a choice between these two points of view, it could safely be concluded that attachment theory does offer vital insight into close relationships. Considering the importance of close relationships through the life span, attachment theory, thus can at least offer essential insight into this crucial domain of human functioning. The importance of close relationships is paramount at times when there is high degree of activity for self definition. In addition, presence of supportive relationships is
75 extremely b eneficial at times where individuals go through challenging transitional experiences. The experience of college students involves both of these characteristics. Therefore, research with attachment theory which not only offers insight into their interper sonal life but also into adaptation might help tailor mental health services to the needs and nature of this group. Along with academic and career concerns, college students also bring a great deal of interpersonal and affective issues to counseling. Con sequently, greater insight in affect regulation and interpersonal domains of college students could enrich counseling services. Accordingly, the current study is aimed at examining college studentsâ€™ traits such as depressiveness and anxiousness in light o f attachment. It also incorporates personal meaning and gender. By using personal meaning as a measure of well being, the study is intended to strike a balance between negative and positive correlates of attachment. Including gender is useful because it may reveal meaningful differences or similarities in terms of relationality (attachment), affect regulation (depression and anxiety) and purposefulness (personal meaning).
76 CHAPTER 3 METHODOLOGY Statement of Purpose The purpose of this study was to investigate if college studentsâ€™ attachment orientation and gender had any significant relationships with measures of well being (personal meaning) and of vulnerability to psychop athology (depressiveness and trait anxiety). This chapter introduces the research hypotheses, delineation of variables, population, and sampling procedures. It also presents the data collection, the data analysis procedures, the measurement instruments a nd the limitations of the study. Hypotheses This study investigated the following are null hypotheses: Ho1: There is no significant relationship between college studentsâ€™ attachment status and their level of personal meaning Ho2: Gender, personal meani ng, depressiveness and trait anxiety do not significantly predict college studentsâ€™ placement in attachment categories (secure, fearful, preoccupied, and dismissing; secure versus insecure). Ho3: There is no significant relationship between college stude ntsâ€™ attachment status and their level of depressiveness. Ho4: There is no significant relationship between college studentsâ€™ attachment status and their level of trait anxiety. Ho5: There is no significant relationship between college studentsâ€™ gender and their attachment status, their levels of personal meaning, depressiveness, and trait anxiety. Ho6: There is no significant relationship between college studentsâ€™ level of personal meaning and their level of depressiveness.
77 Ho7: There is no significant relationship between college studentsâ€™ level of personal meaning and their level of trait anxiety. Ho8: There is no significant relationship between college studentsâ€™ level of depressiveness and their level of trait anxie ty. Ho9: There is no significant relationship among college studentsâ€™ attachment status, level of personal meaning, level of depressiveness, and level of trait anxiety. Delineation of Relevant Variables Independent Variables College studentsâ€™ attachmen t status and gender were the two independent variables of this study. Attachment status was defined as college studentsâ€™ relational orientation based upon perceived degrees of anxiety and avoidance in the context of romantic relationships. Attachment sta tus was be measured using the Experiences in Close Relationships Revised (ECR R; Fraley, Waller, & Brennan, 2000). Thus for the purpose of this study, attachment is conceptualized inline with the research tradition of Hazan and Shaver (1987) who found adu lt attachment in romantic relationships as the adult equivalent of attachment in childhood. Moreover, attachment is viewed according to the models by Brennan et al. (1998) and Fraley et al. (2000). Unlike the categorical model of Hazan and Shaver (1987), these researchers view attachment in terms of dimensions of anxiety and avoidance. As recommended by Brennan et al. (1998) comparisons with previous studies using categorical models will be done in the following manner: (a) Low scores on both avoidance and anxiety dimensions correspond to the secure category. (b) High scores on the avoidance and low scores on anxiety dimensions will correspond to avoidant (dismissing) category. (c) High scores on the anxiety and low scores on the avoidance dimension wil l correspond to the anxious ambivalent (preoccupied) category of
78 attachment. (d) High scores on both anxiety and avoidance dimensions will correspond to the fearful classification of attachment. This four category model was originally developed by Bartho lomew and Horowitz (1991). In the analyses where attachment was conceptualized as a categorical variable, personal meaning, factors depressiveness, and trait anxiety were used as the independent variables. The categorization of attachment was intended t o make comparisons with prior research possible since most of such studies used attachment as a categorical variable. Outcome Variables College studentsâ€™ levels of personal meaning, depressiveness, and trait anxiety were the three outcome variables of this study. When conceptualized as a categorical variable, attachment classification was also used as an outcome variable. Personal M eaning Personal meaning refers to the extent to which college students attribute meaningfulness and purposefulness to th eir lives. Personal meaning will be assessed by the Life Regard Index Revised (LRI R; Debats, 1998). Higher scores on the LRI R indicate a greater level of perceived meaningfulness. Depressiveness Depressiveness refers to an individualâ€™s proneness to exp eriencing depressive symptomatology. Depressiveness was conceptualized in accordance with the research tradition of Blatt (1974), and Blatt, Dâ€™Afflit, and Quinlan (1976) who viewed the construct as consisting of two factors; dependency and self criticism. Depressiveness will be measured using Depressive Experiences Question naire (DEQ; Blatt et al., 1976) .
79 Higher scores on DEQ indicate a greater likelihood of college studentsâ€™ sensitivity to experiencing depressive symptoms. Trait Anxiety Trait anxiety is defined as college studentsâ€™ relatively stable proneness to perceiving life situations as threatening and reacting to these situations with elevated states of anxiety. Trait anxiety will be measured by the trait scale of State Trait Anxiety Inventory (S TAI Form Y; Spielberger, 1983). According to the theoretical framework of Spielberger, Gorsuch and Lushene (1970) higher scores on the STAI indicate greater likelihood of proneness to anxiety in a variety of situations. Population The target population of this study was traditional age (18 23) undergraduate students enrolled at the University of Florida. Sampling Procedure A convenience sample was used for this study. It consisted of undergraduate students in four personal growth classes taught at the College of Education. Two interpersonal communication classes, one stress and anxiety management and one alcohol and drug abuse class, were sampled. Students in these classes were informed about the survey by their respective instructors. All the studen ts present in these classes, who were between ages of 18 23 and who volunteered to participate were included in the study. While none of the students present in these classes declined participation in the study, 12 could not do so because they were older than 23. In addition, 2 students had taken the survey in another class, and were also excluded from the survey. The participation was voluntary and students received extra credit from the instructors of the above courses.
80 A total of 166 individuals com pleted the survey. Eleven of these students (6 males 5 females) either did not fully complete the survey packet or completed inappropriately. Thus the responses of these persons were not included in the analysis. As a result of this deletion a total of 155 personsâ€™ responses were used in the analyses ( N = 155). Fewer than 1/3 rd of participants were males (48 persons, 31 %), whereas over 2/3 rd were females (107 persons, 69%). Data Collection Procedures The time of the administration of the survey was arr anged with the instructors of the above courses. These instructors informed the students at least a week ahead of the time. Upon presenting to these classes, the respective instructors introduced the researcher to the students. Students were fully infor med about their rights as research participants. They were also assured about the confidentiality of their responses. Upon obtaining their consent (see Appendix A), students were given the survey packets. Twelve students indicated that they were older t han 23. These students were not included in the study and they left the classroom. The awarding of these students with extra credit point was left to the discretion of their instructors. When all the students completed the surveys, they were fully infor med about the study and welcomed to ask any questions they had. Furthermore, they were given contact information for any questions or concerns they might later have. Instrumentation This study used the Experiences in Close Relationships Revised (ECR R; Fr aley et al., 2000) to measure attachment status; the Life Regard Index Revised (LRI R; Debats, 1998) to assess personal meaning; the Depressive Experiences Questionnaire
81 (DEQ; Blatt et al., 1976) to measure depressiveness; and the trait scale of the State Trait Anxiety Inventory (STAI Form Y; Spielberger, 1983) to assess trait anxiety. The survey also inquired the gender and ages of the participant. The Experiences in Close Relationships Questionnaire Revised (ECR R Fraley, Waller, & Brennan, 2000) The Experiences in Close Relationships Questionnaire (ECR; Brennan, Clark, & Shaver, 1998) is a 36 item Likert type self report measure of adult attachment. More specifically, it measures adult attachment within the context of romantic relationships. The qu estionnaire has two subscales each represented by 18 items. These subscales are Anxiety and Avoidance. The Anxiety scale measures oneâ€™s self reported degree of anxiety in romantic adult relationships, whereas Avoidance assesses the extent of avoidance of intimacy in such relationships. The ECR R differs from the majority measures in that it does not specify types. It rather places individualsâ€™ attachment orientations on the continuum of these two dimensions. The security of attachment is conceptually p laced at lower levels of these two dimensions. It is worth noting that the scores on these two factors can be converted to place respondents into three or four categories. The ECR emerged from a close look at measures of attachment, including the first pr ocedure utilized by Ainsworth et al. (1978), which is known as the Strange Situation. Brennan, Clark and Shaver (1998) recognized that both psychometrically and conceptually these measures were in fact assessing intimacy related anxiety and avoidance. Su ch conceptualization is practical. For instance, it eliminates constraints of prototypical models.
82 In other words, viewing attachment orientation on a two dimensional space is more inclusive of various attachment organizations, including disorganized a ttachment (Main & Solomon, 1986). Since the work of Brennan et al. (1998) involved the comparison of various self report measures of adult attachment, their findings could be interpreted as criteria for concurrent validity of the ECR. One way in which res earchers typically test concurrent validity of an attachment instrument is to compare percentages of attachment types with those of other measures (Brennan et al., 1998; Fraley et al., 2000). Such comparisons are challenging since the measures use prototy pes that are not only diverse in numbers (i.e. two, three, four etc.) but also are conceptually different. Consequently, the conclusions drawn from these comparisons are often tentative. Brennan et al. (1998) compared scores on anxiety and avoidance dime nsions of the ECR to rates of Bartholomew and Horowitzâ€™s (1991) classifications. They combined Bartholomew and Horowitzâ€™s four category classification into secure versus insecure. Only over half (52.8%) of individuals classified as secure on Bartholomew and Horowitz measure were placed as secure according to the ECR. Thus 47.2% of the secure group was classified as insecure on the ECR. About eleven per cent of persons classified as insecure were placed as secure by the ECR. 88.8% of those classified as insecure by Bartholomew and Horowitz were placed as insecure with the ECR. These findings point to some considerable convergence between the two measures. However, they also show the difference in assessing the secure classification. More specifically, the ECR appears to be significantly more conservative in placing individuals in the secure category than the Bartholomew and Horowitzâ€™s (1991) measure. Brennan et al. attribute this difference in the ECRâ€™s ability
83 to more precisely discriminate different degrees of insecure attachment. Further comparisons of these two measures revealed that the two ECR dimensions are conceptually highly similar with Bartholomewâ€™s (Bartholomew & Horowitz, 1991; Griffin, & Bartholomew, 1994) two axes. Brennan et al. found that the Anxiety dimension corresponded to Bartholomewâ€™s Self model dimension, whereas Avoidance was similar with her Other model dimension. In short, these findings could be interpreted as evidence for satisfactory concurrent validity of the ECR. Brenna n et al. (1998) found a correlation coefficient of .12 between the two factors of the ECR, which illustrates that the dimensions of anxiety and avoidance are independent constructs, and are thus orthogonal. Of all the measures used in the study the resear chers identified 60 subscales for statistical comparisons. Anxiety and Avoidance accounted for 62.8% of variance in the 60 subscales. The highest loading subscales on the factor 1 (Avoidance) were; Avoidance of Intimacy (Rothbard et al., 1993); Discomfor t with Closeness (Feeney, Noller, & Hanrahan, 1994); and Self Reliance (West & Sheldon Keller, 1994) with correlations of .91, .90, and .88 respectively. On the other hand, the subscales with highest loadings on the factor 2 (Anxiety) were Preoccupation ( Feeney et al., 1994); Jealousy/Fear of Abandonment (Brennan & Shaver, 1995); and Fear of Rejection (Rothbard et al., 1993) with corresponding correlations of .86, .85 and .83. Fraley et al. used the Item Response Theory (IRT: van der Linden & Hambleton, 19 97) analysis of self report measures of adult attachment in revising the ECR. The item response theory models are â€œdesigned to represent relations between an individualâ€™s item response and an underlying latent trait (Fraley et al., 2000, p. 351).
84 The IRT analysis illustrated that the ECR had better measurement properties than did other instruments. The authors obtained 50 to 100% improvement in measurement precision of the ECR. Fraley et al. (2000) indicate that similar with other measures of attachment (i.e. the Adult Attachment Scale AAS, Collins & Read, 1990; the Relationship Style Questionnaire RSQ, Griffin & Bartholomew, 1994; the Experiences in Close Relationships ECR, Brennan et al., 1998) the ECR R has limitations in precisely measuring high level s of attachment security. The authors attribute this limitation to the item pool from which it was constructed. Fraley et al. utilized complex methods of the IRT to examine psychometric properties of four of the most commonly used self report adult attach ment instruments (the Adult Attachment Scale AAS, Collins & Read, 1990; the Relationship Style Questionnaire RSQ, Griffin & Bartholomew, 1994; the attachment questionnaire unnamed, Simpson, 1990; the Experiences in Close Relationships ECR, Brennan et al., 1998). Their findings revealed that The ERC (Brennan et al., 1998) had superior test information functions. For example, it had greater degree of measurement precision. Thus they recommend that it may be a preferable alternative to the other three measu res. However, similar with the others, the ECR scales fail to measure the secure end of each dimension (Anxiety and Avoidance) as precisely as it does the insecure end. Hence, they attempted to improve the ERC scales by looking at item discrimination valu es without increasing the number of items. In doing so they used the 323 item pool by Brennan et al. and performed principal axis factor analysis (for details of these IRT procedures see Fraley et al, 2000). The authors improved the ECR scalesâ€™ ability t o measure a wide trait range with high degree of precision. However, this
85 improvement was limited by constraints of the original item pool. Fraley et al. indicate that the pool didnâ€™t have sufficient numbers of items capturing the low end of the two dime nsions (Anxiety and Avoidance). Thus they obtained median Beta1 values of â€“ 1.67 and â€“ 1.86 for Anxiety and Avoidance respectively. The items with low Beta1 values also tended to have low discrimination values. The correlation between alpha and Beta1 was . 59 for Anxiety and .68 for Avoidance. Therefore, Fraley and colleagues selected items with highest discrimination values and came up with 18 items for each of the two factors. Thirteen of Anxiety (72%) and 7 of Avoidance (39%) scale items were from the o riginal ECR. Due to this overlap of the items they refer to the new instrument as Experiences in Close Relationships Revised (ECR R). The comparisons of the test information functions of the ECR and the ECR R substantially favored the latter. Moreover, F raley et al. report that they improved the ECRâ€™s measurement precision by 50 to 100% without increasing the number of items. Despite this improvement the ECR R still does have limitation in assessing the lower end of the two dimensions, thus security of a ttachment. Finally, Fraley and colleagues examined reliability coefficients of the ECR R in comparison with the other four instruments (including the original ECR). The ECR R had higher test re test reliability coefficients (ranging from .93 to .95) then the other measures. The original ECR was constructed by using the highest factor loading items of various existing self report measures of adult attachment. As indicated by Brennan et al. (1998) and Fraley et al. (2000) it has demonstrated superior psych ometric properties than
86 the other measures. The ECR R on the other hand could be thought of as an even more refined measure than the original ECR. Thus it appears to be the most preferable instrument of its kind. The scores on the factors of attachment ( anxiety and avoidance) in this study were calculated according to Fraley (2003). Using a classical test theory approach, attachment scores were calculated by summing studentsâ€™ scores on these two factors. Similarly, the scores on the dimensions were conv erted into four categories of attachment according to Brennan et al. (1998). The four categories were secure, fearful, preoccupied, dismissing. These categories were originally developed by Bartholomew and Horowitz (1991). These categories were also con verted into secure versus insecure for further analysis of the data. In this categorization fearful, preoccupied and dismissing classifications were combined into the insecure category. The Life Regard Index Revised (LRI R; Debats, 1998) Battista and A lmond (1973) developed the original Life Regard Index (LRI), which is a 28 item Likert scale. The authors aimed at creating the LRI as a value independent instrument that operationalizes the construct of positive life regard (Debats, 1998). The authors u sed the phrase value independent to refer to a lack of affiliation with any specific belief system. Battista and Almond preferred to the term life regard to refer to oneâ€™s perception of life as essentially meaningful. They believed that the presence of a framework was an essential prerequisite to a sense of meaning. Furthermore, they proposed that the extent to which the goals of this framework are actualized would give a person a global sense of fulfillment with life. Hence, they constructed the questi onnaire into two subscales: the Framework Scale (FS) and the Fulfillment Scale (FU). These two
87 scales are divided into positive and negative clusters, thus composing a total of four clusters. The clusters are: positive framework, negative framework, posi tive fulfillment, and negative fulfillment. Each cluster is represented by seven subsequent items. The authors indicate that FS â€œmeasures the ability of an individual to see his [or her] life within some perspective or context, and to have derived some g oals, purpose in life or life view from themâ€ (p. 411). FU measures the extent to which one has fulfilled or feels being in the process of actualizing this framework and goals. The questionnaire originated from two fundamental questions: â€œWhat is the na ture of individualsâ€™ experience of their lives as meaningful? And what are the conditions under which individuals will experience their lives as meaningful?â€ (Debats, 1998, p. 239). The questionnaire is intended to assess oneâ€™s overall positive life regar d. As mentioned above, Battista and Almond (1973) intended to construct the LRI as a value independent instrument. Thus the items focus on oneâ€™s perceived degree of dedication to any sets of values, as opposed to some particular values. Findings of Nick els and Stewart (2000) supported this claim. The authors compared scores of the Rokeach Value Survey (RVS; Rokeach, 1973) with the LRI. The RVS is a 36 item questionnaire, designed to measure specific belief systems or value orientations. Looking at the associations between the scores on the LRI and individual items on the RVS, Nickels and Stewart found low correlation coefficients ranging from of .16 and .30 for only four out of 36 items. The rest of the items (32/36) were not correlated with the LRI. Debats (1998) revised the original measure by Battista and Almond (1973). He randomized distribution of items belonging to each of the four clusters. In the original form (Battista, & Almond, 1973) the 28 questions were in four subsequent clusters of
88 sev en items. Debats reasoned that such homogenous blocks would lead to biased responses. In addition, he eliminated descriptive responses such as much , completely , and very from three original items. Moreover, instead of using the 5 point rating scale Deba ts used a 3 point scale. He allocated 1, 2, 3 to correspond to do not agree , no opinion , and agree in scoring of the questionnaire. Thus 1, 2, 3 were used in scoring positive items, while 3, 2, 1 were used for negative items. Battista and Almond (1973) r eported high reliability and validity estimates for the original LRI. They investigated psychometric properties of the Index working with a sample of college students ( N = 229). They found a test retest reliability estimate of .94. The subscales of Fram ework and Fulfillment highly correlated with the total scores of the Index; .94 and .93 respectively. The authors found that the LRI significantly differentiated positive and negative life regard groups without the confounding impact of social desirabilit y, which only accounted for 4 % the variance. Battista and Almond also tested the concurrent validity of the LRI by comparing scores on the LRI and with scores on Rosenberg Self Esteem Scale (1965). The correlation coefficient between scores of the two m easures was .62 and significant. Debats (1990) examined psychometric properties of the LRI with a sample of Dutch college students. He found Cronbach alpha estimates of internal consistency ranging from .86 (Index), and .80 (Fulfillment) to .79 (Framework ). With respect to intercorrelations of the scales and the Index, Debats calculated Pearson correlations. The Fulfillment scale correlated with the Index .88, and the Framework and the Index .87, whereas the two subscales correlated .54. These statistic s indicate that the subscales are highly related to the Index but are relatively independent.
89 Furthermore, Debats (1990) looked into relationships between subscales of the LRI measures of well being. He found chi square values attesting the construct vali dity of the LRI. For example, chi squares for happiness and life satisfaction scales were 30.23 and 17.65 respectively ( df = 2; p < .001). The total score on the Index was also found to relate in expected directions to measures of pathological symptomato logy with correlations between total LRI scores and anxiety and depression reaching .32 and .37, respectively. Harris and Standard (2001) tested psychometric properties of the LRI Revised with a sample of adults ( N = 91) with varying demographic characte ristics. The subscales of Framework (FS) and Fulfillment (FU) were correlated at .80. The test retest reliability coefficients at 8 9 weeks were satisfactory for the full index, FS and FU (.87, .82, and .81 respectively). Similarly, Debats reported test retest reliability ranging from .73 to .80. Due to the small size of their sample, Harris and Standard did not detect sufficient evidence supporting the theorized two factor structure. With regard to concurrent and discriminative validity of the LRI Revis ed, Harris and Standard found moderate to strong correlations with measures of similar constructs (i.e. Life Attitude Profile Revised; Reker, 1992; Personal Meaning Profile; Wong, 1998). There is a need for further testing of psychometric properties of the LRI Revised with large and diverse samples. However, the existing preliminary findings do report satisfactory psychometric properties for the LRI Revised as a measure positive life regard.
90 The scores on LRI Revised were calculated following directions by Debats (1998). Such calculation involves assigning points from 1 to 3 to the regular items and 3, 2, and 1 to the reversed items and then summing all. Depressive Experiences Questionnaire (DEQ; Blatt et al., 1976) The DEQ (Blatt, Dâ€™Affliti, & Quinlan, 1976) consists of 66 Likert type items assessing dependency and self criticism, which are considered to be fundamental personality traits associated with vulnerability to depression. Although Blatt and colleagues identified efficacy as a third factor, due in part to theoretical and clinical reasons often the first two factors have been used and investigated in studies involving the DEQ. In effect, the first two factors are thought to be associated with two types of depressive experiences, namely, anacliti c and introjective depression. The former type of depression is marked with â€œfeelings of helplessness and weakness, by fears of being abandoned, and by wishes to be cared for, loved and protectedâ€ (Blatt et al., 1976, p. 383). The latter â€œis characteriz ed by intense feelings of inferiority, guilt, and worthlessness and by a sense that one has failed to live up to expectations and standardsâ€ (p. 383 384). It is important to note, however, that the two types are not mutually exclusive. One can experience both types of depression simultaneously. Drawing from works of several major figures (i.e. Bibring, 1953; Seligman, 1975; Beck, 1967; & Jacobson, 1971) in the area of depression research, Blatt et al. hypothesized that the DEQ would capture experiences as sociated with both types of depression. Hence, they expected the DEQ to assess these two underlying clusters of characteristics that contribute to predisposition to depression.
91 Although the items on the DEQ reflect experiences often reported by clinica lly depressed persons, the DEQ doesnâ€™t assess symptoms of clinical depression. In constructing the DEQ, Blatt et al. worked with a sample of college students (N=660). The followings were the items with the highest factor loading for dependency and self cr iticism respectively: â€œI often think about the danger of losing someone close to meâ€ and â€œAfter an argument I feel very lonely;â€ â€œThere is a considerable difference between how I am now and how I would like to beâ€ and â€œI often feel guiltyâ€ (Zuroff & Fitzp atrick, 1995, p. 256). Individuals scoring high on dependency scale show preoccupation with feelings of abandonment and loneliness, whereas those with higher scores on self criticism do so with professional or academic striving and feelings of worthlessne ss (Whiffen et al., 2000). Blatt et al. (1976) tested the items of the DEQ with various other measures. They report overall satisfactory psychometric properties for the subscales of Self Criticism, and Dependency. However, the authors rarely reported sta tistical values to confirm their propositions. They compared the DEQ factors to a modified form of the Self Rating Depression Scale (Zung, 1973) for nonclinical subjects. Their sampleâ€™s mean ( N = 128) was slightly lower (46.20) than the cut off score (50 ) established by Zung. In addition, Self Criticism but not Dependency significantly correlated with the total score on the Self Rating Depression Scale ( r = .54). Zuroff and Fitzpatrick (1995) compared a college sampleâ€™s ( N = 160) scores on the DEQ and th e Sociotropy Autonomy Scale (SAS; Beck, Epstein, Harrison, & Emery, 1983). The SAS constructs are theoretically similar with those of Self Criticism and Dependency of the DEQ. More specifically, Sociotropy is relevant to Dependency
92 whereas Autonomy is as sociated with Self Criticism. In fact, Zuroff and Fitzpatrick (1995) confirmed these two associations with correlation coefficients of .75 and .61 respectively. These correlations are statistically significant. Zuroff, Moskowitz, Wielgus, Powers, and Fra nko (1983) tested the reliability and construct validity of the DEQ with a college sample. They administered the DEQ three times; pretest, second testing with an interval of 5 weeks and the third testing 13 weeks from the pretest. Zuroff et al. found tha t DEQ measures were independent at pre testing all r â€™s < .15, and all p â€™s > .25. The measures were independent at post testing with the exception of a small correlation at the third testing r = .30 ( N = 39; p = .05). These researchers did not detect a sign ificant change by time in mean scores of the Self Criticism scale. On the other hand, they did find a significant but small time effect for the means on the Dependency scale ( F (1, 31) = 11.9, p < .001). The mean scores increased from 1.43 to .28, which was a change smaller than the standard deviation of .20. Zuroff et al. interpreted these findings as reasonable evidence for stability over time. Zuroff, Igreja, and Mongrain, (1994) found similarly convincing evidence for the DEQâ€™s test retest reliabil ity (12 months) with r = .79. The same authors also detected high internal consistencies (Cronbachâ€™s a = .75). Zuroff, Quinlan, and Blatt (1990) found similar alpha values in their study with a large college sample ( N = 1154; 779 females, and 373 males). They report Cronbachâ€™s alphas for females as .81 for Dependency and .75 for Self Criticism. Corresponding male alpha value were 80 for Dependency and .77 for Self Criticism. These findings are in support of reliability of the DEQ.
93 To further examine t he concurrent validity of the DEQ, Zuroff et al. (1983) compared scores of males and females with their scores on measures of guilt, self esteem, and locus of control (Guilt Scale, Mosher, 1966; Self Esteem Scale, Oâ€™Brien, & Epstein, 1974; & Locus of Contr ol Scale, Rotter, 1966). Their comparisons were made based on gender and the two scales of the DEQ. The findings were in agreement with Blattâ€™s (1974) model. In other words, the various subscales of these tests were in general related to the DEQ subscal es in theoretically expected directions and degrees. Fuhr, and Shean (1992) examined test retest (2 nd test given 3 4 weeks later) reliability of the DEQ. They found a high coefficient of .86. The authors also tested the scaleâ€™s concurrent validity by comp aring it to the Beck Depression Inventory (BDI; Beck, Ward, Mendelson, Mock, & Erbaugh, 1961) and selected scales (Deference, Autonomy, Succorance, and Abasement) of the Edwards Personal Preference Schedule (EEPS; Edwards, 1959). Their results indicated t hat the Dependency scale was positively correlated with the BDI ( r = .63, p < .001) and the EPPS Abasement scale ( r = .52, p < .001) and negatively correlated with the EPPS Deference scale ( r = .29, p < .01). The Self Criticism scale was positively corre lated with the BDI ( r = .40, p < .001) and negatively correlated with EPPS Deference scale (r= .29, p< .01) and the EPPS Succorance scale (r= .31, p< .01). Considering the theoretical differences between the DEQ scales and BDI and those of the EPPS, the se findings could be considered as supportive of the validity of the DEQ. In conclusion, there has been a considerable body of research supporting the validity and reliability of the Dependency and Self Criticism scales of the DEQ. Blaney and Kutcher (199 1) indicate that the DEQ has been the most extensively researched
94 measure among instruments of its kind (i.e. the Sociotropy Autonomy Scale by Beck, et al., 1983; the Anaclitic and Introjective Dysfunctional Attitude Scale by Mongrain & Zuroff, 1989). Giv en this accumulated work, the DEQ was used in this study to measure proneness to depression. Scores on DEQ were obtained using calculation procedures of Santor, Zuroff, and Fielding (1997). These procedures involve a series of computer programs, which pro vide separate scores for each factor of depressiveness and a score for depressiveness. In other words, the depressiveness score is not calculated by merely summing the scores on the two subscales. The State Trait Anxiety Inventory, Trait Version, Form Y ( STAI T: Spielberger, 1983) The State Trait Anxiety Inventory (STAI) was developed by Spielberger et al. (1970). It is a brief self report instrument assessing both state and trait anxiety for clinical and empirical purposes (Spielberger & Sydeman, 1994). The STAI conceptualizes and measures state anxiety as the intensity of feelings of anxiety at a given time. On the other hand, it assesses trait anxiety as oneâ€™s general tendency to feel an array of situations as threatening (Ramanaiah, Franzen, & Schi ll, 1983). Furthermore, Spielberger et al. (1970) referred to state anxiety (S Anxiety) â€œas a temporal cross section in the emotional stream of life of a person, consisting of subjective feelings of tension, apprehension, and worry, and activation (arousa l) of the autonomic nervous system (Spielberger, & Sydeman, 1994, p. 295). Thus the intensity of S Anxiety varies depending upon time and circumstances. Trait anxiety (T Anxiety) was conceptualized as proneness to experiencing a wide range of stressful s ituations as threatening.
95 Thus Spielberger et al. (1970) hypothesized that individuals with high T Anxiety would be more likely to respond to stressful situations with more frequent and higher levels of S Anxiety. The inventory is a widely used and reli able self report measure of anxiety. In fact, it is translated to over 40 languages and is considered to be the most commonly used measure of anxiety across cultures (Hishinuma et al., 2001). Moreover, Kennedy et al. (2001) report that by 1989 there had been over 3000 published studies that had used the STAI. Since the current study utilized the T Anxiety scale (STAI T) but not S Anxiety scale, only psychometric properties of this scale are introduced here. The STAI T is consisted of 20 statements. The r espondents are asked to circle the number that best describes how they generally feel: 1 (almost never), 2 (sometimes), 3 (Often) and 4 (almost always). The following two statements are examples of items reflecting the presence and absence of t anxiety. A nxiety present: â€œI worry too much over something that really doesnâ€™t matter; Anxiety absent : â€œI am content; I am really a steady personâ€ The original STAI by Spielberger et al. (1970) is known as the Form X. Based upon the great deal of research accumulat ed for over a decade Spielberger (1983) revised the Form X in an effort to develop purer measures of state and trait anxiety as well as to better discriminate between experiences of anxiety and depression (Hishinuma et al., 2001). This revised version of STAI is known as the Form Y.
96 Spielberger (1983) reports test retest stability coefficients for the Y Form of T Anxiety to range from .73 to.86 for college students and .65 to .75 for high school students. Alpha coefficients for internal consistency of t he scale is reported to have a median of .90 for various populations such as working adults, college and high school students, and military recruits. Spielberger also reports item remainder correlations as .50 or higher for all T Anxiety items. Spielberger (1983) eliminated items with depressive content such as â€œat times I think I am no good at allâ€ or those highly related to anger than anxiety from the item pool to obtain greater validity. Furthermore, he tested concurrent validity by obtaining T Anxiety scaleâ€™s correlations with commonly utilized trait anxiety measures such as Manifest Anxiety Scale (MAS; Taylor, 1953) and The Anxiety Scale Questionnaire (ASQ; Cattell, & Scheier, 1963). Correlations with these two scales ranged from .73 to .85. Spielberg er and Sydeman (1994) noted that the T STAI has two major advantages to these two scales. First the T STAI measures anxiety without including as much content of depression or of anger as these scales do. Second, it is half in length as the other two, thu s it doesnâ€™t take nearly as much time to administer. Regarding the construct validity of the T STAI, Spielberger (1983) reports that with the exception of personality disorders, the scale differentiated between normal and clinical subjects. In other words , clinical samples obtained overall higher scores on the T STAI than nonclinical ones. Spielberger considers the lower T STAI scores of individuals with personality disorders as further evidence of construct validity of the scale, reasoning that such pers onsâ€™ unique ways of experiencing anxiety is a defining characteristic of their condition.
97 Along with extensive psychometric work done by Spielberger (1983) with larger samples of various populations, the STAI has been utilized in research and clinical pr actice worldwide. This popularity of the inventory is also evidence for itâ€™s widely agreed upon psychometric properties. Thus the current study used this scale to assess trait anxiety. The scores on STAI were calculated using the procedures specified in the test manual by Spielberger. Data Analytic Procedures The data of this study were examined using simultaneous multiple regression analysis, multinomial and binary logistic regression analyses, correlation matrix and t test procedures. When examining a ttachment as a continuous independent variable, simultaneous multiple regression analysis was conducted for each outcome variable; personal meaning, depressiveness, and trait anxiety. Also, in order to predict individualsâ€™ membership in each of four attac hment classifications, multinomial logistic regression analysis was conducted. In addition, binary logistic regression analysis was used to predict college studentsâ€™ placement in secure versus insecure attachment categories. Moreover, sets of t tests wer e used to examine gender differences on all the measures of the study; attachment related anxiety, attachment related avoidance, personal meaning, self criticism, dependency and trait anxiety. Finally, to further investigate interrelations between each pair of the variables, correlation matrix was used for all the variables except for gender due to its categorical nature.
98 CHAPTER 4 RESULTS Summary and Chapter Overview The purpose of this study was to investigate relationships between attachment status, gender, personal meaning, depressiveness and trait anxiety among traditional age college students. This chapter introduce s the sample, descriptive statistics, results of hypothesis testing, and a summary of the results. The descriptive statistics are presented in comparison with norm groups and samples of previous studies. Hypothesis testing was done using simultaneous mul tiple regression and multinomial logistic regression analyses. In addition, t tests were conducted to examine gender differences. Finally, correlations between all the pairs of the variables were obtained to further examine interrelationships. Descriptiv e Statistics The sample of this study consisted of volunteer undergraduate students recruited from four personal growth classes in College of Education. A total of 166 students completed the survey. Eleven of these persons (6 males 5 females) either did not complete the survey packet or completed it inappropriately. Thus the responses of these persons were not included in the analysis. As a result of this deletion, responses of 155 were used in the analyses ( N =155).
99 Fewer than 1/3 rd of participants were males (48 persons, 31 %), whereas over 2/3 rd were females (107 persons, 69%). Fewer than 1/3 rd of participants were males (48 persons, 31 %), whereas over 2/3 rd were females (107 persons, 69%). The ages of these stud ents varied from 18 to 23. ( Table 4. 1) Table 4. 1 Frequencies for Gender 107 69.0 69.0 69.0 48 31.0 31.0 100.0 155 100.0 100.0 Female Male Total Valid Frequency Percent Valid Percent Cumulative Percent Table 4. 2 Descriptive Statistics for Attachment Related Anxiety, Attachment Related Avoidance, Personal Meaning, Dependency, Self Criticism, and Trait Anxiety This study investigated five primary variables; attachment status, gender, personal meaning, depressiveness and trait anxiety. Of these variables, attachment status and depressiveness consist of two factors. The dimensions (factors) for attac hment status are attachment related anxiety and attachment related avoidance. The factors for depressiveness are dependency, and self criticism. These four dimensions were treated as separate variables. 155 2.8896 1.1763 155 2.7523 1.1863 155 71.90 10.42 155 132.2482 15.2489 155 111.2129 17.0503 155 37.18 8.60 155 Attachment Anxiety Attachment Avoidance Personal Meaning Dependency Self Criticism Trait Anxiety Valid N (listwise) N Mean Std. Deviation
100 Table 4. 2 includes descriptive statistics for 6 variables ; attachment related anxiety, attachment related avoidance, personal meaning, dependency, self criticism, and trait anxiety. For comparison purposes, descriptive statistics on the Experiences in Close Relationships Revised (Fraley et al., 2000) a re reported along with those by Fraley (2003). Fraley (2003) reports descriptive statistics gathered from individuals who took the ECR R on line (N > 22,000; 78 % female, 22 male). The means of the current sample were lower than the Fraley (2003) sample with the exception of male avoidance scores. The standard deviations of both samples seem similar. ( Table 4. 3) Table 4. 3 Descriptive Statistics for Attachment Related Anxiety and Attachment Related Avoidance in Comparison to those of Fraley (2003) Gender Anxiety Avoidance Mean SD Mean SD Fraley (2003) Male 3.64 1.33 2.88 1.15 Female 3.64 1.33 2.95 1.19 Current sample Male 2.75 1.09 3.19 1.14 Female 2.94 1.21 2.5 5 1.15 Distribution of college students across attachment categories seem to be similar with that of a sample by Bartholomew and Horowitz (1991), who used their four category measure in assessing attachment. Percentages of their sample were as the foll owing; 57% secure, 15% fearful, 10% preoccupied, and 18% dismissing. As illustrated in Table 4. 4, with the exception of the secure category, percentages of the current sample of were slightly higher. The percentage for the secure attachment was considera ble lower for the current sample. This finding is not surprising since the ECR R is known to be more conservative in measuring secure attachment (Fraley et al., (2000).
101 Table 4. 4 Distribution o f Participants i nto the Attachment Categories As illustrated in Table 4. 5, means and standard deviations for the Life Regard Index Revised (Debats, 1998) were lower than those of a general population sample reported by Harris and Standard (2001). On the other hand, they were higher th an a college sample by Nickels and Stewart (2000). Table 4. 5 Descriptive Statistics for the Life Regard Inventory Revised in Comparison to Those by Nickels & Stewart (2000); and Harris & Standard (2001) Mean SD Current sample 71.90 10.42 Ni ckels & Stewart (2000) 66.03 12.24 Harris & Standard (2001) 73.06 10.57 Scores on the Depressive Experiences Questionnaire (DEQ; Blatt et al., 1976) were compared to those reported by Santor et al. (1997) from an undergraduate sample at McGill Uni versity, Montreal, Canada. As seen in Table 4. 6, male scores on dependency for both samples seem similar, whereas the male scores on self criticism of the McGill sample were overall higher than the current sample. Female scores on dependency were higher than those of the McGill sample. Female scores on self criticism were lower than those of the McGill sample. 72 46. 5 46.5 46.5 25 16.1 16.1 62.6 23 14.8 14.8 77.4 35 22.6 22.6 100.0 155 100.0 100.0 Secure Fearful Preoccupied Dismissing Total Valid Frequency Percent Valid Percent Cumulative Percent
102 Table 4. 6 Descriptive Statistics for Dependency and Self Criticism in Comparison to a McGill University Sample Gender Mean Standard Deviation McGill Dependency Male 124.22 14.27 Female 130.72 15.98 McGill Self Criticism Male 122.17 18.14 Female 119.53 18.22 Dependency Male 125.37 13.99 Female 135.32 14.83 Self Criticism Male 113.93 11.41 Fema le 109.99 18.96 As seen in Table 4. 7, the mean scores on trait anxiety appear lower than norms for college students reported by Spielberger (1983). The means for males and females of the norm group were 38.30 and 40.40 with standard deviations of 9.18 and 10.15 respectively. Table 4. 7 Descriptive Statistics for Trait Anxiety in Comparison to N orm s by Spielberger (1983) Gender Mean SD Current Sample Male 34.79 7.98 Female 38.25 8.68 Spielberger Male 38.30 9.18 (1983) Female 40.40 10.15 Results Hypothesis One Ho1: There is no significant relationship between college studentsâ€™ attachment status and their level of personal meaning. A simultaneous multiple regression analysis was conducted to test this hypothesis. While the out come variable was personal meaning, the independent variables were attachment related anxiety, attachment related avoidance, gender, dependency, self -
103 criticism, and trait anxiety. As shown in Table 4. 8, this model was significant ( F (6, 148) = 16.696, p < .0001. Adjusted R Square = .379) and it accounted for 37.9 % of the variance in personal meaning. According to this model, only trait anxiety ( t (1, 154) = 3.714, p < .0001), and self criticism ( t (1, 154) = 3.248, p < .001) were significantly related to personal meaning. Both of these relationships were negative. On the other hand, neither attachment related anxiety ( t (1, 154) = 1.914, p = .058) nor attachment related avoidance ( t (1, 154) = 1.070, p = .286) had a significant relationship to per sonal meaning. Thus the null hypothesis was not rejected. In other words, there was no significant relationship between college studentsâ€™ attachment orientation and their degree of personal meaning. Table 4. 8 Simultaneous Multiple Regression Output Usi ng Personal Meaning as the Dependent Variable In short, neither attachment related anxiety nor attachment related avoidance was 104.646 7.281 14.372 .000 1.301 .680 .147 1.914 .058 .692 .647 .079 1.070 .286 3.742E 02 1.609 .002 .023 .981 4.813E 02 .049 .070 .989 .324 .172 .053 .281 3.248 .001 .385 .104 .318 3 .714 .000 (Constant) Attachment Anxiety Attachment Avoidance Gender Dependency Self Criticism Trait Anxiety Model 1 B Std. Error Unstandardized Coefficients Beta Standardize d Coefficients t Sig. Dependent Variable: Personal Meaning a. .635 a .404 .379 8.21 .404 16.696 6 148 .000 Model 1 R R Square Adjusted R Square Std. Error of the Estimate R Square Change F Change df1 df2 Sig. F Change Change Statistics Predictors: (Constant), Trait Anxiety, Gender, Attachment Avoidance, Dependency, Attachment Anxiety , Self Criticism a.
104 significantly related to personal meaning. Hence, the null hypothesis was not rejected. In other words, there was no significant relationship between college studentsâ€™ attachment status and their level of personal meaning. For further examination of the relationship between attachment and personal also see the next hypoth esis. Hypothesis Two Ho2: Gender, personal meaning, dependency, self criticism, and trait anxiety do not significantly predict college studentsâ€™ placement in any of the four attachment categories; secure, fearful, preoccupied, and dismissing. L ogistic regression analys es were used to test this hypothesis. Attachment classification, a categorical variable with four levels, was the outcome variable . This variable was also classified into two categories (secure and insecure). In these analyses, gender, personal meaning, dependency, self criticism, and trait anxiety were the independent variables . A multinomial logistic regression analysis with main effects (gender, personal meaning, dependency, self criticism, and trait anxiety) and all possible inter action effects was conducted. The results of this model showed that the sample size was not sufficient for the inclusion of the interactions. The difference between the deviance for the interaction model (259.131) and the deviance for the main effects (3 47.912) was divided by the difference between the two degrees of freedom (447 372= 75) and it was equal to 88.781, which is smaller than the chi square value of 90.5312 (70< df <80, a= 0.005). Given this likelihood ratio test, with the exception of inter actions with gender, no other interactions were included in the model. The following models were run to test the null hypothesis in several ways.
105 In the first model, a multinomial logistic regression analysis was utilized to examine the main of effects of gender, personal meaning, dependency, self criticism, and trait anxiety. As illustrated in Table 4. 9, this model was overall significant accounting for between 25.5 and 27.7% of the variance in attachment style. According to this model, only gender a nd personal meaning were significant. Individuals with lower scores on personal meaning were more likely to be in the fearful category of attachment while those with higher scores were more likely to be in the secure one. Gender was significant regarding all the paired comparisons of attachment classifications with the secure one, with the exception of the preoccupied category: In secure versus dismissing comparison, females were more likely to be in the secure category while males were more likely to be in the dismissing one. In the fearful versus secure comparison, females were more likely to be in the secure category, whereas males were more likely to be in the fearful one. No significant main effects were found regarding dependency, self criticism, or trait anxiety in predicting placement in the four categories of attachment. For interactions with gender see the third analysis. Table 4. 9 Multinomial Logistic Regression Analyses Output for Attachment Style as the Outcome Variable with Four Levels Main Effects Only 393.571 347.912 45.659 15 .000 Model Intercept Only Final 2 Log Likelihood Chi Square df Sig. .255 .277 Cox and Snell Nagelkerke
106 In the second model a binary logistic regression analysis was conducted to investigate if the independent variables predicted studentsâ€™ likelihood of b eing in the secure versus insecure categories of attachment. In this analysis fearful, preoccupied and dismissing categories were recoded into the insecure classification. As illustrated in Table 4. 10, this model was overall significant accounting for 13 .3 to 17.7% of the variance in attachment classification. However, as individual variables, personal meaning, dependency, self criticism, and trait anxiety did not predict studentsâ€™ placement in secure versus in secure attachment categories. 1.833 3.697 .246 1 .620 1.358E 02 .029 .223 1 .637 .987 .932 1.044 1.711E 02 .016 1.195 1 .274 .983 .953 1.014 1.129E 03 .018 .004 1 .949 .999 .965 1.034 4.507E 02 .035 1.613 1 .204 1.046 .976 1.121 1.290 .496 6.767 1 .009 .275 .104 .728 0 a . . 0 . . . . 1.203 4.337 .077 1 .781 6.333E 0 2 .030 4.478 1 .034 .939 .885 .995 8.093E 03 .019 .186 1 .666 1.008 .972 1.046 2.916E 02 .022 1.722 1 .189 1.030 .986 1.075 2.771E 02 .040 .477 1 .490 1.028 .950 1.112 1.271 .591 4.628 1 .031 .281 8.815E 02 .893 0 a . . 0 . . . . 2.489 4.123 .364 1 .546 4.824E 02 .031 2.491 1 .115 .953 .897 1.012 9.559E 03 .018 .281 1 .596 1.010 .975 1.046 3.409E 02 .020 2.841 1 .092 1.035 .994 1.077 2.378E 02 .043 .301 1 .583 .977 .897 1.063 .760 .734 1.073 1 .300 2.138 .508 9.006 0 a . . 0 . . . . Intercept MEANING DEPENDEN SELFCRIT TRAITANX [GENDER=0] [GENDER=1] Intercept MEANING DEPENDEN SELFCRIT TRAITANX [GENDER=0] [GENDER=1] Intercept MEANING DEPEN DEN SELFCRIT TRAITANX [GENDER=0] [GENDER=1] Attachment Status Dismissing Fearful Preoccupied B Std. Error Wald df Sig. Exp(B) Lower Bound Upper Bound 95% Confidence Interval for Exp(B) This parameter is set to zero because it is redundan t. a.
107 Tab le 4. 10 Multinomial Logistic Regression Analyses Output for Attachment S ecurity a s the Outcome Variable with Two Levels Main Effects Only The third model used multinomia l regression analysis to examine if the independent variables and their interactions with gender predicted placement in the four attachment classifications. As seen in Table 4. 11, this model was overall significant accounting for 31.6 to 24.4% of the vari ance in attachment classification. No significant main or interaction effects were found. Table 4. 11 Multinomial Logistic Regression Analyses Output for Attachment Style as the Outcome Variable with Four Levels Gender Interaction Effects Included 22.061 5 .001 22.061 5 .001 22.061 5 .001 Step Block Model Step 1 Chi square df Sig. 192.033 .133 .177 Step 1 2 Log likelihood Cox & Snell R Square Nagelkerke R Square .794 .412 3.722 1 .054 2.212 .040 .022 3.174 1 .075 1.041 .003 .012 .080 1 .777 1.003 .017 .014 1.510 1 .219 .983 .022 .029 .558 1 .455 .979 1.361 2.902 .220 1 .639 .256 GENDER(1) MEANING DEPENDEN SELFCRIT TRAITANX Constant Step 1 a B S.E. Wald df Sig . Exp(B) Variable(s) entered on step 1: GENDER, MEANING, DEPENDEN, SELFCRIT, TRAITANX. a. 393.571 334.602 58.969 27 .000 Model Intercept Only Final 2 Log Likelihood Chi Square df Sig.
108 .316 .344 Cox and Snell Nagelkerke 3.85 8 6.529 .349 1 .555 6.713 8.066 .693 1 .405 823.212 1.120E 04 6048797212 0 a . . 0 . . . . 9.550E 03 .048 .040 1 .842 .990 .902 1.088 1.110E 02 .026 .183 1 .669 .989 .940 1.041 4.764E 02 .039 1.490 1 .222 1.049 .972 1.132 1.951E 02 .056 .122 1 .727 1.020 .914 1.138 1.142E 02 .062 .034 1 .853 .989 .876 1.115 0 a . . 0 . . . . 9.190E 03 .033 .078 1 .781 .991 .929 1.057 0 a . . 0 . . . . 5.702E 02 .044 1.666 1 .197 .945 .866 1.030 0 a . . 0 . . . . 1.360E 02 .075 .033 1 .856 1.014 .876 1.174 0 a . . 0 . . . . .804 7.742 .011 1 .917 4.112 9.610 .183 1 .669 1.637E 02 1.083E 10 2476495.579 0 a . . 0 . . . . 6.259E 02 .053 1.394 1 .238 .939 .847 1.042 1.403E 02 .032 .192 1 .661 .986 .926 1.050 7.759E 02 .047 2.702 1 .100 1.081 .985 1.185 6.331E 02 .067 .895 1 .344 .939 .823 1.070 5.988E 03 .066 .008 1 .927 1.006 .885 1.144 0 a . . 0 . . . . 3.109E 02 .041 .574 1 .449 1.032 .952 1.118 0 a . . 0 . . . . 6.676E 02 .054 1.504 1 .220 .935 .841 1.041 0 a . . 0 . . . . .159 .089 3.244 1 .072 1.173 .986 1.395 0 a . . 0 . . . . 8.994 13.316 .456 1 .499 7.826 14.066 .310 1 .578 2505.652 2.667E 09 2.354E+15 0 a . . 0 . . . . 2.972E 02 .094 .099 1 .753 .971 .807 1.168 7.986E 02 .066 1.473 1 .225 1.083 .952 1.232 .110 .082 1.795 1 .180 1.117 .950 1.313 .440 .261 2.831 1 .092 .644 .386 1.075 1.827E 02 .100 .033 1 .855 .982 .807 1.195 0 a . . 0 . . . . 7.852E 02 .069 1.306 1 .253 .924 .808 1.058 0 a . . 0 . . . . 8.612E 02 .085 1.022 1 . 312 .917 .776 1.084 0 a . . 0 . . . . .458 .266 2.972 1 .085 1.581 .939 2.663 0 a . . 0 . . . . Intercept [GENDER=0] [GENDER=1] MEANING DEPENDEN SELFCRIT TRAITANX [GENDER=0] * MEANING [GENDER=1] * MEANING [GENDER=0] * DEPENDEN [GENDER=1] * DEPENDEN [GENDER=0] * SELFCRIT [GENDER=1] * SELFCRIT [GENDER=0] * TRAITANX [GENDER=1] * TRAITANX Intercept [GENDER=0] [GENDER=1] MEANING DEPENDEN SELFCRIT TRAITANX [GENDER=0] * MEANING [GENDER=1] * MEANING [GENDER=0] * DEPENDEN [GENDER=1] * DEPENDEN [GENDER=0] * SELFCRIT [GENDER=1] * SELFCRIT [GENDER=0] * TRAITANX [GENDER=1] * TRAITANX Intercept [GENDER=0] [GENDER=1] MEANING DEPENDEN SELFCRIT TRAITANX [GENDER=0] * MEANING [GENDER=1] * MEANING [GENDER=0] * DEPENDEN [GENDER=1] * DEPENDEN [GENDER=0] * SELFCRIT [GENDER=1] * SELFCRIT [GENDER=0] * TRAITANX [GENDER=1] * TRAITANX Attachment Status Dismissing Fearful Preoccupied B Std. Error Wald df Sig. Exp(B) Lower Bound Upper Bound 95% Confiden ce Interval for Exp(B) This parameter is set to zero because it is redundant. a.
109 A fourth multinomial regression analysis was conducted to examine the data on males and females separately. Again, personal meaning, dependency, self criticism, and trait a nxiety were the independent variables. As illustrated in Table 4. 12, the results showed that none of these variables predicted malesâ€™ or femalesâ€™ placement in the four attachment categories. Given these results, the null hypothesis was partially rejecte d regarding gender and personal meaning since they both partially predicted attachment classification. However, it was not rejected regarding dependency, self criticism, trait anxiety and the interactions of all the dependent variables with gender. Tabl e 4. 12 Multinomial Logistic Regression Analyses Output for Attachment S ecurity a s the Outcome Variable with Two Levels Main Effects Only .240 .263 .113 .263 .287 .123 Cox and Snell Nagelkerke McFadden Cox and Snell Nagelkerke McFadden Female Male 259.094 229.764 29.330 12 .004 119.514 104.838 14.676 12 .260 Model Intercept Only Final Intercept Only Final Gender Female Male 2 Log Likelihood Ch i Square df Sig.
110 Hypothesis Three Ho3: There is no significant relationship between college studentsâ€™ attachment status and their level of depressiveness. Simultaneous regression analyses were conducted to test this hypothesis for each factor of depressiveness (dependency and self criticism). Dependen cy and self criticism were used in separate regression models as the outcome variables. Attachment related anxiety, attachment related avoidance, gender, personal meaning, and trait anxiety were the independent variables. 2.855 4.737 .363 1 .547 2.097E 02 .039 .295 1 .587 .979 .908 1.056 2.029E 02 .020 .994 1 .319 .980 .942 1.020 9.373E 03 .021 .206 1 .650 .991 .951 1.032 3.311E 02 .050 .444 1 .505 1.034 .938 1.139 4.916 5.692 .746 1 .388 5.660E 02 .039 2.158 1 .142 .945 .876 1.019 1.706E 02 .026 .442 1 .506 1.017 .967 1.070 1.084E 02 .027 .160 1 .689 1.011 .959 1.066 9.618E 02 .058 2.750 1 .097 1.101 .983 1.234 1.168 4.532 .066 1 .797 4.799E 02 .034 2.030 1 .154 .953 .892 1.018 1.336E 03 .020 .005 1 .946 1.001 .963 1.041 2.434E 0 2 .021 1.296 1 .255 1.025 .983 1.068 1.844E 02 .048 .145 1 .703 1.019 .926 1.120 3.858 6.529 .349 1 .555 9.550E 03 .048 .040 1 .842 .990 .902 1.088 1.110E 02 .026 .183 1 .669 .989 .940 1.041 4.764E 02 .039 1.490 1 .222 1.049 .972 1.132 1.951E 02 .056 .122 1 .727 1.020 .914 1.138 .804 7.742 .011 1 .917 6.259E 02 .053 1.394 1 .238 .939 .847 1.042 1.403E 02 .032 .192 1 .661 .986 .926 1.050 7.759E 02 .047 2.702 1 .100 1.081 .9 85 1.185 6.331E 02 .067 .895 1 .344 .939 .823 1.070 8.994 13.316 .456 1 .499 2.972E 02 .094 .099 1 .753 .971 .807 1.168 7.986E 02 .066 1.473 1 .225 1.083 .952 1.232 .110 .082 1.795 1 .180 1.117 .950 1.313 .440 .2 61 2.831 1 .092 .644 .386 1.075 Intercept MEANING DEPENDEN SELFCRIT TRAITANX Intercept MEANING DEPENDEN SELFCRIT TRAITANX Intercept MEANING DEPENDEN SELFCRIT TRAITANX Intercept MEANING DEPENDEN SELFCRIT TRAITANX Intercept MEANIN G DEPENDEN SELFCRIT TRAITANX Intercept MEANING DEPENDEN SELFCRIT TRAITANX Attachment Status Dismissing Fearful Preoccupied Dismissing Fearful Preoccupied Gender Female Male B Std. Error Wald df Sig. Exp(B) Lower Bound Upper Bound 95% Confidence Interval for Exp(B)
111 As illustrated in Table 4. 13, while using dependency as the outcome variable, the analysis resulted in a significant model (F (5, 149) = 7.949, p < .0001. Adjusted R square = .184), which accounted for 18.4 % of the variance in dependency. It showed that trait anxiety ( t (1, 154) = 3 .394, p < .001), gender ( t (1, 154) = 2.328, p = .021), attachment related avoidance ( t (1, 154) = 2.308, p = .022), and attachment related anxiety ( t (1, 154) = 2.168, p = .032) were significantly related to dependency. Gender and attachment related a voidance were negatively related dependency. On the other hand, there was no significant relationship between personal meaning and dependency ( t (1, 154) = 1.023, p = .308). According to this model the null hypothesis was rejected for the dependency fact or of depressiveness. In other words, both factors of attachment were significantly related to dependency factor of depressiveness. Table 4. 13 Simultaneous Multiple Regression Output Using Dependency Factor of Depressive ness as the Dependent Variable .459 a .211 .184 13.7742 .211 7.949 5 149 .000 Model 1 R R Square Adjusted R Square Std. Error of the Estimate R Square Change F Change df1 df2 Sig. F Change Change Statistics Predictors: (Con stant), Trait Anxiety, Gender, Attachment Avoidance, Attachment Anxiety, Personal Meaning a. 103.935 14.356 7.240 .000 2.389 1.102 .184 2.168 .032 2.463 1.067 .192 2.308 .022 6.009 2.582 .183 2.328 .021 .136 .133 .093 1.0 23 .308 .546 .161 .308 3.394 .001 (Constant) Attachment Anxiety Attachment Avoidance Gender Personal Meaning Trait Anxiety Model 1 B Std. Error Unstandardized Coefficients Beta Standardize d Coefficients t Sig. Dependent Variable: D ependency a.
112 Likewise, a simultaneous multiple regression analysis was used to test the relationship between self criticism and factors of attachment. Self criticism was the dependent variable, while attachment related anxiety, attachment related avoidance, gender, personal meaning, and trait anxiety were the independent variables. As illustrated in Table 4. 14, this analysis resulted in a significant model ( F (5, 149) = 29.657, p < .0001. Adjusted R Square = .4 99). It accounted for almost 50 % of variance in self criticism. According to this model trait anxiety (1, 154) = 5.496, p < .0001), personal meaning (1, 154) = 3.2639, p < .001), gender (1, 154) = 3.394, p < .005), and attachment related anxiety (1, 15 4) = 3.394, p =.002) were significantly related to self criticism. The relationship between personal meaning and self criticism was negative. Of all the independent variables only attachment related avoidance was not significantly related to self critici sm. Therefore, the null hypothesis was only partially rejected. In other words, only the anxiety factor of attachment had a significant relationship with self criticism. The null hypothesis was not rejected regarding the relationship between avoidance a nd self criticism. Table 4. 14 Simultaneous Multiple Regression Output Using Self Criticism Factor of Depressiveness as the Dependent Variable .706 a .499 .482 12.2717 .499 29.657 5 149 .000 Model 1 R R Square Adjusted R Square Std. Error of the Estimate R Square Change F Change df1 df2 Sig. F Change Change Statistics Predictors: (Constant), Trait Anxiety , Gender, Attachment Avoidance, Attachment Anxiety, Personal Meaning a.
113 The hypothesis was further tested with multinomial and binary logistic regression analyse s. A multinomial logistic regression model attachment was used as a categorical variable with four classifications; secure, fearful, preoccupied and dismissing. In the binary model attachment was classified as secure an d insecure (fearful, preoccupied, and dismissing). Attachment classification was the independent variable, whereas gender, personal meaning, dependency, self criticism were the independent variables. These analyses were conducted to examine if dependency significantly predicted college studentsâ€™ membership in categories of attachment. The result showed that dependency, self criticism or their interactions with gender did not significantly predict college studentsâ€™ placement in either four or two categori es of att achment. ( Tables 4. 9 & 4. 10) In summary, dependency was significantly related to both factors of attachment while self criticism was only related to attachment related anxiety. Therefore, with the exception of the relationship between self cri ticism and attachment related avoidance, the null hypothesis was rejected . 96.682 12.790 7.559 .000 3.052 .982 .211 3.107 .002 .790 .951 .055 .831 .407 6.718 2.300 .183 2.921 .004 .387 .118 .236 3.269 .001 .787 .143 .397 5 .496 .000 (Constant) Attachment Anxiety Attachment Avoidance Gender Personal Meaning Trait Anxiety Model 1 B Std. Error Unstandardized Coefficients Beta Standardize d Coefficients t Sig. Dependent Variable: Self Criticism a.
114 Hypothesis Four Ho4: There is no significant relationship between college studentsâ€™ attachment status and their level of trait anxiety. To test this hypothesis simultaneous mu ltiple regression analyses was used. Trait anxiety was the dependent variable. Attachment related anxiety, attachment related avoidance, gender, personal meaning, dependency, and self criticism were the independent variables. As shown in Table 4. 15, thi s analysis resulted in a significant model ( F (6, 148) = 24.262, p < .0001. Adjusted R Square = .496), which accounted for 49.6 % of the variance in trait anxiety. This model showed significant relationships between trait anxiety and all the variables ex cept for the factors of attachment. Personal meaning ( t (1, 154) = 3.714, p < .0001), self criticism ( t (1, 154) = 5.273, p < .0001), dependency ( t (1, 154) = 3.078, p = .002), and gender ( t (1, 154) = 2.988, p = .003) were significantly related to trai t anxiety. Trait anxiety was negatively related to personal meaning and gender. Neither attachment related anxiety ( t (1, 154) = .101, p = .919) nor attachment related avoidance ( t (1, 154) = 1.213, p = .227) had a significant relationship with trait anx iety. Thus the null hypothesis was not rejected. There was no significant relationship between college studentsâ€™ scores on factors of attachment and their level of trait anxiety. Table 4. 15 Simultaneous Regression Output Using Trait Anxiety as the Depe ndent Variable .704 a .496 .475 6.23 .496 24.262 6 148 .000 Model 1 R R Square Adjusted R Square Std. Error of the Estimate R Square Change F Change df1 df2 Sig. F Change Change Statistics Predictors: (Constant), Self Criticism, Gender, Dependency, Attac hment Avoidance, Attachment Anxiety, Personal Meaning a.
115 These results were confirmed by logistic regression analyses in which attachment was a categorical variable with either four (secure, fearful, preoccupied and dismissing) or two level s (secure and insecure). Attachment classification was the outcome variable, while trait anxiety was among the independent variables. These analyses intended to examine if trait anxiety significantly predicted likelihood of membership in the categories o f attachment. The results showed that neither trait anxiety nor its interaction with gender significantly predicted college studentsâ€™ membership in the four or two att achment categories. See Tables 4. 9 and 4. 10 for the results. Hypothesis Five Ho5: Th ere is no significant relationship between college studentsâ€™ gender and their attachment status, levels of personal meaning, depressiveness, and trait anxiety. A series of independent t tests and multinomial logistic regression analysis were used to test t his hypothesis, which was partially rejected. As reported in Table 4. 16, there were significant differences between males and females only on attachment related avoidance, dependency, and trait anxiety. 15.425 8.453 1.825 .070 5.294E 02 .522 .007 .101 .919 .594 .490 .082 1.213 .227 3.542 1.185 .191 2.988 .003 .221 .060 .268 3.714 .000 .111 .036 .196 3.078 .00 2 .201 .038 .399 5.273 .000 (Constant) Attachment Anxiety Attachment Avoidance Gender Personal Meaning Dependency Self Criticism Model 1 B Std. Error Unstandardized Coefficients Beta Standardize d Coefficients t Sig. Dependent Varia ble: Trait Anxiety a.
116 In other words, there were no significant differe nces between males and females on measures of attachment related anxiety, personal meaning, and self criticism. There was no significant difference between males and females on attachment related anxiety ( t (153) = .932, p = . 353). On the other hand, th ere was a significant difference between males and females on attachment related avoidance ( t (153) = 3.226, p = .002). More specifically, males seem to score significantly higher on attachment related avoidance than females, means being 3.19 and 2.55. S ee Table 4. 16 t tests results according to gender. Given these results, the null hypothesis was partially rejected. It was rejected with respect to attachment related anxiety. However, it was not rejected with respect to attachment related avoidance. The relationship between gender and attachment was also examined with a series of logistic regression analyses, in which attachment was the outcome variable with either two (secure and insecure) or four categories (secure, fearful, preoccupied, and dismiss ing). This model only included main effects. Gender was one of the independent variables. As illustrated in Tables 4. 9 and 4. 10 , this analysis showed that gender significantly predicted likelihood in attachment classifications. More specifically, when comparing secure versus dismissing categories, females were more likely to be in the secure category while males were more likely to be in the dismissing one. Similarly, while comparing secure versus fearful categories, females were more likely to be in t he secure category whereas males were more likely to be in the fearful category. Gender did not predict individualsâ€™ placement in the preoccupied category of attachment. Another multinomial logistic regression model including main and interaction effect s of gender was conducted to predict placement in four categories of attachment.
117 No significant main effects or interaction effects of gender were found. These results were confirmed with an additional multinomial regression analysis which examined data on males and females separately. This model used attachment as the outcome variable with four levels. Personal meaning, dependency, self criticism, and trait anxiety were the independent variables. No main effects of these variables were found for eith er males or females in predicting placement in the attachment categories. A binary logistic regression model was used to investigate if gender predicted security of attachment. In this model attachment was classified as secure and insecure (combination of fearful, preoccupied, and dismissing categories.). The results showed that gender did not significantly predict security of attachment. These results show that that gender significantly predicted attachment regarding the secure, fearful and dismissin g categories when only main effects were investigated to predict placement in four categories of attachment. Further analyses did not show any significant gender main effects or interactions regarding any variables, including attachment classification. S ee Tables 4. 9 and 4. 10 for the result. Results of t tests also showed that there was no significant difference between male and female means on personal meaning ( t (153) = .034, p = .973). Thus the null hypothesis was not rejected for personal meaning. As seen in Table 4. 16 the mean scores on personal meaning were 71.85 and 71.92 for males and females respectively.
118 Table 4. 16 Independent Sample t Tests Output for Attachment Related Anxiety, Attachment Related Avoidance, Personal Meaning, Depen dency, Self Criticism and Trait Anxiety According to Gender There was no significant difference between male and female means on the self criticism factor of depressiveness. On the other hand, th ere was a significant difference between males and females on dependency ( t (153) = 3.928, p < .0001). More specifically, females scored overall higher than males on the dependency scale. 48 2.7581 1.0950 107 2.9486 1.2113 48 3.1979 1.1401 107 2.5524 1.1568 48 71.85 8.88 107 71.92 11.08 48 125.3792 13.9987 107 135.3297 14.8319 48 113.9375 11.4118 107 109.9907 18.9670 48 34.79 7.99 107 38.25 8.68 Gender Male Female Male Female Male Female Male Female Male Female Male Female Attachment Anxiety Attachment Avoidance Personal Meaning Dependency Self Criticism Trait Anxiety N Mean Std. Deviation .739 .391 .93 2 153 .353 .968 99.471 .335 .008 .931 3.226 153 .002 3.244 91.758 .002 2.125 .147 .034 153 .973 .037 111.478 .971 .510 .476 3.928 153 .000 4.016 95.516 .000 14.512 .000 1.336 153 .184 1.601 140.199 .112 .183 .670 2.351 153 .020 2.427 97.849 .017 Equal variances assumed Equal variances not assumed Equal variances assumed Equal variances not assumed Equal variances assumed Equal variances not assumed Equal variances assumed Equal variances not assume d Equal variances assumed Equal variances not assumed Equal variances assumed Equal variances not assumed Attachment Anxiety Attachment Avoidance Personal Meaning Dependency Self Criticism Trait Anxiety F Sig. Levene's Test for Equality of Va riances t df Sig. (2 tailed)
119 Therefore, the null hypothesis was only partially rejected rega rding depressiveness and its factors. Finally, the t test showed that there was a significant gender difference on trait anxiety ( t (153) = 2.351, p = .02). Overall, females scored higher on trait anxiety than did males. Thus the null hypothesis was rej ected regarding trait anxiety. Hypothesis Six Ho6: There is no significant relationship between college studentsâ€™ level of personal meaning and their level of depressiveness. To test this hypothesis, a simultaneous multiple analysis was conducted using p ersonal meaning as the outcome variable. Attachment related anxiety, attachment related avoidance, dependency, self criticism, and trait anxiety were the independent variables. According to this model, there was no significant and negative relationship b etween personal meaning and self criticism ( t (1, 154) = 3.348, p < .001). There was no significant relationship between personal meaning and dependency factor of depressiveness ( t (1, 154) = .989, p = .324). Keep in mind that in these models personal m eaning was the dependent variable and factors of depressiveness were among the independent variables ( Table 4. 8). These results remained the same when the factors of depressiveness were dependent variables and personal meaning was among the independent va riables ( Tables 4. 13& 4. 14 ) . Given these results, the null hypothesis was only partially rejected. There were no significant relationships between personal meaning and the dependency factor of depressiveness. There was significant relationship between pe rsonal meaning and self criticism.
120 Therefore, the null hypothesis was rejected regarding self criticism and personal meaning. It was not rejected with respect to the dependency factor of depressiveness. Hypothesis S even Ho7: There is no significant rel ationship between college studentsâ€™ level of personal meaning and their level of trait anxiety. Simultaneous multiple regression analysis was used to test this hypothesis. Regardless of whether personal meaning or trait anxiety was the dependent variable there was a significant and negative relationship between personal meaning and trait anxiety. See Tables 4. 8 and 4. 15 for outputs of these results. These results were confirmed when personal meaning was the dependent variable and only trait anxiety was used as the independent variable. As reported in Table 4. 17 there was a significant and negative relationship between personal meaning and trait anxiety ( F (1, 153) = 60.969, p < .0001. Adjusted R Square = .280; t ( 153) = 7.808 p < .0001). Personal mea ning accounted for 28 % of variance in trait anxiety. In short, the null hypothesis was rejected. That is to say, there was a significant relationship between college studentsâ€™ level of personal meaning and their level of trait anxiety. Table 4. 17 Simult aneous Multiple Regression Analysis Output For Personal Meaning And Trait Anxiety .534 a .285 .280 8.84 Model 1 R R Square Adjusted R Square Std. Error of the Estimate Predictors: (Constant), Trait Anxiety a.
121 Hypothesis Eight Ho8: There is no significant relationship between college studentsâ€™ level of depressiveness and t heir level of trait anxiety. As shown in Table 4. 15 among other variables, factors of depressiveness were the independent variables in the simultaneous multiple regression equation predicting trait anxiety. These analyses showed a significant relationship between trait anxiety and both factors of depressiveness; self criticism ( t (1, 154) = 5.273, p < .0001) and dependency ( t (1, 154) = 3.078, p < .002). Furthermore, simultaneous regression analyses were conducted using factors of depressiveness as the ou tcome variables, while trait anxiety was among the independent variables ( Tables 4. 13 & 4. 14). Again, trait anxiety was significantly related to dependency and self criticism. Therefore, the null hypothesis was rejected. In other words, there was a sign ificant relationship between college studentsâ€™ level of depressiveness and their level of trait anxiety. Hypothesis Nine Ho9: There is no significant relationship among college studentsâ€™ attachment status, level of personal meaning, level of depressiveness , and level of trait anxiety. To test this hypothesis a correlation matrix was used. As seen in Table 4. 1 8 , the results showed that the hypothesis was rejected for the most part. The following are some of the strongest correlation coefficients, which are listed from higher to lower; self a 95.956 3.162 30.346 .000 .647 .083 .534 7.808 .000 (Constant ) Trait Anxiety Model 1 B Std. Error Unstandardized Coefficients Beta Standardize d Coefficients t Sig. Dependent Variable: Personal Meaning a.
122 criticism and trait anxiety (r= .583), personal meaning and self criticism ( r = .553), personal meaning and trait anxiety ( r = .534), attachment related anxiety and self criticism ( r = .468), personal meaning and attach ment related anxiety ( r = .414) . Also see Table 4. 1 8 for all the coefficients. Table 4. 18 Correlation Matrix for Attachment Related Anxiety, Attachment Related Avoidance, Personal Meaning, Dependency, Self Criticism, and Trait Anxiety These results indicate that attachment related anxiety and trait anxiety were related to all the variables. Self criticism and depressiveness were related to all the variables except dependency, which only had a weak correlation with attachme nt related anxiety ( r = .213). Finally, attachment related avoidance was significantly related to all the variables, except the dependency factor of depressiveness. In short, the null hypothesis was rejected for most of the interrelations among the varia bles. 1.000 . 155 .345 ** 1.000 .000 . 155 155 .414 ** .305 ** 1.000 .000 .000 . 155 155 155 .213 ** .133 .089 1.000 .008 .099 .272 . 155 155 155 155 .468 ** .336 ** .553 ** .131 1.000 .000 .000 .000 .103 . 155 155 155 155 155 .390 ** .226 ** .534 ** .321 ** .583 ** 1.000 .000 .005 .0 00 .000 .000 . 155 155 155 155 155 155 Pearson Correlation Sig. (2 tailed) N Pearson Correlation Sig. (2 tailed) N Pearson Correlation Sig. (2 tailed) N Pearson Correlation Sig. (2 tailed) N Pearson Correlation Sig. (2 tailed) N Pe arson Correlation Sig. (2 tailed) N Attachment Anxiety Attachment Avoidance Personal Meaning Dependency Self Criticism Trait Anxiety Attachment Anxiety Attachment Avoidance Personal Meaning Dependency Self Criticism Trait Anxiety Correlat ion is significant at the 0.01 level (2 tailed). **.
123 Summary The purpose of this study was to investigate if college studentsâ€™ attachment orientation and gender had any significant relationships with measures of well being (personal meaning) and those of vulnerability to psychopathology (depressivenes s and trait anxiety). In the analyses attachment was an independent variable while used as a continuous variable. In additional analyses attachment was used as a categorical variable consisting of four classifications (secure, fearful, preoccupied, and d ismissing) as well as of two categories (secure and insecure). These categorizations were made in order to make comparisons with findings of previous research most of which involved the use of attachment as a categorical variable. The first hypothesis e xamined the relationship between attachment orientation and personal meaning. When attachment was used as a continuous variable consisting of dimensions of anxiety and avoidance, neither factor had a significant relationship to personal meaning. Further investigation of these two constructs was done in hypothesis two. The second hypothesis investigated if gender, personal meaning, dependency, self criticism, and trait anxiety significantly predicted individualsâ€™ placement into attachment classifications . Attachment was the outcome variable with four levels; secure, fearful, preoccupied, and dismissing. In an additional analysis it was categorized into secure versus insecure. Of the five independent variables only gender and personal meaning significan tly predicted studentsâ€™ likelihood for being the individual categories of attachment . The results showed that when the four categories were used p ersonal meaning significantly predicted the fearful and secure categories. Individuals with higher
124 scores we re more likely to be in the secure category while those with lower scores were more likely to be in the fearful category, which corresponds to high scores on avoidance and anxiety. Gender also significantly predicted attachment categories with the excepti on of the preoccupied one. When comparing dismissing versus secure attachment, females were more likely to be in the secure category while males were more likely to be in the dismissing one. When comparing fearful versus secure categories, females were m ore likely to be in the secure categories whereas males were more likely to be in the fearful one. On the other hand, there were no more significant main or interaction effect in any of the analyses regarding any of the other variables ( dependency, self c riticism, and trait anxiety ) The third hypothesis sought relationships between attachment orientation and depressiveness. When attach ment was used as an independent continuous variable, the simultaneous regression analysis s howed that both attachment fact ors were significantly related to the dependency factor of depressiveness. On the other hand, only attachment related anxiety was significantly related to self criticism. There was no significant relationship between attachment related avoidance and self criticism. When attachment was used as an outcome categorical variable, neither factor of depressiveness nor their interactions with gender significantly predicted college studentsâ€™ membership in the two (secure versus insecure) or four attachment catego ries of attachment (secure, fearful, preoccupied, and dismissing) Hence, for the most part, factors of depressiveness were significantly related to dimensions of attachment. However, factors of depressiveness did not significantly predict college student s â€˜membership in specific attachment classification .
125 The fourth hypothesis examined if attachment orientation was significantly related to trait anxiety. The results indicated that neither factor of attachment was significantly related to trait anxiety. Likewise, when att achment was conceptualized as a categorical variable, trait anxiety or its interaction with gender did not significantly predict studentsâ€™ placement in specifi c attachment classifications. In the fifth hypothesis male and female stude nts were significantly different only on attachment related avoidance, dependency, and trait anxiety. Male scores were overall higher on avoidance whereas female scores were higher on dependency and trait anxiety. Males and females did not significantly differ on other measures. When attachment was conceptualized as a categorical variable with four levels, gender was significantly predicting of membership in 3 of the four categories; secure, fearful , and dismissing. When comparing dismissing versus secu re attachment, females were more likely to be in the secure category while males were more likely to be in the dismissing one. When comparing fearful versus secure categories, females were more likely to be in the secure categories whereas males were more likely to be in the fearful one. However, there were no more significant main or interaction effect in any of the analyses regarding any of the other variables ( dependency, self criticism, and trait anxiety ). In the sixth hypothesis the relationship be tween personal meaning and factors of depressiveness was examined. The results indicated that only the self criticism factor of depressiveness had a significant and negative relationship with personal meaning. Dependency did not have a significant relati onship with personal meaning.
126 In the seventh hypothesis the relationship between personal meaning and trait anxiety was examined. In all the analyses personal meaning and trait anxiety had a significant and negative relationship. In the eighth hypothe sis the relationship between trait anxiety and depressiveness was examined. Trait anxiety was significantly related to both factors of depressiveness (dependency and self criticism). In the ninth hypothesis of the study correlations among all the variable s of the study were examined. Attachment related anxiety and trait anxiety were significantly related to all the variables of the study. Personal meaning and self criticism were significantly related to all the variables except dependency. In short, th e result of this study showed mixed results regarding relationships of attachment and gender to personal meaning, depre ssiveness, and trait anxiety. Attachment factors were only related to factors of depressiveness with the exception of attachment related avoidance and self criticism. Gender significantly predicted studentsâ€™ placement in 3 of the 4 categories of attachment. The findings of the study are only partially in line with the existing literature.
127 CHAPTER 5 DISCUS S ION In t his study relationships between attachment status and measures of well being and vulnerability to psychopathology were investigated . More specifically, the relationships among attachment status, gender , personal meaning, depressiveness, and trait anxiety among traditional â€“ age college students (18 23) were examined . Personal meaning was chosen as a measure of well being, while depressiveness and trait anxiety were measures of vulnerability to psychopath ology. There were three goals for the study: to contribute to the understanding of college student functioning; to contribute to the research in the fields of student mental health and attachment; and to provide insights for clinical work with young ad ult college students. This chapter consists of the studyâ€™s summary, discussion of results, and implications for theory, research and clinical practice, its limitations, and recommendations for future research. Summary This study sought to ex tend attachment research (Bowlby, 1969, 1973, 1982; Ainsworth et al., 1978; Main, 1990; Hazan & Shaver, 1987; Brennan et al., 1998; Fraley et al., 2000) to college student mental health. The focus was on relationships between attachment status, gender, personal meaning, depressiveness, and trait anxiety.
128 Attachment status was conceptualized in accordance with the research tradition of Brennan et al. (1998) and Fraley et al. (2000). Consequently, attachment orientation was viewed on dimensions of anxiety, and avoidance in close romantic relationships. Att achment was also conceptualized as a categorical construct in order to make comparisons to findings of previous research. According to Brennan et al. (1991) this categorization should be done in the following manner: (a) Low scores on both avoidance and a nxiety dimensions correspond to the secure category. (b) High scores on the avoidance and low scores on anxiety dimensions correspond to avoidant (dismissing) category. (c) High scores on the anxiety and low scores on the avoidance dimension correspond to the anxious ambivalent (preoccupied) category of attachment. (d) High scores on both anxiety and avoidance dimensions correspond to the fearful classification of attachment. These four classifications were originally developed by Bartholomew and Horowit z (1991). These categories were also converted intro secure and insecure classifications. While secure category remained the same, the insecure category was composed of fearful, preoccupied, and dismissing classifications. The concept of personal meani ng was adapted from Battista, and Almond (1973). Thus it refers to oneâ€™s beliefs about the degree to which he or she perceives fulfilling a â€œpositively valued life framework or life goalâ€ (Battista, Almond, 1973, p. 409). Personal meaning was measured wi th the LRI R (Debats, 1998). Depressiveness was adopted from the research tradition of Blatt et al. (1976) who viewed it as consisting of three factors: dependency, self criticism and efficacy. Researchers have often dismissed the self efficacy subscal e (i.e. Blatt, Quinlan, Chevron, McDonald, & Zuroff, 1982; Santor et al., 1997; Santor & Zuroff, 1998; Zuroff et al.,
129 1990). Thus only the first two factors were used in this study. Dependency refers to excessive reliance â€œon others to provide a sense of well being (Blatt et al., 1976, p. 528). On the other hand, self criticism has to do with tendencies of excessive perfectionism. Trait anxiety was adopted from Spielberger (1983) who defined it as the degree to which persons perceive a variety of situatio ns as threatening and as the tendency to respond to such circumstances with increased anxiety. As noted by its creators, the construct and the instrument tap into individual differences in the frequency and intensity of experiencing such anxiety states. S imultaneous multiple regression analysis, multinomial and binary logistic regression analyses, independent t tests, and correlation matrix were used to examine relationships between attachment status, gender, personal meaning, depressiveness, and trait anx iety. These analyses resulted in mixed findings, which were only partially inline with the findings of prior research. Discussion of Results Attachment and P ersonal M eaning The first general question of this study inquired the relationship between attachm ent status and personal meaning. A relationship was expected because the existing literature reports links between attachment status and various measures of well being. However, no studies were found in the literature that specifically addressed attachme nt status and personal meaning. One study has done so rather indirectly by examining attachment status and sense of coherence (Rood, 2000). As conceptualized by Antonovsky (1987) sense of coherence involves personal meaning as one of its dimensions. Roo d (2000) did find a positive relationship between secure attachment and
130 sense of coherence. He also found a negative relationship between sense of coherence and insecure styles of attachment. Findings of the current study did not show any significant rela tionships between factors of attachment (anxiety and avoidance) and personal meaning. However, when attachment was used as a categorical outcome variable with four levels; secure, fearful, preoccupied, and dismissing, personal meaning significantly predic ted placement in two of the attachment categories. When comparing secure versus fearful classifications, lower scores on personal meaning predicted fearful attachment whereas higher scores predicted secure attachment. This significance was limited to thi s particular comparison however. When attachment was formulated as secure versus insecure, personal meaning did not predict college studentsâ€™ placement in these two categories. One would expect that since placement in secure versus fearful categories wa s significantly predicted by personal meaning, such significance would also exist regarding secure versus insecure categories. Secure attachment corresponds to low scores on the both factors of attachment, whereas fearful category corresponds to high scor es on these factors. Since personal meaning or its interaction with gender did not predict membership in the other categories of attachment and since the simultaneous regression analysis did not reveal any significant relationships between personal meanin g and dimensions of attachment, it is clear that attachment and personal meaning are only associated with respect to scores on the lower and higher ends of both factors attachment. These findings imply that personal meaning, as measured by the LRI R (Deb ats, 1998), is associated with attachment only when individuals score either very high or very low on both attachment related anxiety and attachment related avoidance.
131 According to Bartholomew and Horowitz (1991) individuals in the fearful categories hav e negative internal working models of self and others. On the other hand, the preoccupied category is associated with positive internal working models of others and negative working models of self. Dismissing category is linked with positive models of se lf and negative models of others. Secure category corresponds to positive internal models of self and other. Considering the results of this study and how Bartholomew and Horowitz conceptualized the four categories of attachment some tentative inferences can be drawn. Initial attachment research on infant caregiver relationship used a three category model (Ainsworth et al., 1978) which did not include the fearful classification. Therefore, developmental inferences made my attachment researchers did not include this category. These inferences often identify preoccupied and dismissing categories as a risk factor for less favorable developmental outcomes. On the other hand, as Main (1991) pointed out, not all infants with insecure attachment grow to deve lop psychopathology. Main indicated that those who did not have a formed attachment style were more likely to be at such risk. Individualsâ€™ interpersonal relationships constitute one of the most vital domains of their experience. Thus an overall ease, o ptimism and fulfillment in that domain might contribute to a more positive outlook about the life one is leading. Consequently, it is possible that when individuals have positive views of themselves or others such positivity might ease their dealings with challenges of life by providing a sense of resourcefulness. Conversely, a lack of positive views of self and others might be associated with a sense of alienation and meaninglessness.
132 According to attachment theory , i ndividu a ls form specific attachmen t orientations in early years of development. Therefore, they might form their meaning systems in ways that are congruent with these orientations. Accordingly, regardless of what specific insecure attachment style they utilize , persons might find their l ives purposeful and fulfilling as long as such styles do not predispose them to highly challenging developmental pathways as it might be the case in disorganized or fearful attachment . One would also expect links between fearful attachment, depressivenes s, and trait anxiety. Since these associations were not established by the findings of this study it is possible that individuals with preoccupied style of attachment might use unique affect regulation strategies. Typically high attachment related anxiet y is related to hyperactivating strategies, whereas high avoidance is associated with deactivating affect regulation strategies. Given that individuals with fearful attachment have high degrees of anxiety and avoidance, their affect regulation tactics mig ht be more complex than those in the secure, preoccupied or dismissing categories. When attachment was conceptualized as a continuous variable, there was a significant relationship between personal meaning, trait anxiety and self criticism. Both trait a nxiety and self criticism had a negative relationship with personal meaning and when combined they accounted for greater variance than did the two factors of attachment. Each set accounted for 33 % and 19 % of the variance respectively. No previous studi es examining personal meaning, trait anxiety and self criticism were found for comparison. A general vulnerability to anxiety might interfere with oneâ€™s felt comfort with the self and with oneâ€™s involvements. Accordingly, higher degrees of trait anxiety might impact the overall satisfaction and meaning individuals attribute to their present
133 experiences. In addition, ongoing high degrees of anxiety might interfere with setting goals and following through with their execution, which might in turn interfere with the potential fulfillment one can obtain from such pursuits. In other words, higher levels of ongoing anxiety might interfere with purposeful behavior, which would in turn impact individualsâ€™ attributions of meaning to their lives. Indeed, personal meaning, as used in this study, is considered to consist of a positive life framework and positive goal fulfillment. To put in simple terms, higher scores on personal meaning indicate that persons conceive their lives as having purpose (framework) and th eir personal involvements and activities as serving to these goals thus leading to some degree of fulfillment. Blatt et al. (1976) referred to self criticism as having to do with individualsâ€™ evaluations of their actual selves (as opposed to the ideal self ). As a factor of depressiveness it was correlated with negative evaluations of self (as in guilt and self blame) and of oneâ€™s experience. Self criticism is also referred to as maladaptive perfectionism, which involves having extremely high standards for achievement and being highly critical of oneâ€™s performance. Consequently, a significant negative relationship between personal meaning and self criticism is inline with the existing body of literature on self criticism. Another reason for the unexpected findings regarding factors of attachment and personal meaning might be due to the measurement instrument. Although the Life Regard Index Revised (LRI R; Debats, 1998) has well documented psychometric properties, it is not widely used. This sample had hi gh scores on the LRI R with a mean score of 71.9. Due to their age and their experience in college, traditional age students
134 have stressful schedules, complex challenges, uncertainties about their future, and they are in the process forming their adult id entities. Hence, it could be argued that given this high degree of developmental activity, college students may not be as clear about their life goals (framework) and may not be as content with their current lives (fulfillment) as indicated by scores on t he LRI R. In other words, it is reasonable to argue that growth at such developmentally sensitive periods might inevitably coincide with considerable degree of discontent with ones existing life in order for significant change to occur. Therefore, furthe r research is needed to investigate the validity of the LRI R. Males and females did not differ on personal meaning. This similarity might be attributed to similarities in the participantsâ€™ characteristics. For instance, regardless of their gender, parti cipants were of above average academic standing as students in a highly competitive institution. Since the samples were from students in 3000 and 4000 level courses, the participants were likely to be in the last two years of their undergraduate st udies. They typically have chosen their major area of study. As a result, regardless of their gender, these students might perceive a sense of being on track and having a sense of purposefulness. This study examined studentsâ€™ self reports of their reflec tions of how meaningful they find their lives. It did not focus on specific reasons for their attributions. Males and females might find their lives meaningful for different reasons. This finding is consistent with prior studies on personal meaning. De bats (1998) reports that no gender differences were found in any previous samples.
135 Attachment and Depressiveness This study investigated relationships between factors of attachment and depressives. In addition it examined if factors of depressiveness significantly predicted individuals placement in categories of attachment. These inquiries resulted in mixed results. When attach ment was used as an independent continuous variable, the simultaneous regression analysis s howed that bot h attachment factors were significantly related to the dependency factor of depressiveness. On the other hand, only attachment related anxiety was significantly related to self criticism. There was no significant relationship between attachment related a voidance and self criticism. However, w hen attachment was used as an outcome categorical variable, neither factor of depressiveness nor their interactions with gender significantly predicting college studentsâ€™ membership in the two (secure versus insecure ) or four categories of attachment (secure, fearful, preoccupied, and dismissing) . In short, while , factors of depressiveness were for the most part significantly rela ted to dimensions of attachment, factors of depressiveness did not significantly predict college students â€˜membership in specific attachment classifications . Prior research and theory suggest relationships between factors of attachment (anxiety and avoidance) and those of depressiveness (dependency and self criticism). For instance, a strong relationship was expected between dependency and attachment related anxiety. Individuals with high degrees of dependency are on the one hand, characterized with having ongoing intense longings for intimate relationships and on the other hand, not having much faith that these needs will be met (Zuroff & Fitzpatrick, 1995). As a
136 result, they are thought to have fears of abandonment. These experiences have a great deal of theoretical relevance to the characteristics of anxious ambivalent (pr eoccupied) attachment (Kobak, Sudler, & Gamble, 1991; Zuroff & Fitzpatrick, 1995). This relationship was endorsed by the findings of this study. There was a significant positive relationship between attachment related anxiety and dependency. Moreover, o f all the variables dependency had the strongest correlation with attachment related anxiety ( r = 21). According to attachment research dependency should have a significant but negative relationship to attachment related avoidance. For instance, Zuroff and Fitzpatrick (1995) found a significant negative correlation of r = .28. The results of this study did show a similar but weaker relationships with a correlation coefficient of r = .13 between dependence and avoidance. In this study, attachment related anxiety had a significant relationship to self criticism. Persons with attachment related anxiety are known to have negative internal working models of self while having positive models of others. Such self perception could also be part of the perfectio nist strivings of self critics. There was a positive correlation between these two variables ( r = 46). According to prior research a significant relationship would be expected between self criticism and attachment related avoidance. Zuroff and Fitzpatric k (1995) report that self critical individuals have significant concern about obtaining approval of others, whereas avoidant persons do not. They also add that self critical persons report low self esteem while avoidant persons often report positive self images. In this study, there was no significant relationship between self criticism and attachment related avoidance.
137 Zuroff and Fitzpatrick (1995) showed that self criticism was associated with fearful attachment (Bartholomew & Horowitz, 1991) rather t han dismissing attachment style (high scores on both factors of attachment). This finding was not confirmed by this study when attachment style (four levels) was used as the outcome variable. Self criticism, dependency or their interactions with gender d id not significantly predict placement in categories of attachment, regardless of whether a two or four category model was used. Thus, there were discrepancies between results of the two sets of analyses, namely those that used attachment as a continuous independent variable and those in which attachment was a categorical dependent variable. A possible reason for these discrepancies might be that as recommended by Fraley et al. (2000) converting scores on the ECR R may limit the psychometric utility of th e test. Fraley (2003) highly opposes to categorization of scores on the instrument noting that it would reduce its precision and, thus statistical power. In the current study such categorization was made simply for the convenience of making comparisons with previous research easier since most of these studies used categorical models of attachment. However, given the results, this may not advisable for future research. Attachment and Trait Anxiety This study inquired if there is a significant relations hip between attachment status and trait anxiety. No significant relationship was detected between attachment and trait anxiety. Neither attachment related anxiety nor attachment related avoidance was significantly related to trait anxiety. Furthermore, trait anxiety or its interaction with gender did not significantly predict attachment categories. No studies were found in the literature that specifically examined these two variables. However, one would expect that
138 high scores on attachment related anx iety and low scores on avoidance, which correspond to the preoccupied attachment style, would be significantly related to trait anxiety. Typically, persons with this attachment style are thought to have hyperactive affect regulation strate gies (Main, 1990). These strategies involve anxiety regarding parent/partnerâ€™s availability and their responsiveness to the personâ€™s needs. Thus some degree of anxiety in close relationships would be expected. A significant and positive relationship bet ween attachment related anxiety and trait anxiety would be consistent with the premises of attachment theory. One reason for the contrary results might be that the anxiety inherent in the preoccupied attachment style (high score on the attachment related anxiety scale) might be specific to the relational domain. It may not be generalized to other areas of functioning. As a result, the individual may not perceive him or herself as an overall anxious person. A negative relationship between attachment relat ed avoidance and trait anxiety would also be expected. Since such individuals (dismissing) are known to use deactivating strategies of affect regulation, they would be expected to perceive lower degrees of anxiety in a variety of situations. This expecta tion also was not endorsed by the findings of this study. It is possible that the deactivating strategies of affect regulation could be particularly related to close interpersonal relationships and may not be generalized to a variety of other domains. The se findings pose some threat to the premises of attachment theory, which asserts that affect regulation strategies that were originally developed with respect to the availability and proximity of the attachment figure are later generalized to other domains of functioning. The findings of this study show that attachment is not related to a general
139 vulnerability to experiencing anxiety states in a variety of situations. This implies that affect regulation strategies specified by attachment theorists may be specific to close interpersonal relationships. However, such conclusions should be made with caution because this study found a relationship between attachment and vulnerability to depression. This might mean that affect regulation strategies formed in e arly close relationships are related to certain affective states but not others. Future research should address relationships between attachment orientation and a variety of affective experiences. Gender Kobak, Sudler, and Gamble (1991) advocate that being female constitutes a significant risk factor for depression, particularly in adolescents. Attachment researchers typically associate preoccupied attachment with females and dismissing attachment with males. For instance, in their st udy with adolescents, Rosenstein and Horowitz (1996) found that males were more likely to have a dismissing attachment style than the preoccupied one (66 % vs. 34 %) and to have dismissing attachment style than females (75 % vs. 25 %). On the other hand, females were more likely to have a preoccupied attachment style than a dismissing one (68 % vs. 25) and to have preoccupied style than males (63 % vs. 37 %). Results of this study indicated that when comparing dismissing versus secure attachment, females were more likely to be in the secure category while males were more likely to be in the dismissing one. When comparing fearful versus secure categories, females were more likely to be in the secure categories whereas males were more likely to be in the f earful one. Although attachment research frequently reports that females are
140 more likely to be in the preoccupied attachment category, this claim was not endorsed by the findings of this study. The results were similar when attachment was used as contin uous variable. Male and female students did differ on attachment related avoidance. Males scored significantly higher than females on avoidance, which is inline with the previous literature. On the other hand, males and females did not d iffer on the anxiety factor of attachment. Likewise, with the exception of dependency factor of depressiveness, no gender differences were detected for the rest of the variables of the study; personal meaning, self criticism, and trait anxiety. The limi ted gender differences found in this study could be due to several unique aspects of studentsâ€™ experience in college. Students of this age group differ from other populations. First, although such information was not obtained from the sample, it is safe to assume that the majority of the participants were not living in the same town with their parents. As such, regardless of their gender these students might have similar concerns and needs about intimate relationships. That could be so particularly in t he absence of their family. Second, at ages 18 through 23, part of these studentsâ€™ developmental tasks is to acquire competencies in close relationships. These similarities in their developmental needs might be part of the reason for the unexpected simil arities in the participantsâ€™ scores on attachment related anxiety. Fraley (2003) also did not find any gender differences on attachment related anxiety. These results were obtained from a general population sample of over 20 thousand persons who filled ou t Experiences in Close Relationships Revised (Fraley et al, 2000). Therefore, a lack of gender differences could be due to the instrument itself.
141 As mentioned above, unlike findings of Fraley (2003) the current study found a significant difference betwee n males and females on attachment related avoidance. This finding is consistent with studies using other instruments measuring adult attachment. Earlier studies found that compared to females, males were more likely to use deactivating affect regulations strategies, which correspond to the dismissing attachment category. However, attachment related anxiety or preoccupied attachment was not related to either gender. This result might be due to the nature of the attachment instrument (ECR R; Fraley et al. , 2000). Personal Meaning and Depressiveness First, considering that prior research found significant degrees of shared variance between depressiveness and symptoms of clinical depression, a significant negative relationship is expected between depressiv eness and personal meaning. Clinical depression is related to negative perceptions about the self (worthlessness) and about circumstances of oneâ€™s life (helplessness and hopelessness). These negative evaluations are thus associated with felt meaninglessn ess. Second, considering that depressiveness involves negative evaluations of oneâ€™s actual self (perfectionism/self criticism) and excessive anxiety and reliance on others (dependency), one would expect these characteristics to be associated with fluctuat ions in a sense of purpose and fulfillment. Results of this study showed that self criticism had a significant and negative relationship with personal meaning. Dependency did not have a significant relationship to personal meaning. Individuals scoring h igh on self criticism, often referred to as perfectionists, are thought to have high standards for themselves. Thus these persons are overly critical of their performance. Consequently, a significant negative relationship
142 between self criti cism and personal meaning is consistent with the conceptualization of both factors. Persons with perfectionism are thought to have ongoing self consciousness about their performance in various tasks, and have tendencies of ever seeking self improvement. This close observance and monitoring of oneâ€™s performance is highly likely to co occur with a certain degree of ongoing anxiety. Such critical attitude might impact personsâ€™ evaluations of their performance on their hard to achieve goals. There was no sig nificant relationship between personal meaning and the dependency factor of depressiveness. One possible explanation is that regardless of their interpersonal neediness, individuals scoring high on dependency might in fact find reasonable degrees of purpo se and fulfillment in their lives by striving to attain their dependency needs. They might find their lives having purpose even if at a given time such needs are not met. This study did not use the specific subscales of personal meaning (framework and fu lfillment). If these subscales were used it is possible that high scores on dependency could correspond to high scores on the framework scale. On the other hand, their scores on the fulfillment scale could have been dependent upon whether or not the part icipants felt their dependency needs were met for the time being. Personal Meaning and Trait Anxiety The results showed a significant but negative relationship between personal meaning and trait anxiety. High levels of ongoing vulnerability to anxiety may affect individualsâ€™ sense of fulfillment. This anxiety might be associated with negative views, expectations and perceptions of oneâ€™s performance on tasks. Individuals with overall higher degrees of anxiety may be more self critical. The findings showe d a correlations of r = .58 between self criticism and trait anxiety. This implies that even if individuals
143 with higher scores on trait anxiety had positive goals and were doing well on fulfilling these goals, they may be more likely to view these accompl ishments more negatively than those lower on trait anxiety. In other words, higher degrees of anxiety may interfere with positive appraisals of ones goals and accomplishments. Given that trait anxiety has to do with fear (anxiety) about a variety of situ ations, it might interfere with a sense of fulfillment and content with oneâ€™s life. Depressiveness and Trait Anxiety Given how Blatt et al. (1976) conceptualized depressiveness there should be a significant and positive relationship between depressiveness and trait anxiety. The results of this study support their conceptualization. Factors of depressiveness were both significantly related to trait anxiety. It is safe to reason that ongoing self criticism (perfectionism) and dependency (neediness), the two aspects of depressiveness , might co occur with anxiety. High scores on self criticism are related to excessive concerns about being perfect. High scores on dependency are indicative of excessive worries about availability and approval of othe rs. Both of these factors imply some ongoing degree of anxiety related to oneâ€™s perception of self and approval and availability of others. Therefore, a significant positive relationship is inline with existing theories about these two concepts. Correlat ions among the Variables of the Study Final part of hypothesis testing involved examining correlations among the variables. With a few exceptions, the variables of the study were significantly correlated. These relationships ranged from low to moderate. Dependency was not significantly related to any of the following variables; attachment related avoidance, personal meaning
144 and self criticism. Given the nature of the two constructs, one might expect a negative relationship between attachment related avo idance and dependency. High scores on avoidance correspond to tendencies of avoiding intimacy in close relationships. On the other hand, high scores on dependency are related to excessive concerns for the approval and availability of others. The results did not confirm this expectation. These results might imply difference between constructs of intimacy and dependency. In other words, what avoidant persons try to stay away from might not be quite the same construct as what dependents seek. A significan t positive relationship between the anxiety scale of attachment and dependency subscale of depressiveness would be expected because both constructs are related to excessive concerns about availability of close relationships. Although such a relationship w as confirmed with the findings of this study, this relationship was weak ( r = .21). One possible reason for this weak relationship might be because of the differences between dependent and anxious personsâ€™ outlooks regarding relationships. Compared to an xious persons, dependent individuals might have more positive expectations about intimate relationships. In fact, anxious attachment style is associated with a preoccupation and ambivalence about intimate relationships. The dependency factor of depressive ness was not significantly correlated with personal meaning. It is possible that regardless of their interpersonal neediness, dependent persons might find satisfactory degrees of purpose and fulfillment in their lives. Although dependency needs are of gr eat concern to these persons, it is possible that even if these needs are not met for the time being, these persons might not give up on the hope to fulfill these needs in the futu r e. Therefore, a lack of fulfillment of these
145 needs at a given time may no t significantly impact their overall satisfaction with their lives. In addition, these needs whether fulfilled or not might provide a sense of meaning to these persons. Correlations found between the other variables ranged from low to modera te. These associations and their directions were essentially inline with the existing research. However, their degrees were somewhat lower than expected. For instance, dependency and attachment related anxiety had a Pearson correlation coefficient of .2 1. Similarly, while a higher negative correlation between dependency and attachment related anxiety would be expected, the results showed a correlation coefficient of only r = .13. No particular reasons can be thought of other than the characteristics of the sample and procedures of the study. For instance, there was no way of controlling or measuring social desirability with the exception of ensuring anonymity. Theoretical Implications There are several theoretical implications of the results. For inst ance, the findings partially highlight the importance of inte rpersonal relationships in functioning, adjustment, well being and vulnerability to pathology . The results illustrate the inseparable nature of domains of human functi oning; such as interpersonal intrapersonal; wellness pathology, and past present experiences. This study on the one hand supports the idea that attachment orientation has relevance to various areas of functioning. However, it shows that there is merit in the argument questioning the scope of attachment (i.e. M allinckrodt , 1995). In other words, attachment orientation might be related to a greater range of human functioning than the critics of attachment theory advocate.
146 On the other hand, this stud y did not nearly support the idea that attachment can function as a metaconstruct capable of integrating a variety of domains of development and functioning (Lopez, 1995). Some findings of this study did not endorse certain assumptions of attachment theory . For example, there was little evidence supporting the relationship between attachment and personal meaning. Similarly, attachment related anxiety, which was expected to have a strong positive relationship with dependency, only had weak relationship. T hese findings might be considered as indicative of the complex, non linear nature of attachment as a construct. Fraley (2003) advocates against categorization of attachment dimensions measured by the ECR R (Fraley et al., 2000). The results changed consid erably depending on whether attachment was used as a categorical or continuous variable. Furthermore, when it was continuous the analyses revealed far more relationships than those found when attachment was used as a categorical variable. This might be c onsidered as support for Fraleyâ€™s claim that categorization might limit the precision of the ECR R thus leading to decrease in statistical power. In fact, conceptualizing attachment as a continuous construct as opposed to categorical might be more useful from a clinical standpoint. Given the research findings documenting that oneâ€™s membership in attachment category is resistant to change over time, changing a personâ€™s attachment orientation can not possibly be the purpose of psychotherapy focusing on atta chment issues. On the other hand, such therapy can pursue change on the continuum of dimensions of attachment.
147 In other words, in stead of viewing attachment in restrictive categories, psychotherapy can more conveniently explore individualsâ€™ tendencies regarding anxiety and avoidance in close relationships. Although attachment researchers often make references to the relationships between attachment orientation and affect regulation, these relationships seems more complex. This complexity is evident in the unexpected finding regarding attachment related anxiety and trait anxiety. The results did not show any significant relationship between these two variables. However, given that persons with high scores on attachment related anxiety are known to u se hyperactive affect regulation strategies (Main, 1990), and to experience excessive anxiety and emotionality, a significant relationship would be expected. The results did not show such a relationship. These two results suggest that whi le attachment orientation might be related to one area of emotionality, it may not be related to another. Therefore, while making any generalizations about complex phenomena such as attachment, requires considerable caution. Further research is necessary to clarify the scope of the impact attachment orientation might have on various domains of functioning. This study confirmed the complexity of interrelations among major personality constructs. The findings illustrated that the presence, lack or varying degrees of such relationships might be mediated by a host of variables. For instance, as indicated by Lamb, Thompson, Gardner, Charnov, and Estes (1984), temperament is one of these factors. Hence, making straightforward and simple genera lizations about relationships between constructs such as attachment might be overly simplistic and, thus not theoretically sound. The mixed results of the study can be taken as support for the
148 intricacy of these interrelationships. Hence further research is needed to reach any degree of precision in the current understanding of the complex relationships among various variables of human functioning. Clinical Implications The findings of this study have several implications for clinical work with college s tudents. They offer partial evidence for the interconnected nature of various domains of human functioning. The findings suggest a flexible approach that takes attachment style into account in assessing, conceptualizing and intervening with issues brough t to counseling by college students. For example, individuals with high scores on avoidance might be more comfortable with a therapeutic approach that focuses on tasks rather than the therapeutic relationship. Conversely, individuals with high scores on attachment related anxiety benefit from a particular emphasis on the relational aspect of therapy. Therapists working with individuals with insecure attachment can explore adaptive aspects of this attachment. For example, a client could have a predominant ly anxious ambivalent attachment style and dependent tendencies. An exploration of how the person functions when dependency needs are not met can be useful. In stead of taking certain findings of attachment theory literally, clinicians should pay attenti on to unique adaptive strategies . Internal working models can also be explored in counseling with college students. This might be done in the context of the clientâ€™s meaning systems in order to attain and facilitate an understand ing of how the personâ€™s interpersonal functioning relates to his or her adaptation and general functioning.
149 Considering that only fearful attachment (high anxiety and avoidance) was related to personal meaning, psychotherapy might foster a sense of perso nal and interpersonal resourcefulness to individuals with this attachment style. On the other hand, given their high degrees of anxiety and avoidance these individuals might be in greater risk to prematurely terminate psychotherapy. Therefore, more flexi ble and person centered approaches might be more useful for these clients. Given the high mean score on personal meaning, lack of gender differences, and some similarity with adult samples from previous studies, the findings of this study imply that third and fourth year college students might have a great deal of resolution in transition to adulthood. Therefore, clinical work with these student s should be mindful of such resourcefulness and maturity. Limitations of the Study Several aspects of sampling, theoretical framework, and instrumentation of this study pose limitations to its internal and external validity. Since a sample of convenience was used the results might not be generalizable to the population. The sample of the study was obtained from pe rsonal growth classes. The individuals who take these classes might in some significant ways differ from those who do not. For example, students who take such classes are often from certain disciplines, such as education, sociology, recreational sciences and business, and they are in their 3 rd or 4 th years of college. A significant limitation of this study has to do with its exclusive use of self report measures. As pointed out by Rood (2000) using such measures, particularly with complex construct such as attachment, without desirability scales may risk internal validity of the study. For instance, desirability might be a factor in high scores on
150 personal meaning. Similarly, as pointed out by Cassidy and Kobak, (1988) relative to the preoccupied attach ment, the dismissing attachment involves down playing distress. Therefore, perhaps the former is more likely to be perceived and reported through self report measures, whereas the latter is underreported. For instance, Searle and Meara (1999) found that college studentsâ€™ emotional expressivity and emotional intentness (attention to emotion and their intensity) varied based on their attachment classification. They found the dismissing persons were least intent and least expressive, whereas preoccupied one s were most intent and most expressive. Hence, these differences might influence individualsâ€™ reports of their emotionality. Accordingly, particularly when only one assessment modality is utilized, this might have significant impact on research outcomes. As pointed out by Roberts et al. (1996) self report measures rest on the presumption that participants can accurately report such constructs. Attachment was not conceptualized merely as a construct of the conscious mind. Attachment researchers (Bowlby, 1980) have also linked internal working models to the unconscious. Hence, future studies should consider also incorporating measures such as the Adult Attachment Interview (AAI; George et al., 1985), which does not rely only on individual self report. In addition, the procedures used in administering the surveys may have posed problems to internal validity. Students who participated in the study received extra credit points from their instructors for their mere participation. Thus their motivation may n ot have been sincere. When lengthy self report measures are involved, perhaps immediate and content related incentives should be used to encourage more accurate and honest participation. Researchers could perhaps provide participants with extra copies an d make
151 the participation somewhat more experiential and personal. Studies could also involve inquiring feedback about individualsâ€™ subjective experiences with the surveys. This feedback might also include their ratings of the accuracy and sincerity of th eir responses. Furthermore, despite increased interest in attachment theory in recent decades, not all premises of the theory are agreed upon. For instance, the likely relationships between attachment dimensions and the examined variables (gender, persona l meaning, vulnerability to depression, and trait anxiety) could be mediated by other factors. Such factors might be explored if both quantitative and qualitative are incorporated in studies with such variables. Finally, the instrumentation might pose som e limitations. Despite the documented psychometric qualities of the instruments, they may pose some limitations. For instance, the Experiences in Close Relationships Revised (ECR R; Fraley et al., 2000) was developed by highly sophisticated statistical an d psychometric techniques. However, aside from the original study by the authors there has not been sufficient use and validation of the instrument. In addition, although this instrument was developed using the Item Response Theory (IRT: van der Linden & Hambleton, 1997) , the scoring was done by using classical test theory. That is to say , due to the complicated and time consuming scoring system of the IRT, for convenience the scores on this measure were calculated by summation. This scoring metho d might compromise some advantages of using an instrument created with the IRT. This might be even more so when converting the scores on the two dimensions of attachment.
152 In addition, the Life Regard Index Revised (LRI R; Debats, 1998) has well docume nted psychometric properties. Nevertheless, it has not been widely used and research done with the instrument appears to be conducted by the proponents of its conceptual bases and mostly done in Europe. Accordingly, the lack of diversity in its applicati on might be of concern. Implications for Future Research A number of suggestions can be made for future research. Some of these have to do with the findings of this study and s ome of them are related to the evolution of research in areas related to the variables of the study. A number of the findings of the study were contrary to those of prior research. These might in part be due to the use of global constructs such as personal meaning, depressiveness, and trait anxiety. Considering the complexity of such variables, future research should incorporate more specific variables into inquiries with such constructs. This incorporation could foster precision in research findings, which could improve the practical utility of such studies. For i nstance, along with investigating attachment orientation and depressiveness, studies could inquire if the interaction between depressiveness and the presence, absence or severity of past or present experience with clinical depression would contribute to di fferences in the relationship between depressiveness and attachment orientation. Future research should utilize college samples with diverse demographic characteristics. While investigating complex concepts such as attachment, comparing responses of stu dents in first years and those in the last years of their undergraduate studies might enrich current understanding of college student development. This also could be accomplished by utilizing longitudinal studies with this population.
153 Similarly, studies can compare other factors such as race, social class and sexual orientation as they relate to attachment and indicators of psychopathology and well being. Attachment theory could be extended to various clinical issues. It could be utilized in investigati ng various aspects of psychotherapy. For example, it might be used in inquiries involving willingness to seek professional psychological help, compliance with treatment (i.e. regular attendance), and clientsâ€™ preferences in counseling styles . Likewise, considering the connection between emotional intensity and preoccupied attachment, further research is needed to investigate if this style of attachment is also linked to specific DSM diagnoses such as bipolar disorder, attention deficit hyper activity disorders and cluster B diagnoses. Attachment research could also involve collaboration with areas such as cognitive psychology. It can also be combined with personality typologies. For instance, are persons with high scores on attachment relate d avoidance (dismissing/avoidant) more likely to have convergent styles of thinking and decision making (Hudson, 1967)? Are they more likely to have realistic typology? Are the anxious persons more likely to use divergent cognitive styles and be social o r artistic personality types (Holland, 1985)? Seeking empirical answers to such questions could stimulate further clarification in how attachment relates to such broad aspects of personality and functioning. Attachment research can also be greatly enriche d by incorporating findings in biological sciences. For example, there is not enough work done on possible contributions of temperament to development of attachment styles. Lamb et al. (1984) questioned that infantâ€™s behaviors in the Strange Situation (A insworth et al., 1978) might in fact be due to their temperament than attachment.
154 Therefore, future studies can address relationships between biological and genetic variables such as temperament to improve credibility of attachment theory. Moreover, futu re research might focus on issues that can foster preventive efforts in mental health service delivery. Among various ways of prevention, attending to chronic issues with low severity is important. When judging the clinical significance of an issue it is common that professionals focus on the immediate impairment and its severity in various aspects of oneâ€™s functioning. It appears that not much emphasis is put on issues that are lower in severity but highly long lasting. For example, the number of studi es dealing with major depression is incomparably higher than that of studies with dysthymia. It is so for various diagnostic classifications. Therefore, considering long term implications of issues such as depressiveness and trait anxiety, which are not severe enough to receive clinical attention, greater scientific attention is needed. Such attention is vital in guiding preventive efforts. No studies focusing on adaptive outcomes of insecure attachment were found in the literature. Given attachment th eoryâ€™s claim that individuals form attachment styles early in life, it is fair to assume that there would be certain unique advantages to having any particular style of attachment. S ince not all insecure individuals have severe psychopatholo gies, qualitative research methodologies could be utilized to examine adaptive advantages of each attachment orientation.
155 APPENDIX A INFORMED CONSENT Ibrahim Keklik Doctoral Candidate University of Florida College of Education Department of Counselor Education Home address: 389 Maguire Village #1 Gainesville, FL, 32603 Phone: (352) 846 7501 INFORMED CONSENT FORM Dear fellow student, This is a doctoral dissertation research, which intends to investigate relationships of college studentsâ€™ (ages 18 23) relational styles, gender, and measures of their emotional experiences and well being. Please read this letter carefully before deciding to participate. The confidentiality of your responses is of paramount importance to the researcher. Thus every necessary step will be taken to ensure the privacy of the information you provide. You do not need to write your names on the answer sheets. Therefore, your identity will remain anonymous. Answering the questionnaires will not take more than 60 minutes. Responding to this questionnaire does not have any immediate benefits to you, aside from receiving extra credit point from your inst ructor. The questionnaires also donâ€™t include any questions that could cause any distress to you. Hence, it is entirely your choice whether to partake in this study or not. You do not have any obligation to answer any question you do not wish to answer.
156 You also have the choice of requesting your responses not be used in the analyses at any time before the study is c oncluded. If you have any questions about your rights as research participants please contact University of Florida Institutional Review Boa rd at: UFIRB office, Box 112250, University of Florida, Gainesville, FL 32611 2250 ; phone: 392 0433. If you have any questions about this study please contact Ibrahim Keklik at firstname.lastname@example.org or Dr. James Archer, J r. at email@example.com . Thank you for your contribution to this study. Sincerely Ibrahim Keklik If you decide not to participate, you are requested to leave the classroom. By signing the sign in sheet you will be consenting to partaking in this study.
157 APPENDIX B THE EXPERIENCES IN CLOSE RELATIONSHIPS QUESTIONNAIRE REVISED The following statements concern how you feel in romantic relationships. We are interested in how you generally experience relationships, not just in what is happening in a current r elationship. Respond to each statement by indicating how much you agree or disagree with it. Write the number in the space provided, using the following rating scale: Sample Items: 1 2 3 4 5 6 7 Disagree Strongly ......... ......... Neutral/ Mixed ......... ......... Agree Strongly ___1. I'm afraid that I will lose my partner's love. ___2. I often worry that my partner will not want to stay with me. ___3. I often worry that my partner doesn't really love me. ___4. I worry that romanti c partners wonâ€™t care about me as much as I care about them. ___5. I often wish that my partner's feelings for me were as strong as my feelings for him or her. ___6. I worry a lot about my relationships. ___7. When my partner is out of sight , I worry that he or she might become interested in someone else. ___8. When I show my feelings for romantic partners, I'm afraid they will not feel the same about me.
158 1 2 3 4 5 6 7 Disagree Strongly ......... ......... Neutral/ Mixed ......... ......... Agree Strongly ___10. My romantic partner makes me doubt myself. ___11. I do not often worry about being abandoned. ___12. I find that my partner(s) don't want to get as close as I would like. ___13. Sometimes romantic partners change their feelings about me for no apparent reason. ___14. My desire to be very close sometimes scares people away. ___15. I'm afraid that once a romantic part ner gets to know me, he or she won't like who I really am. ___16. It makes me mad that I don't get the affection and support I need from my partner. ___17. I worry that I won't measure up to other people. ___18. My partner only seems to notice me when Iâ€™m angry. ___19. I prefer not to show a partner how I feel deep down. ___20. I feel comfortable sharing my private thoughts and feelings with my partner. ___21. I find it difficult to allow myself to depend on romantic partners. ___22. I am very comfortable being close to romantic partners. ___23. I don't feel comfortable opening up to romantic partners. ___24. I prefer not to be too close to romantic partners. ___25. I get uncomfortable when a romantic partner wants to be very close. ___26. I fi nd it relatively easy to get close to my partner. ___27. It's not difficult for me to get close to my partner. ___28. I usually discuss my problems and concerns with my partner. ___29. It helps to turn to my romantic partner in times of need.
159 1 2 3 4 5 6 7 Disagree Strongly ......... ......... Neutral/ Mixed ......... ......... Agree Strongly ___30. I tell my partner just about everything. ___31. I talk things over with my partner. ___32. I am nervous when partners get too close to me. ___33. I feel comfortable depending on romantic partners. ___34. I find it easy to depend on romantic partners. ___35. It's easy for me to be affectionate with my partner. ___36. My partner really understands me and my needs
160 APPENDIX C LIFE REGARD INDEX REVISED Below are 28 statements regarding your attitudes towards some aspects of life, with which you might agree or disagree. Please indicate each time the answer that best represents your opinion STATEMENTS Sample Items: 1. I feel like I have found a really significant meaning for leading my life. __1__ Do not Agree __ 2 __No Opinion __3__ Agree 2. Living is deeply fulfilling. __1__ Do not Agree __ 2 __No Opinion __3__ Agree 3. I really donâ€™t have much of a purpose for living, even for myself. __1__ Do not Agree _ 2 __No Opinion __3__ Agree 4. There honestly isnâ€™t anything that I totally want to do. __1__ Do not Agree __ 2 __No Opinion __3__ Agree 5. I really feel good about my life. __1__ Do not Agree __ 2 __No Opinion __3__ Agree 6. I spend most of time doing things that really arenâ€™t very important to me. __1__ Do not Agree __ 2 __No Opinion __3__ Agree 7. I have really come to terms with whatâ€™s important for me in life. __1__ Do not Agree __ 2 __No Opinion __3__ Agreee
161 REFERENCES Adam, K. S., Sheldon Keller, A. E., & West, M. (1996). Attachment organization, history of suicidal behavior in clinical adolescents. Journal of Consulting and Clinical Psychology, 64, 264 272. Ainsworth, M. D. S. (1964). Patterns of attachment behavior shown by the infant in interaction with his mother. Merrill Palmer Quarterly, 10, 51 58. Ainsworth, M. D. S. (1967). Infancy in Uganda: Infant care and the of attachment . Baltimore: Johns Hopkins University Press. Ainsworth , M. D. S. (1989). Attachment beyond infancy. American Psychologist, 44, 709 716. Ainsworth, M. D. S. (1990). Epilogue: Some considerations regarding theory and assessment relevant to attachments beyond infancy. In M. T. Greenberg, D., Cicc hetti, & E. M. Cummings (Eds.), Attachment in the preschool years (pp. 463 488). Chicago: University of Chicago Press. Ainsworth, M. D. S., Blehar, M. C., Waters, & Walls, S. (1978). Patterns of attachment: A psychological study of the strange s ituation . Hillsdale, NJ: Lawrence Erlbaum. Allen, J. P., Hauser, S. T., & Borman Spurrell, E. (1996). Attachment theory as a framework for understanding sequelae of severe adolescent psychopathology: An 11 year follow up study. Journal of Cons ulting and Clinical Psychology, 64 , 254 263. Antonovsky, A. (1979). Health, stress, and coping: New perspectives on mental and physical well being . San Francisco: Jossey Bass. Antonovsky, A. (1987). Unraveling the mystery of health . San Fr ancisco: Jossey Bass. Archer, J., Jr., & Cooper, S. (1998). Counseling and mental health services on campus: A handbook of contemporary practices and challenges. San Francisco: Jossey Bass.
162 Armsden, G. C., & Greenberg, M. T. (1987). The inventory of parent and peer attachment: Individual differences and their relations to psychological well being in adolescence. Journal of Youth and Adolescence, 16, 427 453. Baldwin, M. W., & Fehr, B. (1995). On the instability of attachment style ratings. Personal Relationships, 2 , 247 261. Bartholomew, K., & Horowitz, L. M. (1991). Attachment styles among young adults: A test of a four category model. Journal of Personality and Social Psyc hology, 61 , 226 244. Bartholomew, K., & Thompson, J. (1995). The application of attachment theory to counseling psychology. Counseling Psychologist, 23 , 484 490 Battista, J., & Almond, R. (1973). The development of meaning in life. Psychiatry, 36 , 409 427. Beck, A. T. (1963). Thinking and depression: Idiosyncratic content and cognitive disorders. Archives of General Psychiatry, 9, 324 333. Beck, A. T. (1967). Depression: Clinical, experimental, and theoretical aspects . New York: Harper and Row. Beck, A. T., Epstein, N., Harrison, R. P., & Emery, G. (1983). Cognitions, attitudes, and personality dimensions in depression. British Journal of Cognitive Psychotherapy, 1, 1 16. Beck, A.T., Ward, C. H., Mendelson, M., Mock, J., & Erbau gh, J. (1961) An inventory for measuring depression. Archives of General Psychiatry 4 , 561 571. Belenky, M. F., Clinchy, B. M., Goldberger, N. R., & Tarule, J. M. (1986). Women's ways of knowing. New York: Basic Books. Benson, M. J., Harris, P. B., & Rogers, C. S. (1991). Identity consequences of attachment to mothers and father among the late adolescents. Journal of Research on Adolescence, 2 , 187 204. Bertalanffy, Ludwig Von. (1968). General systems t heory: Foundations, development, and applications . New York: Braziller. Bibring, E. (1953). The mechanisms of depression. In Greenacre, P. (Ed.), Affective Disorders. (pp. 13 48). New York: International Universities Press. Bishop, J. B., & Trembley, E. L. (1987). Counseling c enters and accountability: Immovable objects, irresistible factors. Journal of Counseling and Development, 65 , 491 494.
163 Blaney, P. H., & Kutcher, G. S. (1991). Measures of depressive dimensions: Are they interchangeable? Journal of Personality As sessment, 56 , 502 512. Blatt, S. J. (1974). Levels of object representation in analytic and introjective depression. Psychoanalytic Study of the Child, 29, 107 157. Blatt, S. J. , Dâ€™Affliti, J., & Quinlan, D. (1976). Experiences of depression i n normal young adults. Journal of Abnormal Psychology, 85, 383 389. Blatt, S. J., Quinlan, M. D., Chevron, E. S., McDonald, C., & Zuroff, D. (1982). Dependency and Self Criticism: psychological dimensions of depression. Journal of Consulting an d Clinical Psychology, 50, 113 124. Blatt, S. J., & Blass, R. B. (1996). Relatedness and self definition: A dialectical model of personality development. In G. G. Noam & K. W. Fischer (Eds.), Development and vulnerability in close relationship s . (pp. 309 338). New Jersey: Lawrence Erlbaum Associates. Blustein, D. L., Prezioso, M. S., & Schulthesis, D. P. (1995). Attachment theory and career development: Current status and future directions. The Counseling Psychologist, 23 , 416 432. Bowen, M. (1978). Family therapy and clinical practice . New York: Jason/Aranson, Inc. Bowlby, J. (1960). Separation anxiety. International Journal of Psycho Analysis, 39 , 350 373. Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachmen t. New York: Basic Books. Bowlby, J. (1973). Attachment and loss: Vol. 2. Separation, anxiety, and anger: New York: Basic Books. Bowlby, J. (1980). Attachment and loss: Vol. 3. Loss, sadness, and depression: New York: Basic Books. Bowlby, J. (1982). Attachment and l oss: Vol. 1. Attachment (2 nd ed.). New York: Basic Books. Bowlby, J. (1988). A secure base: Parent child attachment and healthy human development. New York: Basic Books. Brennan, K. A, Clark, C. L., & Shaver, P. (1998). Self report measur es of adult attachment. In J. A. Simpson, & W. S. Rholes (Eds.), Attachment theory and close relationships . (pp. 46 76). New York: The Guilford Press.
164 Brennan, K. A., & Shaver, P. (1995). Dimensions of adult attachment, affect regulation and romantic relationships functioning. Personality and Social Psychology Bulletin, 21 , 267 283. Brennan , K. A.; Shaver , P., & Tobey , A. E. (1991). Attachment styles, gender, and parental problem drinking. Journal of Social and Personal Relationshi ps, 8 , 451 466. Carnelley, K. B., Pietromonaco, P. R., & Jaffe, K. (1994). Depression, working models of others, and relationship functioning. Journal of Personality and Social Psychology, 66 , 127 140. Cassidy, J. (1999). The nature of the chil dâ€™s ties. In J. Cassidy & P. Shaver (Eds.), Handbook of attachment: Theory, research and clinical applications . (pp. 3 20). New York: The Guilford Press. Cassidy, J., & Kobak, R. (1988). Avoidance and its relation to other defensive processes. In J. Belsky & T. Neworski (Eds.), Clinical implications of attachment (pp. 300 323). NJ: Erlbaum. Cattell, R. B., Scheier, I. H. (1963). Handbook for IPAT Anxiety Scale (2 nd ed.). Champaign, IL: Institute for Personality and Ability Testin g. Chickering, A. W.,& Reisser, L. (1993). Education and identity . (2 nd ed.). Jossey Bass Publishers. San Francisco. Cole Detke, H., & Kobak, R. (1996). Attachment processes in eating disorders and depression. Journal of Consulting and Clinica l Psychology, 64, 282 290. Cole, P. M., & Putman, F. W. (1992). Effect of incest on self and social functioning: A developmental psychopathology perspective. Journal of Consulting and Clinical Psychology, 60, 174 184. Collins, N. L., & Read, S. J. (1990). Adult attachment, working models, and relationship quality in dating couples. Journal of Personality and Social Psychology, 58 , 644 663. Cowan, P. A., Cohn, D., Cowan, C. P., & Pearson, J. L. (1996). Parents' attachment histories an d children's internalizing and externalizing behavior: Exploring family systems models of linkage. Journal of Consulting and Clinical Psychology , 64, 53 63. Creasey, G., Hesson McInnis, M. (2001). Affective responses, cognitive appraisals, and confli ct tactics in late adolescent romantic relationships: Associations with attachment orientations. Journal of Counseling Psychology, 48, 85 96.
165 Creasey, G., Kershaw, K., & Boston, A. (1999). Conflict management with friends and romantic partners: T he role of attachment and negative mood regulation expectancies. Journal of Youth and Adolescence, 28, 523 543. Crittenden, P. M. (1988). Relationships at risk. In J. Belsky, & T. Nezworsky (Eds.), Clinical implications of attachment . (pp. 1 36 174). Hillsdale, NJ: Erlbaum. Crittenden, P. M. (1990). Internal representational models of attachment relationships. Infant Mental Health Journal, 11 , 259 277. Cummings, E. M., & Cicchetti, D. (1990). Toward a transactional model of rel ations between attachment and depression. In M. T. Greenberg., D. Chicchetti & E. M. Cummings (Eds.). Attachment in the preschool years: Theory, research and N. intervention (pp. 339 372). Chicago: University of Chicago Press. Cummings, E. M., & D avies, P. (1996). Emotional security as a regulatory process in normal development and the development of psychopathology. Development and Psychopathology, 8 , 123 139. Davis, M. H., Morris, M. M., Kraus, L. A. (1998). Relationship specific and global perceptions of social support: Associations with well being and attachment. Journal of Personality and Social Psychology, 74 , 468 481. Del Carmen, R., & Huffman, L. (1996). Epilogue: Bridging the gap between research on attachment and psych opathology. Journal of Consulting and Clinical Psychology, 6 , 291 294. Delvey, J., Jr. (1985). Beyond the black screen: The patientâ€™s search for an emotional container in the therapist. Psychotherapy, 22, 583 586. Debats, D. L. (1990). The Life Regard Index: Reliability and validity. Psychological Reports, 67 , 27 34. Debats, D. L. (1998). Measurement of personal meaning: The psychometric properties of the Life Regard Index. In P. T. P Wong & P. S. Fry (Eds.), The human quest f or meaning: A handbook of psychological research and clinical applications. (pp. 237 261). Mahwah, NJ: Lawrence Erlbaum Associates, Publishers. Dozier, M. (1990). Attachment organization and treatment use for adults with serious psychopathologica l disorders. Development and Psychopathology, 2, 47 60. Dozier, M., Stevenson, A, L., Lee, S. W., & Velligan, D. I. (1992). Attachment organization and familial overinvolvement for adults with serious psychopathological disorders. Development and Psychopathology, 3, 475 489.
166 Edwards, A. L. (1959). Edwards Personal Preference Schedule . New York: Psychological Corporation. Engels, R. C. M. E., Finkenauer, C., Meeus, W., & Dekovic, M. (2001). Parental attachment and adolescentsâ€™ emotio nal adjustment: The associations with social and relational competence. Journal of Counseling Psychology, 48 , 428 439. Enns, M. W., Cox, B. J., & Larsen, D. K. (2000). Perceptions of parental bonding and symptom severity in adults with depression : Mediation by personality dimensions. The Canadian Journal of Psychiatry, 45, 263 268. Erikson, E. (1968). Identity: Youth and crisis. New York: W. W. Norton. Evans, L., & Wertheim, E. H. (1996). Intimacy patterns and relationship satisfactio n of women with eating problems and the mediating effects of depression, trait anxiety and social anxiety. Journal of Psychosomatic Research, 44, 355 365. Fagony, P., Leigh, T., Steele, M., Steele, H., Kennedy, R., & Matton, G. (1996). The relation of attachment status, psychiatric classification, and response to psychotherapy. Journal of Consulting and Clinical Psychology, 64 , 22 31. Fairbairn, W. R. D. (1952). An object relations theory of the personality. New York: Basic Books. Feeney, J. A. (2000). Implications of attachment, style for patterns of health and illness. Child: Care, Health and Development, 26 , 277 288. Feeney, J. A., & Noller, P. (1990). Attachment styles as predictors of adult romantic relationships. Journal o f Personality and Social Psychology, 8, 181 215. Feeney, J. A., & Noller, P., & Hanrahan, M. (1994). Assessing adult attachment. In M. Sperling & W. H. Berman (Eds.), Attachment in adults: Clinical and developmental perspectives (pp. 128 152). New York: Guilford Press. Flanagan, C. (1999). Early socialization: Sociability and attachment. London: Routledge Taylor & Francis Group. Fraley, C. R. (2003). Information on the Experiences in Close Relationships Revised (ECR R): Adult Atta chment Questionnaire . [On line]. Available: http://tigger.uic.edu/%7Efraley/measures/ecrr.htm Fraley, R. C., Waller, N. G., & Brennan, K. A. (2000). An item response theory analysis of self report measures of adult attachment. Journal of Personal ity and Social Psychology, 78 , 350 365. Frankl, V. (1963). Manâ€™s search for meaning . Boston: Beacon.
167 Friedman, S. (1997). Time effective psychotherapy: Maximizing outcomes in an era of limited resources . Boston: Alley and Bacon. Fuhr, S. K, & Shean, G. (1992). Subtypes of depression, efficacy, and the Depressive Experiences Questionnaire. The Journal of Psychology, 126 , 495 506. Fulkerson, C. C. (1986). Psychosomatic illnesses. In J. E. Talley, & W. J. K. Rockwell (Eds.), Counse ling and psychotherapy with college students: A guide to treatment (pp. 123 141). New York: Praeger Publishers. George, C., Kaplan, N., & Main, M. (1985). Attachment interview for adults. Unpublished manuscript . University of California, Berke ley. Gilligan, C. (1982). In a different voice: Psychological theory and womenâ€™s development . Cambridge, MA: Harvard University Press. Goldberg, S., Muir, R., & Kerr, J. (1995). Attachment theory: Social, developmental, and clinical perspectives. H illsdale, NJ: The Atlantic Press. Griffin, D. W., & Bartholomew, K. (1994). The metaphysics of measurement: The case of adult attachment. In K. Bartholomew & D. Perlman (eds.), Advances in personal relationships: Vol. 5. Attachment processes in ad ulthood (pp. 17 52). London: Jessica Kingsley. Group for Advancement of Psychiatry. (1990). Psychotherapy with college students: Formulated by The Committee on the College Student (Report No. 130). New York: Brunner/Mazel, Inc. Harlow, H. F., & Zimmerman, R. (1959). Affectional responses in the infant monkey. Science, 130 , 421 432. Harris, A. H. S., Standard, S. (2001). Psychometric properties of the Life Regard Index Revised; A validation study of a measure of personal meaning. Psycho logical Reports, 89 , 759 772. Harwood, R. L., Miller, J. G., & Irizarry, N. L. (1995). Attachment and culture: perceptions of the child in context. New York: The Guilford Press. Hazan, C., & Hutt, M. (1993). Patterns of adaptations: Attachment differences in psychosocial functioning during the first year of college . Unpublished manuscript. Cornell University, Ithaca, NY. Hazan, C. & Shaver, P. (1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52, 511 524.
168 Hazan, C. & Shaver, P. (1990). Love and work: An attachment theoretical perspective. Journal of Personality and Social Psychology, 59 , 270 280. Hishinuma, E. S., Miyamoto, R. H., Nishimura, S. T., Goebert, D. A. , Yuen, N. Y. C., Makini, Jr. G. K., Andrade, N. N., Johnson, R. C., & Carlton, B. S. (2001). Prediction of anxiety disorders using the State Trait Anxiety Inventory for multiethnic adolescents. Journal of Anxiety Disorders, 15 , 511 533. Hollan d, J. L. (1985). Making vocational choices: A theory of careers (2nd Ed). Englewood Cliffs, NJ: Prentice Hall. Hudson, L. (1967). Contrary imaginations: A psychological study of the English schoolboy . Harmondsworth: Penguin. Jacobson, E. (197 1). Depression: Comparative studies of normal, neurotic, and psychotic conditions. New York: York Hoeber. Kennedy, B. L., Schwab, J. J., Morris, R. L., & Beldia, G. (2001). Assessment of state and trait anxiety in subjects with anxiety and depre ssive disorders. Psychiatric Quarterly, 72, 263 276. Kenny, M. E. (1987). The extent and function of parental attachment among first year college students. Journal of Youth and Adolescent, 16, 17 27. Kenny, M. E., Donaldson, G. (1991). Cont ributions of parental attachment and family structure to the social and psychological functioning of first year college students. Journal of Counseling Psychology, 38, 479 486. Kenny, M. E., Lomax, R. G., Brabeck, M., & Fife, J. (1998). Longitudinal pa thways linking reports of maternal and paternal attachments to psychological well being. Journal of Early Adolescence, 18 , 221 243. Kenny, M. E., Perez, V. (1996). Attachment and psychological well being among racially and ethnically diverse first year college students. Journal of College Student Development, 37 , 527 535). Kenny, M. E., & Rice, K. G. (1995). Attachment to parents and adjustment in late adolescent college students: Current status, applications, and future considerations. The Counseling Psychologist, 23 , 433 455. Kemp. M. A., Neimeyer, G. J. (1999). Interpersonal attachment: Experiencing, expressing and coping with stress. Journal of Counseling Psychology, 46, 388 394.
169 Kirkpatrick, L. A., & Davis, K. E. (19 94). Attachment style, gender, and relationships stability: A longitudinal analysis. Journal of Personality and Social Psychology, 66, 502 512. Kirkpatrick, L. A., Hazan, C. (1994). Attachment styles and close relationships: A four year prospec tive study. Personal Relationships, 1 , 123 142. Kirkpatrick, L., Shaver, P. (1992). An attachment theoretical approach to romantic love and religious belief. Personality and Social Psychology Bulletin, 18 , 266 275. Klein , M. (1932). The p sycho a nalysis of children . , London: Hogarth. Klinger, E. (1998). The search for meaning in evolutionary perspective and its clinical implications. In P. T. P Wong & P. S. Fry (Eds.), The human quest for meaning: A handbook of psychological research a nd clinical applications. (pp. 27 51). Mahwah, NJ: Lawrence Erlbaum Associates, Publishers Kobak, R. R., Cole, H. E., Ferenz Gillies, R., Fleming, W. S., & Gamble, W. (1993). Attachment and emotion regulation during mother teen problem solving: A control theory analysis. Child Development, 64, 231 245. Kobak, R. G., & Hazan, C. (1991). Attachment in marriage: Effects of security and accuracy or working models. Journal of Personality and Social Psychology, 60 , 861 869. Kobak, R. R., & Sceery, A. (1988). Attachment in adolescence: Working models, affect regulation, and representations of self and others. Child Development, 59, 135 146. Kobak, R. G., Sudler, N., & Gamble, W. (1991). Attachment and depressive symptoms during adolescence: A developmental pathway analysis. Developmental and Psychopathology, 3, 461 474. Kotler, T., & Omodei, M. (1988). Attachment and emotional health: A life span approach . Human Relations, 41 , 619 640. La Guardia, J. G., Ryan, R. M., Couchman, C. H., & Deci, E. D. (2000). Within person variation in security of attachment: A self determination theory perspective on attachment, need fulfillment, and well being. Journal of Personality and Social Psychology, 79 , 367 385. Lamb , M. E., Thompson, R. A., Gardner, W. P., Charnov, E. L., & Estes, D. (1984). Security of infantile attachment as assessed in the strange situation: Its study and biological interpretation. Brain and Behavioral Sciences, 7, 127 147.
170 Lapsley, D. K., & Edgerton. J. (2002). Separation individuation, adult attachment style, and college adjustment. Journal of Counseling and Development, 80, 484 492. Lapsley, D. K., Rice, K., & Fitzgerald, D. P. (1990). Adolescent attachment, identity, and adjustment to college: Implications for the continuity of adaptation hypothesis. Journal of Counseling and Development, 69, 561 565. Lieberman, A. F., & Pawl, J. H. (1988). Clinical applications of attachment theory. In J. Belsky & T. Nezworsk i (Eds.), Clinical Applications of Attachment . (pp. 327 351). Hillsdale: Lawrence Erlbaum Associates. Liotti G. (1999). Disorganization of attachment as a model for understanding dissociative psychopathology. In: J. Solomon & C. George (Eds.), Att achment disorganization . (pp 291 â€“ 317). New York: Guilford Press. Lopez, G. F. (1995). Contemporary attachment theory: An introduction with implications for counseling psychology. The Counseling Psychologist, 23, 395 415. Lopez, F. G. (2001). A dult attachment, self other boundary regulation, and splitting tendencies in a college sample. Journal of Counseling Psychology, 48 , 440 446. Lopez, F. G., & Gormley, B. (2002). Stability and change in adult attachment style over the first year col lege transition: Relations to self confidence, coping, and distress patterns. Journal of Counseling Psychology, 49 , 355 364. Lopez, G. F., Melendez, M. C., & Rice. K. G. (2000). Parental divorce, parent child bonds, and adult attachment orientati ons among college students: A comparison of three ethnic groups. Journal of Counseling Psychology, 47 , 177 186. Lopez, G. F., Melendez, M. C., Sauer, E. M., Berger, E., & Wyssmann, J. (1998). Internal working models, self reported problems, and he lp seeking attitudes among college students. Journal of Counseling Psychology, 45 , 79 83. Lorenz, K. Z. (1952). King Solomon's ring . New York: Crowell Lyddon, W. J. (1995). Attachment theory: A metaperspective for counseling psychology? Journal of Counseling Psychology, 48, 401 409. Magai, C., Hunziker, J., Mesias, W., & Culver, L. C. (2000). Adult attachment styles and emotional biases. International Journal of Behavioral Development, 24 (3), 301 309.
171 Main, M. (1990). Cross cultural St udies of attachment organization: Recent studies, changing methodologies and the concept of conditional strategies. Human Development, 55 , 123 136. Main, M. (1991). Metacognitive knowledge, metacognitive monitoring, and singular (coherent) vs. mul tiple (incoherent) model of attachment: Findings and directions for future research. In C. M. Parkes, J. Stevenson Hinde, & P. Marris (Eds.), Attachment Across the Life Cycle . (pp. 127 159). London: Routledge. Main, M. (1995). Recent studies in at tachment: Overview with selected implications for clinical social work. In S. Goldberg., R. Muir & J. Kerr (Eds.), Attachment theory: Social, developmental, and clinical perspectives. (pp. 407 474). Hillsdale, NJ: The Analytic Press. Main, M. (19 96). Introduction to the special section on attachment and psychopathology: 2. Overview of the field of attachment. Journal of Consulting and Clinical Psychology, 64 , 237 243. Main, M., & Solomon, J. (1986). Discovery of a disorganized/disoriented at tachment pattern. In M. T. Greenberg, D. Cicchetti, & E. M. Cummings (Eds.), Attachment in the preschool years (pp. 121 160). Chicago: University of Chicago Press. Mallinckrodt, B. (1995). Attachment theory and counseling psychology: Ready to be a prime time player? Journal of Counseling Psychologist, 23, 501 505. Mantynen, P., Happonen, J., & Toskala, A. (2000). Social phobia: Control processes in anxiety provoking situations. Journal of Constructivist Psychology, 13 (3), 199 220. Maslow, A. (1968). Toward a psychology of being . (Rev. ed.). New York: Van Nostrand Reinhold. Meyers, S. A. (1998). Personality correlates of adult attachment style. The Journal of Social Psychology, 138, 407 409. Mikulincer, M., & Flori an, V. (1995). Appraisal and coping with a real life stressful situation: The contribution of attachment styles. Personality and Social Psychology Bulletin, 21 , 408 416. Mikulincer, M., & Florian, V. (1998). The relationship between adult attachment style and emotional and cognitive reactions to stressful events. In Simpson, J. A., Rhodes, W. S. (Eds.), Attachment theory and close relationships (pp. 143 165). New York: Guilford.
172 Mikulincer, M., & Florian, V. (1999). The association be tween parental reports of attachment style and family dynamics, and offspringâ€™s reports of adult attachment style. Family Process, 38 , 243 255. Mikulincer, M., Horesh, N., Eilati, H., & Kotler, M. (1999). The association between adult attachment st yle and mental health in extreme life endangering conditions. Personality and Individual Differences, 27, 831 842. Mitchell, S. T. (2000). Relationality: From attachment to intersubjectivity . Hillsdale, NJ: The Analytic Press. Mohr, J. J. (1999 ). Same sex romantic attachment. In J. Cassidy & P. Shaver (Eds.), Handbook of attachment: Theory, research and clinical applications . (pp. 378 394). New York: The Guilford Press. Mongrain, M., & Zuroff, D. C. (1989). Cognitive vulnerability to depressed affect in dependent and self critical college women. Journal of Personality Disorders, 3 , 240 251. Morelli, G. A., & Tronick, E. Z. (1991). Efe Multiple caretaking and attachment. In J. L. Geweritz & W. L. Kurtines. (Eds.), In tersections with attachment . (pp. 41 51). Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers. Mosher, D. L. (1966). The development and multitrait multimethod matrix analysis of three measures of three aspects of guilt. Journal of Consulti ng Psychology, 30 , 25 29. Mothersead, P. K., Kivlighan, D. M. Jr., & Wynkoop, T. F. (1998). Attachment, family dysfunction, parental alcoholism, and interpersonal distress in late adolescence: A structural model. Journal of Counseling Psychology, 45 , 196 203. Muris, P., Mayer, B., & Meesters, C. (2000). Self reported attachment style, anxiety, and depression in children. Social Behavior and Personality, 28, 157 162. Nickels, J. B., & Stewart, M. E. (2000). The Relationship between life meani ng and commitment to and consistency in life values.. Paper presented at 1 st International Conference on Personal Meaning. Vancouver, Canada. Noam, G. G., & Fischer, K. W. (1996). Development and vulnerability in close relationships . New Jersey: L awrence Erlbaum Associates. Oâ€™Brien, E., & Epstein, S. (1974). Self Report Inventory . Unpublished test. University of Massachusetts, Amherst.
173 Petersen, A. C., Kennedy, R. E., & Sullivan, P. (1991). Coping with adolescents. In M E. Colten & S. Gore (Eds.), Adolescent stress: Causes and consequences. (pp. 91 110). Hawthorne, NY: Aldine De Gruyter. Piaget, J. (1950). The psychology of intelligence . New York: Harcourt Brace. Piaget, J. (1972). Intellectual evolution from adolescenc e to adulthood. Human Development, 15, 1 12. Pianta, R. C., Egeland, B., & Adam, E. K. (1996). Adult attachment classification and self reported psychiatric symptomatology as assessed by the Minnesota Multiphasic Personality Inventory 2. Journa l of Consulting and Clinical Psychology, 64 , 273 281. Pistole, M. C. (1989a). Attachment in adult romantic relationships: Style of conflict resolution and relationship satisfaction. Journal of Social and Personal Relationships, 6, 505 510. Pistole, M. C. (1989b). Attachment: Implications for counselors. Journal of Counseling and Development, 86, 190 193. Pistole, M. C. (1999). Caregiver in attachment relationships: A perspective for counselors. Journal of Counseling and Development, 77, 43 7 446. Raja, S. N., McGee, R., & Stanton, W. R. (1992). Perceived attachment to parents and peers and psychological well being in adolescence. Journal of Youth and Adolescence, 21 , 471 485. Ramanaiah, N. V., Franzen, M., & Schill, T. (1983). A psychometric study of the State Trait Anxiety Inventory. Journal of Personality Assessment, 47 , 531 535. Reker, G. T. (1992). Manual of the Life Attitude Profile Revised (LAP R) . Trent University, Peterborough, ON: Student Psychologists Press. Rice, K. G., & Cummings, P. N. (1996). Late adolescent and parent perceptions of attachment: An exploratory study of personal and social well being. Journal of Counseling and Development, 75, 50 75. Rice, K. G., FitzGerald, D. P., Whaley, T. J. , & Gibbs, C. L. (1995). Cross sectional and longitudinal examination of attachment, separation individuation, and college student adjustment. Journal of Counseling and Development, 73 , 463 474. Rice, K. G., & Mirzadeh, S. A. (2000). Perfection ism, attachment, and adjustment. Journal of Counseling Psychology, 47 , 238 250.
174 Rice, K. G., & Whaley, T. J. (1994). A short term longitudinal study of within semester stability and change in attachment and college student adjustment. Journal o f College Student Development Sep 1994, 35 , 324 330. Roberts, J. E., Gotlib, I. H., Kassel, J. D. (1996). Adult attachment security and symptoms of depression: The mediating roles of dysfunctional attitudes and low self esteem. Journal of Persona lity and Social Psychology, 70 , 310 320. Robertson, J., & Bowlby, J. (1952). Responses of young children to separation from their mother. Courier of the International Childrenâ€™s Center, Paris, 2 , 131 140. Rokeach, M. (1973). The nature of huma n values . New York: Free Press. Rood, L. (2000). The relationship of persona coherence and interpersonal attachment with psychological well being . Doctoral dissertation, University of Florida, Gainesville, Fl. Rosenberg, M. (1965). Society and th e adolescent self image. Princeton: Princeton University Press Rosenblum, L A., & Paully, G. S. (1991). Evolutionary and environmental factors influencing attachment patterns in nonhuman primates. In J. L. Geweritz & W. L. Kurtines (Eds.), Inters ections with attachment . (pp. 23 39). Hillsdale, NJ: M. Lawrence Erlbaum Associates, Publishers. Rosenstein, D. S., & Horowitz, H. A. (1996). Adolescent attachment and psychopathology. Journal of Consulting and Clinical Psychology, 64 (2), 244 2 53. Rothbard, J. C., Roberts, L. J., Leonard, K. E., & Eiden, R. D. (1993, August). Attachment styles and marital interaction behavior . Paper presented at annual meeting of the American Psychological Association, Toronto, Ontario. Rothbard J. C., & Shaver, P. (1994). Continuity of attachment across the life span. In M. B. Sperling & W. H. Berman (Eds.), Attachment in adults. Clinical and developmental perspectives . (pp. 31 71). New York: Guilford. Rotter, J. B. (1966) I nternal E xternal Locus of Control Scale (1966). In B. Blattner (1981). Holistic nursing. (pp. 437 438). Englewood Cliffs: Prentice Hall. Sampson, E. (1988). Indigenous psychologies of the individual and their role in personal and societal functioning. Amer ican Psychologist, 43 , 15 22. Sandell, R., Blomberg, J., & Lazaar, A. (1998). The factor structure of Anthonovskyâ€™s sense of coherence scale in Swedish clinical and nonclinical samples. Personality and Individual Differences, 24, 701 711.
175 Sanford, N. (1962). The American college . New York: Wiley. Sanford, N. (1966). Self and society: Social changes and individual development . New York: Atherton Press. Santor, D. A., & Zuroff, D. (1998). Controlling shared resources: Effects of depe ndency, self criticism, and threats to self worth. Personality and Individual Differences, 24, 237 252. Santor, D. A., Zuroff, D. C., & Fielding, A. (1997). Analysis and revision of the Depressive Experiences Questionnaire: Examining scale perf ormance as a function of scale length. Journal of Personality Assessment, 69 , 145 163. Seligman, M. E. P. (1975). Helplessness: On depression, development and death . San Francisco: Freeman. Shaver, P., Brennan, K. A. (1992). Attachment styl es and the big five personality traits: Their connections with each other and with romantic relationships outcome. Personality and Social Psychology Bulletin, 18 , 536 545. Shaver, P., & Clark, C. L. (1996). Forms of adult romantic attachment and th eir cognitive and emotional underpinnings. In G. G. Noam & K. W. Fischer (Eds.), Development and vulnerability in close relationships. (pp. 29 58). Hillsdale, NJ: Lawrence Erlbaum Associates. Shaver, P., & Norman, A. J. (1995). Attachment theo ry and counseling psychology: A commentary. Journal of Counseling Psychologist, 23 , 491 500. Simpson, J. (1990). The influence of attachment styles on romantic relationships. Journal of Personality and Social Psychology, 59 , 971 980. Simpson, J. (1999). Attachment theory in modern evolutionary perspective. In J. Cassidy & P. Shaver (Eds.), Handbook of attachment: Theory, research and clinical applications . (pp.115 140). New York: The Guilford Press. Spielberger, C. D. (1983). State Trait Anxiety Inventory: STAI (Form Y). Redwood City: Mind Garden. Spielberger, C. D., & Sydeman, S. J. (1994). State Trait Anxiety Inventory and State Trait Anger Expression Inventory. In M. E. Maruish (Ed.), The use of psychological testin g for treatment planning and outcome research. (pp. 292 321). Hillsdale, NJ: Lawrence Erlbaum Associates. Spielberger, C. D., Gorsuch, R. L., & Lushene, R. E. (1970). State Trait Anxiety Inventory for adults (Form X) . Palo Alto, CA: Consulting Psychologists Press.
176 Springer, C. A., Britt, T., & Schlenker, B. R. (1998). Codependence: Clarifying the construct. Journal of Mental Health Counseling, 20, 141 158. Stone, G. L., & Archer, J. Jr. (1990). College and university counseling c enters in the 1990s: Challenges and limitations. Counseling Psychologist, 18, 539 607. Talley, J. E., & Rockwell, W. J. (1985). Counseling and psychotherapy services for university students . Springfield, IL: Thomas. Taylor, J. A. (1953). A p ersonality scale for manifest anxiety. Journal of Abnormal Social Psychology, 48, 285 290. Torquati, J. C., & Vazsonyi, A. T. (1999). Attachment as an organizational construct for affect, appraisals, and coping of late adolescent females. Journal of Youth and Adolescence, 28 , 545 562. Toth, S. L., & Cicchetti, D. (1996). Patterns of relatedness, depressive symptomatology, and perceived competence in maltreated children. Journal of Consulting and Clinical Psychology, 64 , 32 41. Van der Linden, W. J., & Hambleton, R. K. (1997). Handbook of modern item response theory . New York: Springer Volling, B. L., Notaro, P. C., & Larsen, J. J. (1998). Adult attachment styles: Relations with emotional well being, marriage, and parenting. Family Relations, 47, 355 367. West, M. L., & Sheldon Keller, A. E. (1994). Patterns of relating: An adult attachment perspective . New York: Guilford. Whiffen, V. E., Aubeâ€™, J. A., Thomson, J. M., & Campbell, T. L. (2000). Attachment beli efs and interpersonal contexts associated with dependency and self criticism. Journal of Social and Clinical Psychology, 19 , 184 205. Yerkes, R. M., & Dodson, J. D. (1908). The relation of strength of stimulus to rapidity of habit formation. J ournal of Comparative Neurology and Psychology, 18, 459 482. Zeanah, C. H. (1996). Beyond security: A reconceptualization of attachment disorders in infancy. Journal of Consulting and Clinical Psychology, 64 , 42 52. Zuroff, D. J., Igreja, I., & Mongrain, M. (1994). Dysfunctional attitudes, dependency, and self criticism as predictors of depressive mood states: A 12 month longitudinal study. Cognitive Therapy and Research, 14 , 315 326.
177 Zuroff, D. C., Fitzpatrick, D. K. (1995). Depress ive personality Styles: Implications for adult attachment. Personality and Individual Differences, 18, 253 265. Zuroff, D. C., Moskowitz, D. S., Wielgus, M S., Powers, T. A. , & Franko, D. L. (1983). Construct validity of the Dependency and Se lf Criticism scales of the Depressive Experiences Questionnaire. Journal of Research in Personality Assessment, 17 , 226 241. Zuroff, D. C., Quinlan, D. M., & Blatt, S. J. (1990). Psychometric properties of the Depressive Experiences Questionnair e in a college population. Journal of Research in Personality Assessment, 55 , 226 241. Zung, W. W. (1973). From art to science. Archives of General Psychiatry, 29 , 328 337.
178 BIOGRAPHICAL SKETCH Ibrahim Keklik was born in Adiyaman, Turkey. His parents were tobacco farmers who lived in a small village in Southeastern Turkey. Ibrahim was the younger of two children who were the only alive children left alive of many that had di ed due to poor living circumstances and lack of medical care. During his early schooling in village elementary and middle schools, he was highly praised for his school performance, which he then decided to use to â€œgo beyond the mountainsâ€ that surrounded the geography of his childhood. From early age s until he finished college Ibrahim farmed tobacco along with the family. Ibrahim received his undergraduate degree in psychological guidance and counseling from Cukurova University in 1993. He completed co ursework requirements for the masterâ€™s degree offered in the same program while teaching in a rural elementary school. Before starting his thesis he received a scholarship from the Turkish Ministry of Education to study abroad, in 1995. After a year of training in the English language at the Middle East Technical University Ankara, Turkey, Ibrahim moved to the U.S in 1996. He completed his masterâ€™s in counseling with an emphasis on college student development from Old Dominion University, Norfolk, VA, i n 1998. In August 1998, Ibrahim was admitted to the student personal doctoral program at the Department of Counselor Education, University of Florida. After the first year of coursework, he changed his degree program to the mental health track. Ibrahim h as done
179 internships in a variety of settings. He taught undergraduate courses for over two years. He has been working at the Crisis Stabilization Unit of Meridian Behavioral Healthcare during last two years of his doctorate . Upon completion of the doctoral degr ee program, Ibrahim plans to go back to Turkey to teach, practice and conduct research in counseling. He aspires to take active roles in national and international professional organizations.