4.“. MICHIGAN STATE HUHIW l!’ minimummnmmuu'tii 3 1293 00917 3562 l! This is to certify that the dissertation entitled LATE ADOLESCENT CHILDREN OF ALCOHOLICS: AUTONOMY AND RELATEDNESS IN THEIR RELATIONSHIPS WITH THEIR PARENTS presented by VIRGINIA CAROL WRIGHT has been accepted towards fulfillment of the requirements for Ph.D. degreein Psycho'logy 4455* ? Major professor Susan Frank Date September 11, 1992 M5 U (3 an Affirmative Action/Equal Opportunity lnm’tun’on 0-12771 LIBRARY Michigan State Unlverslty PLACE IN BET URN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE MSU Is An Affirmative Action/Equal Opportunity Institution czbimmpms-pJ LATE ADOLESCENT CHILDREN OF ALCOHOLICS: AUTONOMY AND RELATEDNESS IN THEIR RELATIONSHIPS WITH THEIR PARENTS By Virginia Carol Wright A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1992 ABSTRACT LATE ADOLESCENT CHILDREN OF ALCOHOLICS: AUTONOMY AND RELATEDNESS IN THEIR RELATIONSHIPS WITH THEIR PARENTS By Virginia Carol Wright This study examined the relationship between paternal alcoholism and late adolescents' capacity for autonomy from, and relatedness with, their parents. Subjects were drawn from a large sample of college undergraduates from intact families and were selected on the basis of their own reports of the negative consequences of their fathers' drinking. Interviews provided information concerning the capacity of adolescent children of alcoholics (ACOAs) and comparison subjects for "healthy" separation/individuation (i.e., balancing the needs for increased autonomy and maintained relatedness with parents). Additionally, subjects reported on their perceptions of their parents' ability to resolve conflicts in a positive manner, without involving the adolescent in parental disputes. It was hypothesized that ACOAs would describe more difficulties with separation/individuation than would comparison subjects, and that parent marital conflict would account for some but not all of these difficulties. Interviews were coded for five aspects of autonomy (independence, decision-making, personal control, self-assertion, and self-other responsibility) and five aspects of relatedness (closeness, communication, concern, empathy, and respect). Factor analyses reduced these data by identifying three relationship dimensions: self-reliance, connectedness and emotional autonomy. Moreover, cluster analyses identified six parent/adolescent relationship categories reflecting distinct patterns of self-reliance, connectedness and emotional autonomy. Analyses indicated that, as expected, adolescents from alcoholic and non-alcoholic families differed in their descriptions of their relationships with parents. Importantly, it was found that parent marital conflict not only mediated, but also moderated, the effects of paternal alcoholism on separation/individuation, and that some group differences were not evident until marital conflict was accounted for. The importance of exploring contextual factors (such as marital conflict) when studying the effects of paternal alcoholism was discussed. ACKNOWLEDGMENTS I would like to express my sincere appreciation to the many people whose help and support facilitated the completion of this project. The study presented here is part of a larger research project undertaken by many dedicated people, all of whom deserve recognition and thanks. Susan Jackson, Lisa Crandell, and Lisa Sackett contributed enormous amounts of time and energy by assisting with both the interviewing of subjects and coding of interviews. Other interviewers for the project included: Linda Burke, Carol Laub, Carla Monestere, Mollie Poorman, Maria Townsend, and Marilyn Charles. Two undergraduate research assistants, Kai Younkin and Nicolle Barnes, deserve special credit for their extraordinary commitment to the project. Many thanks to all. Overseeing this research group was Susan Frank, who also served as the chairperson of my dissertation committee. From her, I learned not only the methods of good research, but also the value of critical thinking and necessity of diligent perseverance to complete a project with a sense of accomplishment and pride. I will always be grateful for her very important contributions to my graduate training. I also offer thanks to the other members of my dissertation committee: Frank Floyd, Gary Stollak and Robert Zucker. The four members of my committee also formed the core faculty of the Child/Family iv Clinical Psychology graduate program. I take this opportunity to thank them for both their contributions to this project, and their roles as teachers and mentors over the past five years. Finally, I reserve special thanks to my family and friends for their encouragement and support. Through verbal encouragement, silent support, and active assistance, my friends and family succeeded in letting me know that I was never alone in the process of completing this project. I express very special thanks to my friend, Lisa Sackett, for many years now of collaboration and friendship. Undoubtably, my greatest thanks go to Ron Lapporte. I cannot express in words the many things he has been to me, and I am quite sure that it was not easy being friend, partner, editor and cheerleader all at once. Yet, he successfully managed to be all that I needed, and he did so with unyielding calm. For his contributions to this project and all aspects of my life, I am deeply thankful. TABLE OF CONTENTS Page LIST OF TABLES O O O O O O O O O O O O O O O O O O O O O O O O Vii-i CHAPTER 1. Introduction to the problem . . . . . . . . . . . 1 CHAPTER 2. Review of the literature . . . . . . . . . . . . 6 Separation/Individuation in children of alcoholics . . . . . . . . . . . . . . . . 6 Competence . . . . . . . . . . . . . . . . . . . . .10 Separateness/Connectedness . . . . . . . . . . . . l3 Emotional Autonomy . . . . . . . . . . . . . . . . 17 Parental Conflict . . . . . . . . . . . . . . . . . 21 Relationships with mothers versus fathers . . . . . 24 Summary . . . . . . . . . . . . . . . . . . . . . . 28 CHAPTER 3. Hypotheses . . . . . . . . . . . . . . . . . . 30 CHAPTER 4. Method . . . . . . . . . . . . . . . . . . . . 32 ’ Subjects . . . . . . . . . . . . . . . . . . . . . . 32 Procedures . . . . . . . . . . . . . . . . . . . . . 33 Measures . . . . . . . . . . . . . . . . . . . . . . 35 Parental alcoholism . . . . . . . . . . . . . . . 35 Parental conflict . . . . . . . . . . . . . . . . 37 Autonomy and relatedness with parents . . . . . . 38 CHAPTER 5. Results . . . . . . . . . . . . . . . . . . . 43 Test of hypothesis 1 . . . . . . . . . . . . . . . . 43 vi Test of hypothesized differences in emotional autonomy, self-reliance, and connectedness between ACOAs and comparison subjects . . Additional analyses 0 O O O I O O O O O O I Descriptions of the relationship categories Analyses using the relationship categories . CHAPTER 6 0 Discussion 0 O O C O O O O O O I O O 0 Parent marital conflict . . . . . . . . . . Differences in parent/adolescent relationships for ACOAs and Comparisons . . . . . . . . Self-reliance . . . . . . . . . . . . Emotional autonomy . . . . . . . . . . Connectedness . . . . . . . . . . . . The relationship categories . . . . . . . . Limitations of the study and directions for future research . . . . . . . . . . . . . APPENDICES A. Parent Alcohol Consumption Questionnaire . . . B. Family Situation Checklist (Positive Parent Conflict Resolution Scale) . . . . . . . . . . C. Young Adult/Parent Relationship Interview . . . D. Manual for Coding the Young Adult/Parent Relationship Interview . . . . . . . . . . . . REFERENCE 8 O O O O O O O O O O 0 0 O O O O O O O O 0 vii 43 55 57 62 68 68 69 69 7O 71 74 77 83 86 88 98 119 10. 11. LIST OF TABLES Reliability estimates (Pearson correlations coefficients) for each of the parent/adolescent relationship issues Pearson correlation coefficients for the father/adolescent relationship issues, by Total Sample, ACOA and Comparison (COMP) groups Pearson correlation coefficients for the mother/adolescent relationship issues, by Total Sample, ACOA and Comparison (COMP) groups Factor loadings for the parent/adolescent relationship issues on the three relationship factors Factor scores on the Self-reliance with father dimension (broken down by ACOA group and PCR group) Factor scores on the Emotional autonomy with father dimension (broken down by ACOA group and PCR.group) Factor scores on the Connectedness with mother dimension (broken down by Sex, ACOA group and PCR group) Mean scores of the relationship factors for adolescents in each of the relationship categories Numbers of ACOA and Comparison (COMP) subjects in each of the relationship categories Mean scores of Positive Parent Conflict Resolution for adolescents in each of the relationship categories Numbers of ACOA and Comparison (COMP) subjects in each of the father relationship categories, controlling for parent conflict resolution viii Page 42 44 45 47 50 52 54 56 63 64 67 Chapter 1 Introduction £2_the Problem Past research on adolescent and young adult children of alcoholics has attempted to describe the short- and long-term negative consequences of growing up in a family where one or both parents has or has had a drinking problem. Although relatively new, the children of alcoholics research field has now provided extensive literature indicating the deleterious effects of parental alcoholism and has successfully established that children and adolescents from alcoholic families represent a population at risk. This body of literature has consisted of empirical studies, as well as clinical observations and descriptive information regarding the characteristics that differentiate children of problem drinking parents from those who were raised in non-alcoholic families. In addition, existing research provides evidence for both behavioral and emotional difficulties among adolescent and adult children of problemrdrinking parents. Children of alcoholics have been described as exhibiting numerous behavioral problems both as children and adolescents. Younger children of alcoholics are disprOportionately identified as hyperactive (el- Guebaly & Offord, 1977), conduct disordered (Steinhausen, Gobel & Nestle, 1984; Lund & Landesman-Dwyer, 1979), and truant (Rimmer, 1982). Adolescent and young adult children of alcoholics also have been described as engaging in delinquent behavior (Adler & Raphael, 1983), and as employing less effective problem-solving strategies (Clair & Genest, 1987), displaying controlling (Cermak & Brown, 1982) and passive-aggressive behaviors (Bogdaniak & Piercy, 1987), having difficulty establishing and maintaining intimate relationships (Seixas, 1982; Cermak & Brown, 1982; Woititz, 1978; Wood, 1987), and at risk for alcohol misuse (Schuckit & Sweeney, 1987; Parker & Harford, 1988). In addition to the behavioral manifestations of difficulties associated with growing up with an alcoholic parent, children of alcoholics also appear to be at risk for developing emotional difficulties. For example, research and clinical observations have indicated that children of alcoholics are more prone to higher degrees of neuroticism (Benson & Heller, 1987), poor boundary formation (Cermak & Brown, 1982), depression (Clair & Genest, 1987), low self-esteem (Ackerman, 1983; Hughes, 1977) and excessive dependence upon defensive denial and excessive feelings of guilt (Bogdaniak & Piercy, 1987). Although some contributors to the children of alcoholics literature (e.g., Black, 1979 and Woititz, 1978) maintain that all children of alcoholics suffer negative consequences as adults, this supposition is based almost exclusively on clinical observations and has not been substantiated by controlled empirical research. In fact, a significant number of researchers recognize that many, if not most, adult children of alcoholics are psychologically and socially well-adjusted (e.g., Heller, Sher, & Benson, 1982; Burk & Sher, 1988; Werner, 1986; West & Prinz, 1987). This relatively recent recognition of so-called "invulnerable" children of alcoholics (albeit based predominantly on clinical observations and/or methodologically flawed studies) has inspired a wave of new research designed to identify and describe potential mediating or moderating variables in the relationship between parent alcoholism and various child and adult outcomes (e.g., Adler & Raphael, 1983; Clair 8 Genest, 1987; Rogasch, Chassin & Sher, 1990). Researchers have suggested that the "effects of parental alcoholism need to be viewed within a theoretical framework that considers multiple sources of stress on children" (West & Prinz, 1987, p. 216) and that "the critical issue at this time is the clarification of the parameters of risk, those biological and psychosocial factors that mediate vulnerability" (Seilhamer & Jacob, 1989, p. 169). "Clarification of the parameters of risk" has been initiated through both theoretical and empirical examinations of the experience of growing up in an alcoholic family. In one of the earlier examinations, Robins, West, Ratcliff, and Herjanic (1978) empirically identified several risk factors associated with paternal alcoholism: an increased number of illegitimate children, having a younger father, living in a broken home and having a father who has been arrested. Robins et al. found that these risk factors were predictive of high rates of truancy and school drapout in the offspring of alcoholics, thereby demonstrating the potential mediating properties of these sorts of concomitant factors associated with a father's problem drinking. Other variables which have been hypothesized as potential mediators of outcome include family disharmony (Adler & Raphael, 1983; Benson & Heller, 1987; Werner, 1986), impaired socialization (Ackerman, 1983; Werner, 1986), and disruptions in family system maintenance, such as discontinued family rituals (Wolin, Bennett, Noonan, & Teitelbaum, 1980; Bennett & Wolin, 1986) or poor family organization (Clair & Genest, 1987; Filstead, McElfresh, & Anderson, 1981). Theoretical arguments suggest that relationships between children of alcoholics and their parents constitute a potentially important mediator of outcome. Davis, Stern, and Vandeusen (1978), for example, suggested that "alcoholism may be consistently correlated with behavioral pathologies at the family level, but the most evident and powerful effects may work their way through subsystems" (p.24), including the parent-child subsystem. Similarly, Moore (1982) PrOposed a comprehensive model of the effects of parental alcoholism on children in which he suggested that child adjustment is dependent upon the disruptions of three primary factors: the quality and style of the relationship between parent and child, the style and consistency of parent supervision, and the effectiveness of direct parent socialization. According to Moore, secondary factors include marital conflict, social isolation, and family crisis, and these factors impact child adjustment by disrupting the three primary factors. Given the potential mediating influence of the quality and nature of parent/child relationships in alcoholic families, exploration of the nature of these relationships is critical. The current study empirically investigated the nature of parent/adolescent relationships in alcoholic families. Specifically, interviews provided information concerning the nature of the relationships between children of alcoholics and their parents during late adolescence -- a period when developmental tasks necessitate restructuring of relationships with parents such that increased autonomy from and continuing relatedness with parents is possible. In addition, this study assessed and compared separation/individuation processes among late adolescent children of alcoholics with those of late adolescent children of non-alcoholic parents. Hypotheses concerning the nature of relationships between children and their alcoholic parents have been criticized for their laCk of a theoretical base (Nardi, 1981) and limited empirical support (Searles & Windle, 1990). While clinical and theoretical discussions have emphasized for some time the processes by which children of alcoholics come to exhibit various outcomes, validation of such mediational models through empirical studies has begun relatively recently. There is relatively little in the way of systematic research expounding, in particular, on the nature of the parent/adolescent relationships within alcoholic families. Nevertheless, there is evidence to suggest that children's experiences of their parents as supportive (Clair & Genest, 1987; Benson & Heller, 1987) and as encouraging of autonomy (Benson & Heller, 1987) mediate the relationships between parental alcoholism and the psychological adjustment of adolescent and young adult children. Likewise, a questionnaire study of late adolescent children of alcoholics (Wright, Frank, & Pirsch, 1990) suggests that both parental alcoholism and parental conflict may be more distal predictors and late adolescents' relationships with their parents may be more proximal predictors of deviations in separation/individuation processes for this age group. Chapter 2 Review g£_the Literature Separation/individuation.12.children.g£.alcoholics Theoretical discussions of the effects of parental alcoholism on child outcomes suggest that adolescent children of alcoholics may have difficulties with separation and individuation from the family of origin (e.g., Wood, 1987). Supporting evidence comes predominantly from clinical observations. For example, the inability of some children of alcoholics to disengage from parental needs and make satisfying commitments to love and work (Woititz, 1983; Black, 1981) and their difficulties disassociating their own distress from their parent's alcoholism (Berlin, Davis, & Orenstein, 1988) suggest that individuation may often remain incomplete. Theorists have emphasized different aspects of the separation/individuation process: whereas some emphasize the importance of emotional disengagement, behavioral autonomy, and self-directedness, others underscore an increasingly mutuality in adolescent/parent relationships. Psychoanalytic theorists traditionally belong to the first group in that they describe separation/individuation as a process by which adolescents emotionally disengage from parents in an attempt to separate and become more independent (Blos, 1979). In contrast, others maintain that "healthy individuation" is characterized by increasing symmetry of influence or "mutuality" that can occur only within the context of continued relatedness with parents (Allen, 1990; Grotevant & C00per, 1986; Ryan & Lynch, 1989). More recently, research (Frank, Pirsch & Wright, 1991) has supported the belief of some theorists (Josselson, 1980; Weiss, 1982) that, for most adolescents, separation/individuation is a dialectical process whereby adolescents partially disengage from parents in an attempt to make gains in self- regulation, yet avoid excessive estrangement and insecurity by maintaining a sense of connectedness with their parents. Evidence provided primarily by clinical observations suggest that, in contrast, children of alcoholics may experience extreme connectedness with their parents in some ways and extreme disengagement in others, neither of which provide sufficient opportunity for the development of "healthy autonomy" (Allen, 1990). It is interesting to note, however, that preliminary studies of parent/adolescent relationships suggest that late adolescent children of alcoholics are mgrg_rather than less autonomous from their parents than children of non-alcoholics (Wright, Frank, & Pirsch, 1990). An understanding of these seemingly contradictory findings may come from examining various definitions of autonomy. Hill and Holmbeck (1986), in a review of the adolescent autonomy literature, point out that most researchers have defined autonomy in terms of "freedom from" parents (e.g., opportunities for less supervised activity) rather than more proactive notions of self- governance and, in general, have ignored more intrapsychic aspects of autonomous development. Frank, Avery and Laman (1988) suggest that psychoanalytic formulations of adolescence actually point to several different aspects of autonomy from parents. Their research, which involved interviewing young adults (ages 22-32) about their relationships with their parents, supported three separate notions of autonomy: competence, separateness (vs. connectedness) and emotional autonomy. In particular, increased autonomy involves a)changes in one's sense of competence in dealing with the external environment, b) the ability to "master intrapsychic conflicts and related feelings of shame, guilt or rage associated with undifferentiated parental representations and identification" (p. 731) and c) one's perception of separateness and appropriate boundaries between one's own and one's parents' lives. Young adults who were high on the competence dimension described themselves as able to make decisions without undue input from their parents (high decision making) and as able to cape with life's difficulties without needing their parents' assistance (high independence). A sense of young adults' connection with parents was indicated by the separateness vs. connectedness dimension. In particular, this dimension consisted of five issues: strong emotional ties to the parent (closeness); interest in the parent's well-being (concern); an ability to communicate with the parent about personally meaningful feelings (communication); an ability to see the parent as a complex person (empathy); and continued loyalty and investment to the parent-child relationship rather than investments in new sources of affection and fulfillment (self-other responsibility, i.e., greater emphasis on responsibility to the parent rather than to the self). Finally, the emotional autonomy dimension provided an indication of the extent to which young adults considered themselves (rather than their parents) to be the best evaluators of their own self-worth (self- assertion), the extent to which they were in control of, rather than controlled by, their feelings toward their parents (personal control), and the extent to which the young adults perceived their parents as suitable and competent role models (respect). Evidence suggests that children of alcoholics may face disruptions in each of these three aspects of the separation/individuation process, although it is important to note that our knowledge of the nature of parent/adolescent relationships in alcoholic families is heavily reliant upon clinical observations and much less so on empirical investigations. The relatively few studies which do exist can be criticized on a number of bases, including: a)a lack of control groups, especially groups of children with parents suffering from other types of psychopathology; b)failure to control for additional stress factors which are over~represented in alcoholic families (e.g., marital conflict, unemployment, violence, etc.); c)failure to assess the mental status of the nonalcoholic spouse and the degree to which the nonalcoholic parents may serve as a ' 'protective agent" against negative effects of the alcoholism; d)vague and inconsistent definitions of alcoholism; e)little attention to variables related to parent drinking such as duration, severity, etc.; and f)1ack of attention to the interaction of the sex of the drinking parent and the sex of the child (Seilhamer & Jacob, 1990). Clinical reports are also frequently criticized for their lack of theoretical foundations and causal hypotheses. Despite these methodological shortcomings, an argument can be made 10 for disruptions in the separation/individuation process for adolescent children of alcoholics. Competence. As noted above, Frank et al. (1988) describe "competence" as that aspect of separation/individuation that is related to adolescents' increased ability to make their own decisions and to face life's challenges without excessively relying on their parents. There is difficulty, however, with the "competence" label, as it does not distinguish between a)aelf-reliance within the context of adequate capacity for responsible decision-making ("competence") and b)self- reliance which arises through the perception of the parent as inadequate or unavailable as a source of assistance and advice ("self-reliance"). The literature suggests that adolescent children of alcoholics are highly self-reliant: they do not turn to parents for assistance in making decisions or coping with difficulties (Clair & Genest, 1987; Wright et al. 1990), yet they may be unprepared to make their own decisions responsibly. Frank et al. (1988) identified a similar group of young adults as "pseudoautonomous": they made their own decisions and caped without assistance from the parent, yet were distant from the parent and engaged in ongoing conflict. Clinical accounts suggest that adolescent children of alcoholics may develop a sense of self-reliance (in the sense of minimal reliance on parents for guidance and assistance) earlier than children from non- alcoholic families. Specifically, children of alcoholics are often described as hyper-responsible and somewhat precocious in their sense of responsibility for making important family decisions (e.g., Bogdaniak & Piercy, 1987). Some empirical findings corroborate these descriptions in indicating that children of alcoholics may be more behaviorally 11 autonomous and self-governing as young adults than are children of non- alcoholic parents (e.g., Clair 8 Genest, 1987; Wright et al., 1990). However, while children of alcoholics are likely to make independent decisions and face life's challenges with minimal parental input, they may be somewhat unequipped to do so successfully. Research has suggested that "competence" is fostered by authoritative parenting where warm and supportive parent/adolescent relationships are coupled with moderate parental control over the adolescent (Devereux, 1970). While competence may be related to authoritative parenting, premature self- reliance and independent decision-making may be linked to permissive parenting styles which force children to assume adult roles early in life. This latter form of self-reliance may be less successful, as it occurs without the support and guidance of parents and before adolescents are sufficiently mature to handle life's difficulties on their own. Clinical observations suggest that alcoholic families are more likely to exhibit permissive and even chaotic, rather than the authoritative, parenting styles. Not only is the alcoholic his or herself described as irresponsible and unpredictable (Woititz, 1978; Arentzen, 1978), but because of the problem drinker's continual cycle of intoxication and sobriety, consistent controls in the family as a whole are replaced with chaos and unpredictability in alcoholic homes (Morehouse 8 Richards, 1982). The presence of alcoholism not only increases the risk for inconsistent controls, but for affectional and emotional inconsistency, as well. Hence, children of alcoholics may not be able to rely on parents for the nurturance and support necessary to 12 facilitate adolescent development of competence (i.e., self-reliance based on maturity rather than neglect). In one of the few empirical investigations of emotional inconsistency in alcoholic families, Benson and Heller (1987) reported that young adult daughters of alcoholic fathers perceived themselves as having relationships with their fathers that were more inconsistent with regard to love and affection than did daughters of fathers without a drinking problem. Although this finding is consistent with clinical descriptions of adolescents' relationships with their alcoholic parents, the investigation did not explore sons' perceptions of their relationships with alcoholic parents and did not clearly differentiate between parental alcoholism and problem drinking. Nevertheless, it appears that the warmth, consistency and moderate control exhibited by authoritative parents and hypothesized as important for the development of "healthy autonomy" (Allen et al., 1990) or competence during late adolescence is often lacking in alcoholic families. Theoretical discussions which are heavily reliant upon object relations theory further expound on why children of alcoholics may be at risk for difficulties in obtaining "competence" during adolescence. Fairbairn (1943) suggests that children are utterly dependent upon their parents and thus find it psychically intolerable to accept them as "bad." Consequently, children will apt to experience themselves as bad before sensing the insecurity associated with parents who are unreliable. This phenomenon has been applied to children of alcoholics who are often described as taking on blame and self- punitiveness for their parent's drinking (Woititz, 1983). Similarly, Berlin, Davis and Ornstein (1988) suggest that l3 adolescent children of alcoholics' difficulty separating from their families is due, in part, to the fact that their sense of self is tied to their parent's drinking. These children have "learned to disassociate their parent's drinking from their own personal distress" (p. 589), thus suggesting the degree to which children of alcoholics experience negative affect as self-related rather than other-related. Berlin et a1. further differentiate adolescents' reactive and adaptive distancing from their families. Adaptive distancing is characterized by movement toward relationships and activities which allow for "reparative work." Reactive distancing, on the other hand, results from the disassociation of parental failure and one's subjective distress and is characterized by flight and isolation. This pattern of flight and isolation is often seen in adolescent children of alcoholics who may consequently be self-reliant and make decisions independent of parental input but, because of denial, may be unable to "repair" feelings of inadequacy or distress. Clinical, theoretical and empirical evidence suggest that, despite impediments to the development of competence, many children of alcoholics are unusually self-reliant. Such observations may further suggest that children of alcoholics' sense of self-reliance may be quite high but also may be premature and thus unrelated, or even negatively related, to other aspects of healthy autonomy from parents (Allen et a1. 1990). Separateness vs. connectedness. Some theorists have defined autonomy in terms of balancing needs for separateness and connectedness with their parents (e.g., White, Speisman 8 Costos, 1983). Frank et al. 14 (1988), as noted above, identified a separateness vs. connectedness dimension of autonomy which at the connectedness end of the continuum described between young adults who maintained very close relationships with parents and at the separateness end of the continuum referred to young adults who engaged in less intense relationships with their parents and thus maintained clearer boundaries between their own and their parents' lives. More specifically, young adults who reported high connectedness with their parents described greater emotional ties to the parent, greater concern for the parent's well-being, a greater capacity to see the parent as a complex person, extensive communication with the parent about personally meaningful subjects, and continued loyalty to the parent/child relationship in lieu of new sources of affection. It appears that "optimal" autonomous development involves achieving an apprOpriate balance between connection to and separateness from parents. Disruptions for adolescent children of alcoholics on the separateness vs. connectedness dimension may be quite complex. Clinical descriptions of children of alcoholics suggest that these adolescents might exhibit a high (perhaps even excessive) degree of concern for the parent's well-being, as well as continued loyalty to the parent/child relationship at a cost to external sources of fulfillment (Black, 1981). However, it seems less likely that adolescent children of alcoholics would report the related experiences of closeness and adequate communication. Children of alcoholics' capacity for empathy toward their parents (i.e., their ability to see their parents as complex peOple) is even more difficult to predict. The excessive denial which is characteristic of alcoholic families may impede offspring's ability to see their parents in a realistic manner; however, extreme 15 dysfunction and behavioral inconsistencies within alcoholic families may force children of alcoholics to observe and consider their parents in a more critical and perhaps complex manner. Hence, children of alcoholics may exhibit extreme highs on some aspects and extreme lows on other aspects of the separateness/connectedness dimension. Two commonly acknowledged descriptions of the alcoholic family experience may help to explain the potential difficulty of adolescent children of alcoholics in maintaining a balance between connectedness and separateness: unavailability of parents and role reversals. According to clinical reports, chronic heavy drinking not only diminishes problem drinkers' physical availability (Hecht, 1973), but also their ability to interact with and be empathic or emotionally available to others (Morehouse 8 Richards, 1983). In addition to the alcoholic parent's physical and emotional unavailability, the alcoholic family's tendency to collude with the alcoholic's rationalizations may increase the risk of social isolation of the entire family as they attempt to minimize outside knowledge of the alcoholic's problem. Straussner, Weinstein and Hernandex (1979) suggest that the social isolation experienced by the alcoholic family is compounded by emotional isolation as family members are unable to share their feelings of resentment, embarrassment, etc. with one another. Children of alcoholics' experience of isolation, both emotional and social, may render them less able to invest in interpersonal relationships outside the family and, as such, unable to separate from their parents. At the same time, relationships within these alcoholic families may remain distant and unfulfilling. 16 In addition to becoming isolated from external relationships, alcoholic families are also noted as displaying role conflicts. Nardi (1981) notes the often-cited uncertainty and changability of both the parents' and the child's roles and the high likelihood of role reversals in these families. Children within alcoholic families are often asked to assume responsibility for taking care of the alcoholic and non- alcoholic spouse (Homonoff 8 Stephen, 1979), as well as younger siblings (Bogdaniak 8 Piercy, 1987). It has been further suggested that children who are afforded responsibilities during periods of intoxication often have these responsibilities revoked when sobriety returns (Seixas, 1982; Hecht, 1973). The dramatic shift between the experience of an excess of freedom and parents' subsequent re-capturing of all responsibility and control lead not only to role confusion (Morehouse 8 Richards, 1983), but also to a difficult to rationalize preference for parents when drunk (Seixas, 1982). Furthermore, excessive role reversal and over- concern with the parents' well-being may interfere with children's efforts to separate from their parents and invest in outside relationships and activities. Although many adolescent children of alcoholics may have difficulty separating from their parents due to social isolation or excessive role reversals, it appears that some children of alcoholics are quite distant from their parents. Berlin et al. (1988) explain that adolescent children of alcoholics' difficulty separating from their parents -- due to loyalty to parents and assumed responsibility for their well-being -- may, in fact, result in abrupt disengagement from their families (e.g., through running away or an overinvolvement in the peer culture or activities outside the home). It appears, then, that while Optimal 17 adolescent development entails balancing separateness from and connectedness to parents, children of alcoholics may instead fall on either extreme of this continuum or on both extremes depending upon which aspect of the separateness/connectedness dimension is under consideration. Namely, adolescents from alcoholic families may exhibit high degrees of concern, and possibly loyalty and empathy, for their parents, while exhibiting unusually low levels of communication and closeness with parents. Emotional Autonomy. Historical and contemporary definitions of emotional autonomy have been quite varied. As previously noted, research by Frank et al. (1988) suggest three components of emotional autonomy: the capacity to utilize one's own evaluations (rather than the parent's) for determining self-worth (i.e., self-assertion); the capacity to be in control of, rather than controlled by, negative feelings toward parent (i.e., personal control); and the capacity to maintain respect for one's parents. Classic psychoanalytic writings suggest that adolescence is a period of "reawakened conflicts" with parents and that the resulting tension between parent and child necessitates adolescents' emotional detachment from parents (Freud, 1958). However, the conceptualization of emotional autonomy as detachment from parents has received much criticism for its inattention to the fact that most adolescents maintain close relationships with their parents and do not report excessive tensions with or estrangement from their parents (Douvan 8 Adelson, 1966; Kandel 8 Lesser, 1972; Offer, Ostrov 8 Howard, 1981). As an alternative, Blos (1967) suggests that emotional autonomy 18 refers to the relinquishing of identifications with childish parental representations and dependencies rather than of emotional ties to the "reality" parent. Although he does not ascribe to earlier notions of excessive (yet normative) tensions between adolescents and their parents, Blos does suggest that adolescents experience a moderate degree of inner turmoil as they master psychic conflicts surrounding deidealization such that they are able to replace idealized images of parental omnipotence with more realistic observations of parents' faults, weaknesses and vulnerabilities. Frank et al. (1988) have also described deidealization and emotional autonomy as the recognition of the parent's fallibilities and as involving the "capacity to internally regulate self-esteem without fear of 'archaic guilt' or parental disapproval" (p.736). Presumably, this process is best mastered in the context of continued closeness to and support from parents. As noted above, this may be less available to children of alcoholics. Several clinicians and theorists (Wood, 1987; Brown, 1988; Seixas, 1988) have suggested that children of alcoholics often face difficulties developing emotional autonomy because their relationships with their alcoholic parents have not allowed for the development of a mature and differentiated self. Wood (1987) suggests several reasons why the difficulties displayed by many late adolescent and adult children of alcoholics can be understood as a manifestation of, in object relations terminology, a "damaged self." First, a "healthy self" is able to maintain attachments to other persons despite disappointments and conflicts and this is notably absent in children of alcoholics. Second, a "damaged self" requires constant praise and affirmation, a quality often seen in children of alcoholics. Finally, a stable, secure self is 19 able to tolerate separateness from loved others while a shaky, undifferentiated self requires persistent attachment to the family, often focusing narrowly on the needs of others at the expense of oneself. Clinical observations suggest that children of alcoholics often remain loyal to and protective of their families, often to the detriment of autonomous growth. Clinical observations have supported theoretical suppositions that children of alcoholics may have a diminished capacity for emotional autonomy. Clinicians have described adolescent children of alcoholics as disproportionately affected by feelings of anger and hostility toward their parents (Cork, 1969). High levels of hostility and conflict in alcoholic families are manifest in different ways. Seixas (1982) suggests that conflict in alcoholic families may often take the form of long periods of panting and quiet tension. Because direct expression of conflict may often be thwarted, and because there is frequently a chronic state of tension, members of an alcoholic family may be provided with little Opportunity to share negative feedback in any apprOpriate manner. Family members' feelings of anger and disappointment, then, are frequently expressed during bouts of rage (Straussner et al., 1979). The unpredictability and volatile nature of expressed affect in alcoholic families renders all family members, and perhaps especially the children, distrustful not only of the family environment, but also of the experience and expression of affect. Unresolved feelings of hostility toward parents and a lack of understanding of how to resolve conflict may have serious negative implications for adolescent's capacity to deidealize the parent and 20 develop emotional autonomy. Some neoanalytic writers have suggested that negative feelings toward parents can be so overwhelming that they dictate the child's behavior and hence impair his or her strivings for autonomy (Blos, 1967). From a theoretical viewpoint, one can argue that the deidealization process (the precursor to emotional autonomy) is particularly difficult for children of alcoholics. Some theorists (e.g., Blos, 1967) have suggested that children cannot truly deidealize parental representations unless they have previously idealized the parent. As described above, children of alcoholics may chronically experience unresolved feelings of anger and hostility toward parents. Consequently, these adolescents may never idealize their parents. Rather, they may experience intense negative affect toward their parents which, from the start, is untempered by feelings of love and respect. This first scenario may be contrasted with a second in which the child, rather than holding contempt for his parents, maintains a rigidly idealized view of the parent because of excessive denial of parental shortcomings. Clinical accounts depict denial as a primary defense for the alcoholic and the members of his family (e.g., Seixas, 1982; Straussner et al., 1979). It could be argued, of course, that many children of alcoholics express negative feelings toward their parents and thus appear to exhibit excessive and perhaps premature deidealization. Regardless, when children either excessively denigrate or overidealize their parent, they are unable to use self-standards for self-evaluation. Instead, their self-evaluations are driven by fears of being like the parent or fears of parental disapproval. It is this inability to use self-standards for self-evaluation which is the hallmark of low emotional autonomy. 21 In sum, it appears that adolescent children of alcoholic parents may experience not only disrupted autonomous growth, but also a qualitatively different pattern of competence, separateness and emotional autonomy than do children of non-alcoholic parents. Children of alcoholics may have particular difficulty balancing the needs for separateness from and connectedness with their parents. Furthermore, the involvement of these adolescents in their parents' lives (and vice versa) may result in a pseudo-maturity or pseudo-autonomy whereby children of alcoholics may be overly self-reliant while maintaining excessive intrapsychic connections and/or experiencing both emotional estrangement from and a high degree of conflict with their parents. Parental Conflict A review of theoretical and clinical examinations of children of alcoholics suggests that these adolescents are at risk for special problems, including those associated with the task of becoming autonomous and maintaining relatedness with parents. However, it is important to note that children of alcoholics share many of the same characteristics (both individual and familial) with children of parents with other types of psychOpathology. Evidence for this has been provided, in part, by Theodore Jacob and his colleagues in their comparisons of children and families with alcoholic fathers and depressed fathers (Jacob, Krahn, and Leonard, 1991; Jacob and Leonard, 1986). These researchers have documented, for example, that these two groups of children and families do not differ significantly from one another with regard to parent- or teacher-rated child behavior problems, or family problemrsolving difficulties. Although these 22 findings may be explained, in part, by the proposed theoretical relationship between alcoholism and depression (see Zucker, 1987), many researchers now suggest that adverse outcomes for COAs may be related to non-alcohol-specific effects of parental disturbance (e.g., Chassin, Rogasch 8 Berrera, 1991; Sher, Walitzer, Wood 8 Brent, 1991). One potential factor in this regard is the presence of marital conflict. It has been documented that disordered individuals are more likely to experience marital discord and divorce (Molholm 8 Dinitz, 1972) and that children of discord and divorce are at increased risk for a variety of behavioral and emotional difficulties (Wallerstein 8 Kelly, 1980). Several researchers have suggested that the marital conflict associated with parent alcoholism mediates between parent alcoholism and the quality and style of parent-child relationships and other child outcomes (e.g., Moore, 1982; West 8 Prinz, 1987). It has been well documented that alcoholic couples are at risk for marital discord and divorce (Woodruff, Guze, and Clayton, 1972) and higher than average levels of negativity in communication, especially during periods of intoxication (Jacob, 1986). Additionally, research suggesting that even during periods of sobriety, alcoholic couples experience high levels of conflict and have difficulty handling feelings of hostility (Mitchell, 1958) underscore the importance of recognizing marital conflict as a contributing factor to the functioning of the alcoholic family's members. Empirical support for the impact of marital conflict on parent/adolescent relationships comes from two studies of the relationship between late adolescents and their parents. Frank and Burke (1992) have found that children of divorce experience many of the 23 same relationship patterns (i.e., increased self-reliance and decreased relatedness) during late adolescence as prOposed for children of alcoholics. Furthermore, Wright et al. (1990) showed that late adolescent children of alcoholics perceived their parents as less able to engage in positive conflict resolution and more likely to involve their children in their marital disputes than do children of nonalcoholic parents and that parents' poor conflict resolution skills mediated the effects of parental alcoholism on the late adolescents' experiences of autonomy, relatedness, and mutuality with their parents. A recent review by Sessa and Steinberg (1991) lends insight into why marital conflict may be particularly relevant to the study of separation/individuation processes in adolescence. Although Sessa and Steinberg address the development of autonomy as it relates to divorce rather than to marital conflict, many of the negative outcomes of divorce are, in fact, more related to the conflict precipitating divorce than to the divorce itself (Grych 8 Fincham, 1990). Thus, their discussion can be easily generalized to explain how marital conflict may disrupt adolescents' sense of competence, separateness (vs. connectedness) and emotional autonomy. In particular, Sessa 8 Steinberg (1991) suggest that adolescents' sense of competence may be prematurely deveIOped as parents who are in chronic conflict display what Wallerstein and Kelly (1980) describe as a "diminished capacity to parent." Wallerstein and Kelly describe "diminished parenting" as decreased affection, control, monitoring and communication and note that such parenting necessitates the adolescent's self-sufficiency. Additionally, children who are exposed to excessive 24 parental conflict may, according to Sessa and Steinberg (1991), cape with such conflict by disengaging from the family and seeking outside sources of support. However, children whose parents fight frequently may also feel concerned about one or both of their parents' well-being and consequently may become overly involved in the parents' lives. Hence, parental conflict, like parental alcoholism, may contribute to difficulties for the adolescent in balancing connectedness and separateness with the parent. Finally, marital conflict may instigate early deidealization and interfere with emotional autonomy in that children witness their parents' shortcomings and fallibilities (through their excessive fighting) and may prematurely deidealize images of parental omnipotence or be over-deprecating of their parents. Disruptions in the three aspects of separation/individuation brought about in part by marital conflict are described above as causing early autonomous growth. Sessa and Steinberg argue that premature autonomy may, in fact, be a desirable outcome. However, when autonomous growth occurs in the context of over-involvement in the parents' lives or as a result of premature deidealization, the result may be a sort of "pseudo-autonomy" which has not provided the adolescent with the foundation for healthy separation/individuation. Relationships with mothers versus fathers Because of a high likelihood of marital conflict within alcoholic families, and because marital discord may prevent spouses from establishing and maintaining a strong parental alliance (Christensen and Margolin, 1988), children of alcoholics may be less likely than most children to experience parents as a unified executive system. The 25 existence of an internal division with the parenting subsystem may subsequently create a situation in which the child's relationship with each parent -- independent of one another -- is unusually distinct. For this reason, discussions of children of alcoholic's relationships with parents must account for variability in the quality and nature of relationships with mothers and fathers. This may be particularly true when one parent is alcoholic and the other is not. Some clinicians and researchers have argued that a positive relationship with the nonalcoholic spouse, usually the mother, may serve as a compensatory factor which, in effect, buffers the negative impact of the other parent's problem drinking. For example, Obuchowska (1974) found that fifth grade children with alcoholic fathers who had positive emotional contact with their mothers displayed more social and achievement-oriented behaviors than did those who did not have positive emotional contact with their mothers. Furthermore, the high levels of aggression and resignation displayed by this latter group closely resembled the behaviors displayed by children who had two alcohol- addicted parents. However, some clinical reports suggest that while some adolescents' relationships with their nonalcoholic parent may help to buffer the negative impact of alcoholism, other non-drinking spouses may be too overwhelmed with their own feelings of anger and frustration to be of help their children (Bosma, 1972). In fact, some clinicians suggest that children of alcoholics may ultimately become more resentful toward the nonalcoholic than the alcoholic parent (e.g., Cork, 1969; Seixas, 1982). One important consideration with regard to children's relationship 26 with mothers versus fathers is the gender of the child, even though this factor has often been ignored in outcome studies (West and Prinz, 1987; Benson and Heller, 1987). Furthermore, it appears important to study not only the implication of the child's gender, but also the interaction of parent and child gender. In normative samples, theinteraction of parent and child gender has been documented. Specifically, both sons and daughters experience fathers as "authority" figures who provide advice and practical guidance and mothers as both authority figures and_ intimate confidantes. While the nature Of differences between relationships with mothers and fathers are similar for male and female adolescents, the daughters' experiences Of their relationships with their mothers and fathers appear to be more disparate (Youniss 8 Smollar, 1985). In particular, the intimate aspect Of the mother- daughter relationship is quite different than the mother-son relationship in that daughters appear to perceive their mothers as "persons who need their help as much as they perceive them as persons who can help them" (p. 51). Although the interaction Of parent and child gender within alcoholic families has not been systematically studied, theoretical frameworks can again provide insight in this regard. For example, Nardi (1981) suggests that "both the socialization process and the modeling process must be considered in the differing effects on male and female children of alcoholic mothers or fathers" (p.241). In particular, he argues that parents in alcoholic families may be absent (either emotionally or physically) and children may thus have decreased Opportunities for modeling parents. He further notes that adolescents rely on modeling of same-sex parents and that disruptions in sex-role 27 development resulting from unavailable parents in alcoholic families may have negative implications for the develOpment Of self-esteem, identity, and locus of control. Because the father is typically the identified alcoholic, and hence the most absent, parental unavailability may create particular difficulties for sons. From an analytic viewpoint, the absence Of the same-sex parent may have additional implications, especially during adolescence. In particular, the normal regressive nature Of adolescence is Often manifest in the re-emergence Of oedipal conflicts (Adelson and Doerhman, 1980) and this suggests the need for same-sexed parents to defuse an emotionally intense relationship between children and their Opposite sexed parents. When the same-sex parent is unavailable in this regard, disrupted parent-child relationships may be most evidenced in relationships between fathers and daughters and between mothers and sons. The nature of this disruption might be expected, from this paradigm, to be evidenced in an inapprOpriate degree Of concern for and connectedness with the opposite sex parent. Finally, research by Glen Elder and his colleagues has provided support for the notion that family disruptions may result in differential changes in the various dyadic relationships within the family and may consequently have different implications for male and female Offspring (Elder 8 Caspi, 1990). Elder and Caspi studied changes in family relationships within the context Of the Depression. They found that paternal unemployment and the related increases in fathers' heavy drinking, emotional depression and inconsistency in the discipline Of children resulted in increases in mothers' "power" within the family. 28 Mothers became the providers Of both affection toward and discipline of the children. Male and female children appeared to respond differently to this shift in family roles. In particular, daughters appeared to form a strong bond with their mothers, perhaps forming an alliance "against" the devalued father. In contrast, it appeared that some, if not many, sons perceived their mothers' increased influence as hyper- critical and overbearing. These males appeared to fare best when they engaged in external activities which decreased their interaction within the family (e.g., through military service, jobs outside the home, etc.) and hence accelerated what has previously been described as their self-reliance. There exists no systematic research to support a similar pattern of parent/child relationships within alcoholic families, although the presence Of a "devalued" father and hence stronger mother in alcoholic families is evident. In any case, the interaction Of parent and child gender in alcoholic families appears worthy of further study. Summary Information provided primarily by clinical and theoretical discussions suggests that the nature of the relationship between parents and children in alcoholic homes may be an important mediator Of child outcomes. Yet relatively few empirical investigations have provided information as to the nature of these relationships. The current research attempted to provide empirical validation for clinically and theoretically derived hypotheses concerning separation/individuation in late adolescents from alcoholic and non-alcoholic families. Additionally, the present study attempted to address some of the limitations of past research, most notably by measuring and controlling 29 for marital conflict and by considering the gender of both parents and children when describing the nature Of their relationships. Chapter 3 Hypotheses Hypotheses were as follows: 1) ACOAs will report significantly more parent marital conflict than will children of non-alcoholics. Although the degree of reported marital conflict will partly mediate the effects Of parental alcoholism on adolescents' relationships with their parents, some drinking group differences in the parent/adolescent relationship will remain even after controlling for parent marital conflict. 2) In contrast to the comparison subjects, ACOAs will describe greater self-reliance (i.e., less reliance on parents for dealing with the challenges in their lives or making important decisions). 3) Compared to adolescent children Of non-alcoholics, ACOAs will describe less emotional autonomy (i.e., more difficulty less self- assertion, less personal control, and less respect). 4) Late adolescent children Of alcoholics will describe higher degrees of concern and empathy for their parents, and greater involvement in their parents' lives compared to children Of non- alcoholics, but ACOAs will also report less communication and less closeness in their relationships with their parents. In addition to these specific hypotheses, the study design allowed for an exploration of the implication Of parent and child gender for the 30 31 nature of these relationships within alcoholic families and families where drinking is not a problem. Chapter 4 Method Subjects Data for the prOposed study were collected as part Of a larger study conducted at Michigan State University, on "Development Through the College Years.‘ An initial sample Of approximately 1300 undergraduate students enrolled in Introductory Psychology courses received research credit in exchange for their participation in this larger study. In order to be eligible for inclusion in this study students had to be between the ages Of 17 and 22 and from intact families. Students from non-intact families were not recruited because a child's contact with one or the other parent (usually the father) is Often decreased following a divorce and because we decided to control for, rather than examine, the effects of parent marital status. This means that the families of subjects with an alcoholic parent may be representative of a somewhat less dysfunctional segment of the larger pOpulation of families with a history Of paternal alcoholism. From this larger subject pOOl, a smaller number of students were selected to participate in a second phase Of the experiment. This smaller group Of 125 subjects (27 males and 29 females with an alcoholic father and 35 males and 34 females from non-alcoholic families) constituted the sample used for the current study. Five subjects with 32 33 an alcoholic father also had an alcoholic mother. Procedures As indicated above, the study involved two phases Of data collection. In the initial stages Of data collection, subjects in Phase I attended one two-and-a-half hour testing session during which they completed an extensive battery of questionnaires assessing various aspects of their current relationship with their parents, parental conflict, self-image, drug and alcohol consumption practices and ego identity functioning. Additionally, subjects were asked to report on their parents' alcohol consumption. Subjects participating in Phase I Of the study were informed of the Opportunity for further participation and were asked to read the Informed Consent Agreement for Phase II Of the study before making a decision whether or not to continue. Those students who were interested in continuing were asked to complete a Contact Form; students indicated on this form their assigned code number written on their questionnaires from Phase I, a code name Of their choosing, and their phone number. Students were also informed that volunteering did not guarantee selection for participation in Phase II. Participants for Phase II were selected primarily on the basis of their reports of their parents' drinking behaviors. Half Of the Phase II participants were selected from among volunteers reporting in Phase I that their father had a serious drinking problem during their adolescence, and the other half were selected from among those reporting no history Of a drinking problem in either parent. Subjects in the comparison group were selected such that their fathers' educational and occupational status, 34 based on Hollingshead's (1957) social index scale, were roughly similar to those included in the high risk group. During the latter stages Of data collection, procedures for Phase I were modified in that subjects were asked to complete a much smaller number Of questionnaires, including reports of their parents' drinking~ behaviors, in order to screen subjects for those reporting parental alcoholism. Those selected via this screening procedure were matched with participants from non-alcoholic families from similar SES backgrounds and both groups then completed the entire battery Of questionnaires after agreeing to continued participation in Phase II. All other procedures remained the same as those used in the initial stages Of data collection. Graduate student interviewers who had no knowledge of any Of the students' questionnaire responses contacted the potential interviewees by telephone (the interviewer asked for the student using the code name indicated on the Contact Form). During Phase II, subjects were asked to complete some additional questionnaires (not used in this study) and to participate in a Young Adult/Parent Relationships Interview assessing the experiences Of relatedness and autonomy in their relationships with each Of their parents. Ordering of the interviews was counter-balanced so that approximately one half Of the subjects first described their relationships with their mothers and half were first described their relationships with their fathers. Interviews lasted approximately three hours. Depending upon the scheduling needs Of the subjects, some subjects were interviewed about both parents during one (three hour) meeting (interrupted with a break between mother and father interviews) and others met on two separate occasions. Subjects received additional 35 research credit for participation in this second phase of the experiment. Measures Three constructs pertinent to this study were measured: parental alcoholism, marital conflict, and quality Of the late adolescent/parent relationships. Each measure can be found in the appendices. Parental Alcoholism. Subjects with an alcoholic father were identified via a student report measure, the Parent Alcohol Consumption Questionnaire (Appendix A), using criteria validated by O'Malley, Carey, and Maisto (1986). These investigators validated Schuckit's (1980) assumption that children's reports Of their parents' major alcohol- related problems (e.g., marital separation or divorce, loss Of employment, two or more arrests from drunk driving, harm to health, etc.) can be used tO indicate a family history Of alcoholism. In particular, O'Malley et al. compared reports Of parent drinking practices (i.e., frequency and quantity of consumption and problems caused by alcohol) provided by both young adults and their parents. These investigators were able to maximize true positive reporting Of parent alcoholism by defining a parent as alcoholic if the child reported that the parent had experienced at least one major alcohol- related problem. They further found that they could minimize false negative reporting if, in identifying non-alcoholic parents, they did not include parents who, according to the child, had experienced minor (but not major) alcohol-related problems (e.g., economic distress, shame of family, accidents, etc.). Children's reports of the quantity and frequency of their parents' drinking were less reliable. 36 Accordingly, in this study children with an alcoholic parent were identified as those who reported that their father has experienced one or more major alcohol-related problem. Subjects were included in the comparison group if they reported that neither parent has experienced either major or minor consequences Of parental drinking. There were two exceptions to this inclusion criterion. First, some subjects indicated that their fathers had experienced harm to health because Of their drinking (regarded by O'Malley et al. as a major consequence), and also reported "heaviest levels Of quantity and frequency of alcohol consumption, current or past" on the part Of their father which did not reflect excessive drinking. One possible explanation is that a pre-existing health condition was exacerbated by alcohol consumption so that the parent refrained from drinking (even though he was drinking in non-excessive amounts). There were also some subjects who expressed by writing a note on the questionnaires their personal beliefs that alcohol consumption is, in general, an unhealthy practice and hence reported that their parents had experienced harm to their health as a result Of their drinking, regardless of the amount of alcohol actually consumed by the parent. Including these subjects as children Of alcoholics could result in a number of false positives. To circumvent this possibility, subjects who reported that harm to health was the 2212_major consequence Of paternal drinking were included in the children Of alcoholics group only if the average frequency and quantity of drinking met criteria described in the literature for problem drinking (see Cahalan and Cisin, 1968). Problem drinking was defined as the consumption of five or more drinks on each Of one or more occasions 37 per week 23 three or more drinks nearly every day. Those who indicated harm to health as the only major consequence Of the father's drinking and did not report problem drinking were not included in the sample for the study. Second, subjects with an alcoholic father were included in the children of alcoholics group only if they reported that their fathers were actively drinking when the subject was fifteen years Old or Older (i.e., subjects whose alcoholic fathers stopped drinking before the subject was fifteen year Old are excluded from the experimental and control groups). Social scientists frequently divide the periods of early and middle adolescence at age 14- or lS-years; this division, while somewhat arbitrary, corresponds tO the age at which adolescents leave middle school and enter high school. Moreover, this inclusion criterion ensured that problem drinking was present during at least part Of the subject's adolescent years, and hence during a critical period Of autonomy develOpment. Recent research (Wright, Frank 8 Pirsch, in progress) has indicated that paternal alcoholism that continues past an adolescent Offspring's middle adolescence (i.e., the 14th birthday) has consequences for both the mother/adolescent and father/adolescent relationship which are different from the consequences Of paternal alcoholism which ceased prior to adolescent's middle adolescence. In this study we controlled for this factor by only examining adolescent children Of alcoholics whose father's continued to drink beyond age 14. Parent Conflict. A 14 item Positive Parent Conflict Resolution §gglg_(Frank and Burke, 1992) assessed the late adolescents' perceptions Of their parents' ability to resolve marital conflicts ("My parents are able to resolve disagreements fairly quickly") without involving the 38 adolescent in the parents' disputes ("My father tries to get me to side with him when he fights with my mother", negatively scored); alpha'.86. A capy Of this measure can be found in Appendix B. A validity study on a sample Of 40 undergraduates and their mothers and fathers indicated that the PPCR correlated with parents' report Of marital functioning on the widely used Dyadic Adjustment Scale (Spanier 8 Cole, 1974);.£_for the correlation between students' scores on the PPCR and scores averaged across mothers and fathers on the Dyadic Adjustment Scale was .80. In the same study, the correlation between students' reports on the PPCR and parents' scores on a General Parenting Alliance Scale (Frank, Jacobson, and Hole, 1988) describing parents' ability to work together as parents, was .72. Frank and Burke (1992) found that late adolescents whose parents had divorced within the past five years had significantly lower scores on the PPCR than late adolescents from intact families; also, adolescents describing poor parent conflict resolution styles reported more autonomy and less relatedness in their relationships with their parents. Adolescents' Autonomy and Relatedness in_Relationship with Parents. Subjects' experiences of autonomy and relatedness in their relationship with each parent was assessed via the Young_Adult/Parent Relationship Interview (Frank, Avery, and Laman, 1988). This interview (found in Appendix C) was successfully used by its authors to describe three dimensions Of 22- to 32-year-old young adults' relationships with their mothers and fathers: connectedness, competence, and emotional autonomy. Subjects were asked questions concerning decision making, mutual help- giving and support, frequency Of contacts, conflicts between the 39 subjects' and parents' needs, emotional tensions, communication, feelings Of closeness and concern, and subjects' evaluations Of the parents' strengths and weaknesses (see Appendix C). These responses were then coded on a 5-point scale on ten scales (1 and 5 are indicative of low and high scores, respectively): closeness, communication, concern, empathy, respect, self-Other responsibility, personal control, self-assertion, independence, and decision making. Appendix D contains scoring criteria for each of the ten scales. Two scales assessed issues related to competence: decision making and independence. Scores on decision making reflected adolescents' ability to make important decisions and life choices without undue influence from the parent and without feelings Of anxiety or shame when their own values and preferences do not conform to those Of parents. Ratings Of independence assessed adolescents' perceived ability to cOpe with challenges and difficulties without the help Of the parent. Three other scales assessed emotional autonomy: personal control, self-assertion, and respect. Personal control scores reflected the adolescents' ability tO control his or her behavior and feelings in the relationship with the parent such that he or she was not constantly on guard against or overwhelmed by feelings of rage, fear, or dependency. Scores on the self-assertion scale assessed the extent to which the adolescents considered themselves to be the best evaluators Of their self-worth or, in contrast, monitored their behavior in order to avoid the parent's disapproval, anger, or retribution. Finally, scores on the respect scale described adolescents' evaluation of the parent and included their perceptions of the parent's competence and suitability as a positive role model. 40 Five additional scales assessed connectedness vs. separateness in relation to parents. Self-other responsibility (separateness) described how adolescents resolve conflicts and pulls between Obligations and needs created by their own life circumstances and the needs Of and Obligations tO the parent. The closeness scale reflected adolescents' perceptions Of the degree of emotional connectedness versus estrangement and distance in their relationships with the parent. Scores on the communication scale assessed the degree of Openness (both depth and breadth) in the communication between parent and adolescent. The concern scale assessed the extent tO which adolescents showed evidence of concern about the parent's well-being and interest in the parent's needs and feelings. Finally, the empathy scale reflected adolescents' ability to understand the parent as a complex person and to appreciate the parent's feelings and perspectives, even when these differed from those Of the adolescent. In order to maximize the independence Of the issue ratings, responses pertaining to each Of the autonomy and relatedness issues were typed on separate sheets of paper. Because a number of responses pertained tO more than one issue, some overlap in the content Of the protocols was both expected and observed. Coding Of the late adolescents' interview responses was completed by four clinical psychology graduate students. Although these coders were not told Of the drinking status of subjects' parents, it should be noted that subjects Often (but not in every instance) referred to their parent's problem drinking during the course Of the interview. Hence, in many instances raters were aware Of the presence of parental alcoholism. 41 Training of coders was done using interview responses from the Frank et al. (1988) study that had already been coded by raters not involved in the current study. These protocols were used as the standard for bringing raters up to stands for adequate interrater reliability. reliability of coding. Because the ages of the subjects included in the current sample differed from those Of the previous study, further reliability estimates were Obtained on a subsample Of 20 protocols from each Of the relationship issues collected as part Of the present study. Twenty interviews from the current study were coded by the senior author Of the earlier study (S. Frank; Frank et al., 1988) and then re-coded by graduate student raters to determine interrater reliability. Because interrater reliability was adequate (See Table l), the graduate students coded the remaining protocols for each issue to which they were assigned. Two graduate students coded two dimensions each, and two additional graduate students coded three dimensions each. Spot checks of reliability by S. Frank on randomly selected protocols helped to avoid rater drift. 42 Table 1. Reliability estimates (Pearson correlation coefficients) for each Of the parent/adolescent relationship issues Decision-making .87 Independence .80 Personal Control .89 Self-assertion .88 Self-other responsibility .89 Closeness .88 Communication .87 Concern .92 Empathy .97 Respect .83 Chapter 5 Results Test pf Hypothesis 1 PPCR scores were subjected to a 2 (ACOA Group) X 2 (Sex) analysis of variance (ANOVA). A significant main effect [2K1,121)'8.71, pf.001] for ACOA group indicated that, as expected, children Of alcoholic fathers reported significantly less positive parent conflict resolution (Mr 2.80) than did the comparison subjects (M?3.33). Neither the main effect for Sex nor the two-way interaction effect were significant. Igp£p_pf_hypothesized differences ip_Emotional Autonomy, Self-reliance, and Connectedness between ACOAs and Comparison Subjects Factor analyses reduced the 10 relationship issue scores to a smaller number Of relationship dimensions in order to test hypothesized differences in the late adolescent/parent relationship. Univariate correlations of the father/adolescent and mother/adolescent relationship issues are presented in Tables 2 and 3, respectively. As can be seen, separate correlation matrices for each group (ACOA and COMP) reveal that, for the most part, correlations among the relationship issues were similar for adolescents from alcoholic and non-alcoholic families. Additionally, in several cases where the correlations were significantly different, differences were in the magnitude rather than the direction Of the correlations. However, three exceptions are worth noting. 43 Table 2. 44 Pearson correlation coefficients for the father/adolescent relationship issues, by Total sample, ACOA and Comparison (COMP) groups DM ACOA COMP IN ACOA COMP PC ACOA COMP SA ACOA COMP SO ACOA COMP CL ACOA COMP ACOA COMP CN ACOA COMP ACOA COMP Note: Note: DM a= pf.05; IN PC .46b -.32b .49b -.26 .38a -.22 -- -.09 -- -.21 -- .18 bfp$.01 SA .13 .25 .10 .37b .35b .48b .37b .25 .46b - SO CL CM CN -.37b -.38b -.04 -.32a -.42b -.02 -.29a -.25a -.06 -.27b -.17 .00 -.53b -.46b -.21 .10 .19 .22 .55b .54b .20a .41b .44b .18 .55b .50b .26a .26b .26b .27b .01 -003 .27a .45b .45b .27a -.33b -.28b -.30b -.47b -.31a -.45b -.25a -.33b -.14 -- .74b .47b -- .75b .48b -- .70b .51b -- .43b -- .48b -- .45b groups are significantly different (p$.05, Fisher's Z transformation). EM .22a .16 .25a .22a .14 .30a .07 .03 .16 .12 .22 .04 .12 .18 .04 .08 .05 .15 .09 .04 .17 .29b .30a .28a Boldface type indicates that correlations for ACOA and COMP RS -.47b -.46b -.35b -.28b -.40b .00 .68b .52b .70b .20a -.04 .40b -.05 -.10 -.06 .60b .50b .59b .55b .41b .56b .22a .19 .30a -.03 .06 -.08 Table 3. 45 Pearson correlation coefficients for the mother/adolescent relationship issues, by Total Sample, ACOA and Comparison (COMP) groups DM ACOA COMP IN ACOA COMP PC ACOA COMP SA ACOA COMP SO ACOA COMP CL ACOA COMP ACOA COMP CN ACOA COMP ACOA COMP Note: Note: DM IN PC SA SO CL CM CN -- .54b -.15 .05 .16 -.23a -.33b -.23b -- .62b -.25 .10 .22 -.45b -.52b -.30a -- .48b -.02 .08 .22 -.11 -.15 -.l6a -- -.01 .11 .01 -.12 -.O7 .09 -- -.19 .17 .03 -.31a -.28a -.17 -- .24a .09 .03 .04 .13 .40b -- .55b .08 .44b .47b .24b -- .52b -.05 .55b .50b .29a -- .53b .11 .43b .42b .18 -- .06 .25b .38b .16 -- .10 .26 .32a .23 -- -.09 .32b .38b .08 -- -.32b -.24b -.21a -- -024 -0318 -026 -- -.35b -.28a -.19 -- .58b .37b -- o63b .531) -- .62b .20 -- .43b -- .51b -- .36b a=lp§.05; bfp$.01 EM RS .04 -.32b -003 -043b .06 -.15 .34b -.21a .22 -.33a .44b -.04 .11 .56b .11 .68b .17 .35b .01 .34b .10 .32a .02 .32b .09 -.10 .08 -.17 .04 -.11 .16 .46b .11 .65b .17 .31b .16 .51b .04 .56b .28a .44b .38b .27b .31a .39b .47b .08 -- -.05 -- .02 -- -.08 Boldface type indicates that correlations for ACOA and COMP groups are significantly different (p§.05, Fisher's Z transformation). 46 First, for ACOAs, Independence was negatively associated with various aspects of connection in relation to mothers and fathers, whereas for comparison subjects Independence was positively, but for the most part not significantly, related to connection with parents. Secondly, negative correlations between late adolescents' perceptions Of Decision Making and their experiences Of Closeness, Communication and Concern in relation tO mothers were stronger and more Often statistically significant for ACOAs than for comparison subjects. Thirdly and finally, positive associations between Self-assertion in relation to fathers and various aspects of connection typically were stronger and more Often statistically significant for comparison subjects compared to ACOAs. Overall, the correlations among the relationship issues for ACOAs and comparison subjects were more similar than different. Hence, data for the adolescent/parent relationship issues for both groups combined were subjected to a principal component factor analysis using a varimax solution. Factor solutions are presented in Table 4. These analyses yielded factors which were quite similar to those found in the Frank et al. (1988) study of young adults. As in the earlier study, one factor described an emotional autonomy dimension (i.e., positive loadings for Personal Control, Self-Assertion, Closeness, Communication, and Respect). The emotional autonomy factor derived by Frank et al. also included significant loadings on Personal Control, Self-Assertion and Respect; in addition, in the young adult sample Closeness and Communication contributed to the emotional autonomy factor in relationships with fathers (but not mothers). However, this factor solution differed from that reported by Frank et al. in that for these 47 Table 4. Factor loadings for the parent/adolescent relationship issues on the three relationship factors. Connectedness vs. Emotional Separateness Self-reliance Autonomy Mothers Fathers Mothers Fathers Mothers Fathers Decision making -.30 .08 .76 .60 -.ll -.53 Independence .12 -.09 .76 .82 -.00 -.19 Personal control .05 -.05 .01 .13 .86 .86 Self-assertion -.11 .03 .20 .69 .78 .45 Self-other responsibility -.63 -.77 .19 .23 .14 .02 Closeness .53 .46 -.21 '-.13 .52 .75 Communication .50 .41 -.19 -.07 .63 .74 Concern .71 .70 .06 .18 .27 .31 Empathy .61 .50 .49 .43 .01 -.05 Respect .16 -.01 -.35 -.16 .71 .85 Z variance 17.5 13.4 14.4 18.8 32.7 35.6 Note: Factor loadings Of or greater than .40 are in boldface. 48 adolescents, emotional autonomy from fathers included a sizable negative loading for Decision Making. The emotional autonomy factor accounted for the greatest percent Of the variance for adolescents' relationships with both mothers (32.7 percent) and fathers (35.6 percent). A second factor described a self-reliance dimension (comparable to what Frank et al. referred tO as a Competence factor). As reported by Frank et al. (1988), sizable positive loadings on both Decision Making and Independence characterized this dimension. However, in this adolescent sample (but not in the Frank et al. young adult sample), the self-reliance factor also included positive loadings for Empathy (for both mothers and fathers), as well as Self-Assertion (for fathers only). This factor accounted for 14.4 percent Of the variance for relationships with mothers and 18.8 percent of the variance for relationships with fathers. Finally, as in the Frank et al. study, a third factor described a connectedness vs. separateness dimension (with positive loadings for Closeness, Communication, Concern and Empathy, and a negative loading for Self-Other Responsibility) for both mothers and fathers. This factor accounted for 17.5 percent of the variance for adolescents' relationships with their mothers and 13.4 percent of the variance with fathers. Between.group differences. Regression analyses performed on each Of the three relationship dimensions for mothers and fathers were used to test the remaining hypotheses, i.e., that a)ACOAs would exhibit greater self-reliance, but lower emotional autonomy and connectedness than comparison subjects, and b)these group differences would remain even after controlling for PPCR. Simple t-tests indicated that, 49 compared to the comparison subjects, ACOAs described less emotional autonomy in relation to both mothers (p§.01) and fathers (pf.001) and more connectedness in relation to mothers (pS.05). Regression analyses assessed whether these differences would still be significant after controlling for PPCR. ACOA group, PPCR, and Sex were simultaneously entered on the first "step" Of the regression analyses. In addition, all possible two-way interactions, as well as the three-way interaction between Sex, PPCR and ACOA group were entered on the second and third steps of these analyses tO explore the possibility that PPCR and/or Sex would moderate ACOA group differences in late adolescent/parent relationships. Self-reliance. Perceptions Of positive parent conflict resolution (PPCR) were negatively related to self-reliance in relation to fathers [b=-.20, pf.05]; i.e., when parents were perceived as having more difficulty resolving conflict, adolescents reported more self-reliance in relation to fathers. Neither the main effect for ACOA group nor Sex were statistically significant. However, there was a significant two- way (ACOA group X PPCR) interaction effect for self-reliance in relation to fathers [b'-.22,‘p§.051. TO understand this interaction, subjects were assigned to a Parent Conflict Resolution (Positive vs. Negative; PCR) group: subjects reporting PPCR scores at or above the sample mean (Mf3.09) were included in the Positive PCR group, whereas those reporting PPCR scores below the sample mean were included in the Negative PCR group. Scores on the self-reliance factor were broken down by ACOA group and PCR group. As can be seen in Table 5, differences in ACOA and comparison subjects' reports Of self-reliance were marginally 50 Table 5. Factor scores on the Self-reliance with Father dimension (broken down by ACOA Group and PCR Group) Parent Conflict Resolution Group Negative Positive T-Test ACOA .27 (n=4l) -.30 (n=15) t(1,54)'1.82, p§.074 COMP -.26 (n-18) -.04 (n=51) NS T-test t(1,57)=1.81,.p$.076 NS 51 significant only under the condition of negative parent conflict resolution. In other words, when parents were described as unable to adequately resolve their marital conflicts, ACOAs tended to report more self-reliance in relation to fathers than did comparison subjects. In addition, ACOAs describing negative parent conflict resolution tended to report greater self-reliance in relation to fathers than ACOAs reporting positive parent conflict resolution. In contrast to the findings for fathers, neither the main effects nor any Of the interaction effects accounted for significant variance in self-reliance in relation to mothers. Emotional autonomy. PPCR was positively related to emotional autonomy in relation to fathers [b=.52, pf.001] and mothers [b'.34, 'p5.001]. In addition, and even after controlling for perceptions of parent marital conflict, adolescent children Of alcoholics reported less emotional autonomy from their fathers than did the comparison adolescents [b--.20, p§.051. There also was a two-way (ACOA group X PPCR) interaction effect for Emotional Autonomy in relation to fathers, [b=-.19, pf.05]. This interaction was tested, as before, by assigning subjects to Positive and Negative PCR groups. As can be seen in Table 6, differences between ACOA and comparison subjects in emotional autonomy were significant only under conditions of positive parent marital conflict resolution. Comparison subjects reporting positive PPCR also reported the highest level of emotional autonomy (LSD tests indicated that emotional autonomy for that group was higher than for any other group). However, ACOAs describing positive PPCR described higher emotional autonomy than ACOAs describing negative PPCR. Neither the main effects nor the interaction effects were 52 Table 6. Factor scores on the Emotional autonomy with Father dimension (broken down by ACOA Group and PCR Group) Parent Conflict Resolution Group Negative Positive T-Test ACOA -.64 (n-41) -.13 (n-15) c(1,54)-2.07, 25.05 COMP -.31 (n-18) .67 (n-Sl) t(l,67)-4.29,‘p§.001 T-test NS t(1,64)=-3.45,p§.001 53 significant for emotional autonomy in relation to mothers. As noted above, however, ACOAs did report lower emotional autonomy, before controlling for PPCR. Connectedness. It was hypothesized that ACOAs would report high concern and empathy, but low communication, closeness, and self-Other responsibility in relation to their parents. The implication was that the connectedness factor would not hold up for ACOAs. However, as noted previously (Tables 2 and 3), intercorrelations among the issues defining this factor in Frank et al.'s (1988) work (i.e., communication, closeness, concern, empathy and self-Other responsibility) were similar for ACOAs and comparison subjects. Hence, overall differences were examined by looking at data for the factor. The only significant main effect for connectedness was for Sex: females reported greater connectedness in relation to both fathers [b=.19, p§.05] and mothers [b=.33,‘p$.001]. However, a three-way (ACOA group X PPCR X Sex) interaction effect also was significant for connectedness in relation to mothers [b--.28,'p$.Ol]. Table 7 presents the connectedness scores for mothers, broken down by Sex, ACOA group and PCR group. As can be seen in Table 7, group differences were significant for males, but not for females. Differences in connectedness scores between ACOA and comparison subjects were significant only under the condition of positive parent conflict resolution, with comparison males reporting positive parent conflict resolution describing significantly less connectedness with mothers than ACOA males reporting positive parent conflict resolution. The pattern of scores for females was quite different (with the lowest connectedness scores reported by comparison females who perceived parents as engaged 54 Table 7. Factor scores on the Connectedness with Mother dimension (broken down by Sex, ACOA Group, and PCR Group) Parent Conflict Resolution Group Negative Positive T-Test Males ACOA -.24 (n=18) .16 (n=9) NS COMP -.12 (n-13) -.74 (n=22) (1,33)=2.11 (25.05) T-test NS t(1,29)=2.73, p§.01 Females ACOA .45 (n=23) .75 (n=6) NS COMP -.38 (n-S) .27 (n-29) NS T-test NS NS 55 in excessive conflict). However, these differences did not reach statistical significance. Additional Analyses In order to gain a more holistic understanding of adolescents' relationships with their parents (and the impact of paternal alcoholism and parental conflict on those relationships), factor scores were used in a cluster analysis to further describe different types of parent/adolescent relationships. Cluster analysis yielded six "relationship categories" which were similar for mothers and fathers, and also were relatively similar to the relationship categories described in the Frank et al. (1988) young adult study. Table 8 provides the mean factor scores for the relationship dimensions within each of the relationship categories. All but two subjects were able to be assigned to one of these six categories. A Pseudoautonomous category was defined by low scores on the Emotional Autonomy and Connectedness dimensions, and high scores on the Self-reliance dimension. Subjects who described Role Reversed relationships with their parents reported low scores on the Emotional Autonomy dimension, but high scores on Self-reliance and Connectedness. Dependent adolescents reported low scores on all three dimensions, although scores on the Connectedness factor were less discriminating than scores on the other two factors. Adolescents who described their relationships with parents as Identified reported high scores on both Connectedness scores and Emotional Autonomy, and low scores on the Self- reliance dimension. An Individuating/In Progress category was defined by somewhat low scores on both the Self-reliance and Connectedness dimensions, but moderately high scores on the Emotional Autonomy factor. 56 Table 8. Mean scores of the relationship factors for adolescents in each of the relationship categories Emotional Self- Category, (n) Autonomy Reliance Connectedness M SD M SD M SD Pseudoautonomous Mothers (4) -2.26 .22 .82 .75 -1.26 .65 Fathers (15) -.31 .55 1.31 .15 -.55 .59 Role Reversed Mothers (10) -1.44 .34 .51 .60 .31 .50 Fathers (8) -l.66 .41 1.19 .71 1.05 .64 Dependent Mothers (13) -.79 .66 -l.45 .43 .01 .85 Fathers (30) -.76 .55 -.60 .68 -.66 .73 Identified Mothers (31) .44 .62 .11 .61 1.08 .47 Fathers (13) .32 .71 -.93 .83 1.46 .61 Individuating/In Progress Mothers (54) .45 .78 -.19 .79 -.69 .61 Fathers (51) .62 .60 -.14 .53 -.01 .87 Individuated Mothers (ll) “.11 .48 1.73 .57 .01 .49 Fathers (6) 1.53 .45 1.72 .50 .02 .83 57 Finally, subjects who reported Individuated relationships with parents reported, for the most part, high Emotional Autonomy and Competence scores, with only moderate Connectedness scores. Descriptions pf the Relationship Categories Interview responses by subjects in each of the relationship categories were re-read in order to provide a greater understanding of the nature Of the relationships between adolescents and their parents. Pseudoautonomous. Mean factor scores on each Of the three relationship dimensions were very similar to those found for a corresponding relationship type in the young adult study. Specifically, these adolescents described themselves as self-sufficient and insisted that they could take care of themselves. However, this apparent self- reliance was coupled with intrusive feelings Of guilt, anger or resentment about accepting any parental assistance and these adolescents continued to rely heavily on their parents as evaluators of their self- worth. They felt distanced from their parents, yet this distance seemed "forced" rather than the result Of individuation. She feels a need for me to be perfect and I have a need to live my life even it's imperfect. I try to keep my life separate from hers, my private life, my life outside the house. I can't say "get out of my hair" because that wouldn't be nice; she has a right to worry. ... But she thinks I'm going to be a loser. And that makes me pretty mad. Role reversed. Like the adolescents in the pseudoautonomous group, those in the role reversed group described themselves as self-sufficient and competent to face life's challenges without unduly relying on their parents for support, yet continued to experience feelings of personal inadequacies or emotional discomfort when with the parent. Despite significant feelings Of discomfort, however, these 58 adolescent continued to be very involved in their parent's life. In fact, their connectedness with the parent typically took place in a reversal of roles, with the adolescent frequently expressing a perceived need to take care of their parent. Although this relationship type corresponded rather directly to the Competent/connected group from the young adult study, pervasive role reversals appeared particularly salient for these adolescents and hence this is stressed in the cluster name. Additionally, their profound sense of needing to care for their parent appeared to preclude their ability to make gains in emotional autonomy. I stick up for him when anybody says anything bad about him. I Offer to make him lunch when I'm there. I tried to make sure he had something to eat.. He knows we're suffering with him and that helps him know we're concerned. That's why he doesn't commit suicide.. plus us two kids, we're his bond to life. About his drinking ... I say stop, he says no. For a long time I tried to be as understanding as I could and remain calm and keep my faith in him but now it's a lost cause. I almost just bitch at him, like a coach would scream at his players. I'm surprised I haven't gone looney in those situations. I'm proud of my response. Dependent. This relationship type corresponded closely to that seen in the young adult sample. Dependent adolescents relied on their parent for cOping with major life decisions and events. However, they frequently projected their dependency feelings onto the parent and described the parent's considerable involvement in their lives as typical, rather than excessive ("she wants to take care of me like any mother would"). They also relied on their parent as an evaluator Of their self-worth and Often avoided the parent's anger and disapproval by "not making waves.‘ As with the young adult sample, the connectedness 59 factor did not help discriminate well on this relationship type, especially for mothers. While some subjects described estranged relationships with their mothers, others described extremely close, almost enmeshed, relationships. With fathers, this kind of dependence Often occurred in the context of authoritarian fathering in which the adolescents perceived their dependence as "forced" upon them. I've talked over moving out Of the house. I told her I wanted to move out and she asked me was I ready. I thought I was ready but she talked me out Of it. She just always knows what I need. Always. Always. I think it's totally fine. I don't think I deserve it, to tell the truth. So I'm really satisfied. I don't give her anything like that, so I just feel like I'm taking things from her. He told me I was going to MSU and every time I talked about a different school he said it wasn't any good. I got accepted, it was the only school I applied to. He'd always implanted that I was going to MSU. He told me. I like MSU because it's a Big 10 school and it's better than the other ones. Identified. These adolescents described very Open and close relationships with their parent and they accepted their parent's values and Opinions. Although they continued to be quite dependent on their parent, they experienced this dependence as untroubling. In fact, they described relatively few tensions with their parent, felt little psychological discomfort when.with the parent, and felt free to express themselves openly. They typically described their parent in a very positive light, describing their parent as a good role model. I enjoy just about everything. I really admire my father. A lot of being with him is almost a learning experience because I basically feel that what he says is mostly right. He has a lot to Offer in terms Of business related activities and investments and stuff. He does anything he can for me. He gives to me financially ... he gives to me emotionally. I like it 6O alot. I'm comfortable that I can ask for his advice, or his money if I want to buy something, or whatever. Mean factor scores for the Self-reliance dimension were higher for subjects who were "identified" with their mothers compared to those "identified" with their fathers. It appeared that adolescents in this category remained dependent on their mothers in different ways: they relied on mothers for more practical support (e.g., laundry, care packages, etc.), whereas reliance on fathers Often occurred in the context of advice or provision Of material needs. These adolescents also described relying on their mothers for help in the process of becoming more independent. She gives me money, help me make decisions by sitting down and talking things through; helps me come to my own conclusions. She'd never tell me what to do or think, but she'll help me solidify my own ideas by letting me try them out on her first. Individuating/In Progress. Unlike any Of the relationship types in the young adult study, this category described a develOpmentally relevant and apprOpriate movement toward individuation. Specifically, individuating adolescents continued to rely on their parent for help with value decisions, coping with life's challenges, etc., yet simultaneously had a sense of themselves as making gains in both separateness and emotional autonomy. In fact, many Openly discussed that they are in the process of "growing up" or becoming more independent and that they experienced their parents as generally accepting Of this process. I've grown up. I don't have to worry about him scolding me. We have an adult relationship. Man-to-man more than father and son. I became my own person. They let me fly the cOOp. They let me go. They support the moves I made and if they 61 didn't support it they let it happen. [Mom] will listen, she'll send money. The best thing, though, is letting me do what I want. It's my life, let me learn from my mistakes myself. Different things I've experienced such as jobs, friends, being on my own. They gave me more of a common ground with him and made me realize that he was actually right about a couple things. It made me put more value on his advice where before it went in one ear and out the other. Individuated. Individuated adolescents (like the individuated young adults described by Frank et al.) felt ready to face the challenges of their lives without relying excessively on their parent. They maintained rather clear boundaries between their own and their parent's lives, yet enjoyed the company of their parent and experienced few tensions or conflicts when with them. Adolescents who described individuated relationships with their fathers described their relationship in much the same way as did the individuated young adults. I have enough space and room to be growing up and be by myself, or be with Scott. I wouldn't not want to see him, he's really important to me, to my life. I really value him. But as I grow up, other people come into my life and it stops being just you and your parents ... you have friends from college, from work, boyfriends, people that you want to have room in your life for,and that means your parents may have to move over a little. In relationships with mothers, adolescents in the individuated group reported similarly high levels of self-reliance within the context of moderate closeness, but were slightly more vulnerable to their mother's disapproval and influence. This vulnerability to mother's disapproval was often manifest in conflicts surrounding the adolescent's impending independence and these subjects often described their mothers as having difficulty "letting go." I'm relying on them less now and I like that. I feel more responsible, more adult. Like I'm supporting myself 62 some, and that I can earn and spend my money however I want. And that I'm helping them out, that feels good, too. And she treats us more like grown ups than she used to, because we're mostly adults now. But I think she gets lonely not having anybody around much. She can't let go of her kids. Analyses Using the Relationship Categpries Chi Square analyses revealed no sex differences in either relationship categories with fathers [X2(5,N-123=1.84, n.s.] or in relationship categories with mothers [X2(5,N-123)'8.68, n.s]. In addition, when broken down by Sex and ACOA group, sex difference continued to be unremarkable. ACOA Group status, however, was significantly associated with relationships types for fathers [X2(5,N-123), pf.001] and mothers [X2(5,N-123),'p§.05]. Table 9 presents the number of subjects in each of the relationship types, broken down by ACOA group (n's for males and females are presented, but not included in the chi square analyses). Inspection of the cell frequencies in Table 9 indicates that when clusters were defined by the data for relationships with fathers, adolescent children of alcoholic fathers were overrepresented in the role reversed, dependent, identified and (to a lesser degree) pseudoautonomous categories. Conversely, comparison adolescents were overrepresented in the Individuating and Individuated categories. When describing their relationships with their mothers, subjects in the ACOA group were also more pseudoautonomous and role reversed and comparison subjects were more likely (than ACOAs) to be individuating. Not too surprisingly, additional analyses demonstrated differences among the relationship categories in perceptions of parent marital conflict. Table 10 presents the means and standard deviations of PPCR 63 Table 9. Numbers of ACOA and Comparison (COMP) subjects in each of the relationship categories Mothers Fathers Category ACOA COMP ACOA COMP Pseudoautonomous 4 0 9 6 Column 1 7.4 0.0 16.1 9.0 Males(n,2) 4(14.8) 0(00.0) 5(18.5) 4(12.l) Females 0(00.0) 0(00.0) 4(13.8) 2( 5.9) Role Reversed 7 3 8 0 Column 2 13.0 4.3 14.3 0.0 Males 3(11.1) 2( 5.7) 4(14.8) 0(00.0) Females 4(14.8) 1( 2.9) 4(13.8) 0(00.0) Dependent 5 8 18 12 Column 2 9.3 11.6 32.1 17.9 Males 2( 7.4) 5(14.3) 9(33.3) 7(21.2) Females 3(11.1) 3( 8.8) 9(31.0) 5(14.7) Identified 14 17 9 4 Column 2 25.9 24.6 16.1 6.0 Males 6(22.2) 5(14.3) 4(14.8) 1( 3.0) Females 8(29.6) 12(35.3) 5(17.2) 3( 8.8) Individuating/In Progress 18 36 11 40 Column 1 33.3 52.2 19.6 59.7 Males 9(33.3) 22(62.9) 4(14.8) 19(57.6) Females 9(33.3) 14(41.2) 7(24.l) 21(61.8) Individuated 6 5 1 5 Column 2 11.1 7.2 1.8 7.5 Males 3(11.1) 1( 2.9) 1( 3.7) 2( 6.1) Females 3(11.1) 4(11.8) 0(00.0) 3( 8.8) 2 X (5,N=123) 11.01, p$.05 30.14,p,001 (Total Sample) 64 Table 10. Mean scores of Positive Parent Conflict Resolution for adolescents in each of the relationship categories Category Mothers Fathers E .82 :1 e2 Pseudoautonomous 2.48 (.32) 2.92 (.64) Role Reversed 2.59 (.59) 2.38 (.60) Dependent 3.04 (.37) 2.90 (.41) Identified 3.16 (.50) 3.19 (.37) Individuating/In Progress 3.25 (.43) 3.35 (.39) Individuated 3.09 (.58) 3.24 (.60) £6,122) 4.99, 95.001 9.03, p<.001 Note: Mothers: Role Reversed and Pseudoautonomous are significantly different from all others (p$.05), but not from each other. Fathers: Role reversed is different from all others (25.05); Individuating is different from Pseudoautonomous and Dependent (p§.05). 65 scores for subjects in each of the relationship categories. Significant between group differences were found for both fathers LE(5,122)-9.03, pf.001] and mothers LEKS,122)'4.99,‘p5.05]. A least significant differences (LSD) test indicated that adolescents who experienced their relationships with their fathers as role reversed reported lower PPCR scores than did subjects in all other groups (p$.05). Additionally, those who were individuating from their fathers reported significantly better parent conflict resolution than those in the pseudoautonomous and dependent groups (p§.05). Slightly different findings for parent conflict were noted when adolescents reported on their relationships with their mothers. Specifically, adolescents who described relationships with their mothers as either pseudoautonomous or role reversed reported lower PPCR scores than did adolescents in any of the other groups (p§.05). PPCR scores did not differ significantly between the pseudoautonomous and role reversed group. Given the clear importance of parent conflict as a moderator of ACOA versus comparison differences in earlier analyses, final analyses focused on attempting to understand the role of PPCR in moderating associations between ACOA group and the relationship categories. First, the proportion of ACOA subjects in each category was used as the dependent variable in an analysis of covariance, controlling for PPCR. Results indicated that after controlling for PPCR there continued to be significant differences in the proportion of ACOAs versus comparison subjects in the relationship categories for fathers [2(5,16)'3.16, ‘p§.011, but not for mothers. Chi square analyses were then repeated as described above, controlling for parent conflict by assigning subjects to Positive and 66 Negative parent conflict resolution (PCR) groups as before. Results indicated that, even when controlling for PCR group status, there were significant ACOA Group differences in adolescents' experiences of their relationships with their fathers (Table 11), although these were somewhat stronger in the Positive PCR group [X2(5,N-66)-13.04, pf.05] than in the Negative PCR group [X2(5,N=57)'9.86, p§.08]. However, no differences were noted in the relationships with mothers. The clearest distinction between ACOAs and comparisons in the negative parent conflict resolution group were in the role reversed and identified categories: all subjects in those father categories were children of alcoholics. When subjects reported positive parent conflict resolution, comparison subjects were more likely than ACOAs to describe either individuating/in progress or individuated relationships with their fathers. Thus, very meaningful between group differences were found when comparing relationships between adolescents and their alcoholic or non-alcoholic fathers, even when accounting for differences in parent conflict resolution scores. 67 Table 11. Numbers of ACOA and Comparison (COMP) subjects in each of the Father relationship categories, controlling for parent conflict resolution Negative Conflict Resolution Category ACOA COMP Pseudoautonomous 7 3 Column 2 17.1 18.8 Role Reversed 7 0 Column 2 17.1 0.3 Dependent 14 5 Column Z 34.1 31.3 Identified 5 0 Column 1 12.2 0.0 Individuating/In Progress 7 8 Column 2 17.1 50.0 Individuated 1 0 Column 2 2.4 0.0 2 x (5,N=57) 9.86, 25.08 Positive Conflict Resolution ACOA COMP 2 3 13.3 5.9 l 0 6.7 0.0 4 7 26.7 13.7 4 4 26.7 7.8 4 32 26.7 62.7 0 5 0.0 9.8 2 x (5,N=66) 13.04, 25.05 Chapter 6 Discussion The purpose of this study was to compare late adolescent children of alcoholic and non-alcoholic fathers in terms of their experiences of separation/individuation from their parents. It was hypothesized that adolescent children of alcoholics would experience self-reliance, emotional autonomy, and connectedness in relation to parents differently than adolescent children of non-alcoholics, and that these group differences would continue to be statistically significant even after controlling for predicted group differences in perceptions of parent marital conflict. Parent marital conflict Hypothesis 1 was that ACOAs would report lower positive parent conflict resolution (PPCR) and that PPCR would partly mediate the effects of paternal alcoholism on the parent/adolescent relationship. Results pertaining to parent conflict were straightforward: as expected, late adolescent children from alcoholic families reported significantly more parent marital conflict than did late adolescents from non-alcoholic families. This finding is consistent with extensive past research indicating that alcoholic couples experience high levels of conflict that they find difficult to resolve effectively (e.g., Jacob, 1986; Reider, 1991; Woodruff et al., 1972). Also as expected, 68 69 parent conflict was found to mediate drinking group effects on several aspects of the parent/adolescent relationship (detailed below). What was not anticipated was that parent marital conflict was also found to moderate some of the associations between paternal alcoholism and the parent/adolescent relationship. The implications of this finding are discussed in a later section. Differences i2 parent/adolescent relationships for ACOAs and Comparisons Self-reliance. Hypothesis 2 was that ACOAs would report greater self-reliance in relation to both parents than comparison subjects. Consistent with past research (Clair 8 Genest, 1987; Wright et al., 1990), ACOAs tended to report more self-directedness and less reliance on fathers for guidance and support than did late adolescents from non- alcoholic families. However, these group differences emerged only in the context of negative parent conflict resolution; i.e., only when parents were described as engaging in excessive conflict did ACOAs tend to report greater self-reliance in relation to fathers than comparison subjects. In fact, within group comparisons suggested that parent conflict had greater implications for ACOAs than for children from non- alcoholic families: ACOAs who reported negative parent conflict resolution styles also described more self-reliance in relation to fathers than ACOAs who described positive parent conflict resolution. In short, then, a high degree of self-reliance in relation to fathers appeared to be limited to a combination of paternal alcoholism and parent conflict rather than either one alone. Examination of items on the parent conflict measure used in this study can shed some light on this finding. Specifically, low scores on the PPCR scale used here 70 indicate that parents not only experience a high degree of conflict, but also denigrate the spouse and attempt to form a coalition with the adolescent child against the spouse. Presumably, paternal alcoholism increases the likelihood that adolescents become engaged in their parents' conflict, perhaps by "siding with" their mothers. Subsequently, they may perceive their fathers as not sufficiently competent to offer assistance and advice. It is interesting to note that, at least for this sample, poor parent conflict resolution was also insufficient, in and of itself, to lead to high self-reliance in relation to fathers. Past research (Frank 8 Burke, 1992) indicates that parent conflict only indirectly increases adolescents' self-reliance, through its impact on the deidealization process. In this study, only when parent conflict was combined with paternal alcoholism, was it positively related to self-reliance. Likewise, these data suggest that the sense of self-sufficiency described in ACOAs is not solely the result of having an alcoholic father, but rather becomes important when parent alcoholism co-occurs with a high degree of marital disharmony. Self-reliance in relation to mothers was unrelated to either paternal alcoholism or parent marital conflict. Emotional autonomy. Hypothesis 3 was that ACOAs would report lower emotional autonomy in relation to parents than comparison subjects. As predicted, ACOAs did describe lower emotional autonomy from fathers, but (in contrast to findings for self-reliance) these differences were significant only when adolescents described their parents' marriage as relatively conflict free. In particular, comparison subjects reporting parental harmony enjoyed a relatively high degree of emotional autonomy, whereas ACOAs who described a relatively harmonious parental marriage 71 also reported lower emotional autonomy in relation to their fathers. In contrast, when parents were engaged in excessive conflict, comparison subjects as well as ACOAs reported relatively conflicted relationships (i.e., low emotional autonomy) with their fathers. Seemingly, then, parental harmony somewhat attenuates the negative implications of having an alcoholic father, but parental marital dysfunction appeared to be directly related to conflictual parent/adolescent relationships, regardless of paternal alcoholism. Thus, parent marital conflict may be related to disruptions in the parent/adolescent relationship (at least in terms of adolescents' difficulties with self-assertion, and their feelings of rage, disappointment and disrespect toward the parent) which are similar to the relationship disruptions described by ACOAs. In contrast to the data for fathers, PPCR mediated (rather than moderated) the relationship between ACOA status and emotional autonomy in relation to mothers. In particular, although ACOAs did describe less emotional autonomy in relation to mothers than comparison subjects, group differences disappeared after controlling for PPCR. Connectedness. The final hypothesis was that ACOAs would report high empathy and concern for their parents, but low closeness, communication and self-other responsibility in relation to their parents. In contrast, it was predicted that comparison subjects would report relatively high scores on all five relatedness issues. Results did not support this hypothesis and instead indicated that the intercorrelations among these relatedness issues were similar for the ACOA and comparison groups. Hence, data for the two groups were factor 72 analyzed together, and scores on the resulting connectedness factor were used to test group differences. Data analyses pointed to drinking group differences in adolescents' perceptions of connectedness with mothers, but again these differences were observed only after taking into account contextual factors, i.e., both adolescent gender and parent conflict. Specifically, when parents were described as able to resolve conflicts quickly and in a positive manner, comparison sons described significantly more distance from their mothers than sons of alcoholics. Hence, in a more "favorable" family situation (i.e., one in which there was no parent alcoholism and the marital relationship was perceived as relatively harmonious), sons reported feeling relatively separate from their mothers. Interestingly, in all three of the other groups (i.e., ACOAs who reported both positive and negative parental conflict resolution, and comparisons who reported negative parental conflict resolution) sons reported relatively close relationships with mothers. Seemingly, then, either paternal alcoholism p£_marital conflict may make it more difficult for sons to separate from their mothers. Neither parent conflict nor paternal alcoholism impacted daughters' connectedness in relation to mothers. In contrast to the data for mothers, the only significant finding with regard to connectedness with fathers was that daughters reported a greater sense of closeness, warmth and connection to fathers than did sons. Similar gender differences in connectedness with parents (both mothers and fathers) have been well documented (Chodorow, 1978; Frank et al., 1988). Paternal alcoholism and parent marital conflict were unrelated to late adolescents' sense of connectedness with fathers. In general, results from analyses of the relationship dimensions 73 suggested that many ACOAs experience their relationships with parents in ways that are very similar to adolescents from non-alcoholic families, so that drinking group differences may be somewhat exaggerated in the clinical literature. Although in several instances these differences did emerge in the predicted directions, they were not generally evident until contextual factors (i.e., parent conflict or sex) were accounted for. Poor parent marital conflict resolution as perceived by the adolescent child was linked, by itself, to disruptions in the parent/adolescent relationship (i.e., low emotional autonomy from fathers and high connectedness in sons' relationships with mothers). What was not anticipated was that positive parent conflict resolution was also a moderating variable in that it attenuated the negative influences of paternal alcoholism on both emotional autonomy in relation to fathers and separateness in sons' relationships with mothers. The role of mediating and moderating variables in the observed effects of paternal alcoholism has been increasingly addressed by researchers in the ACOA field (e.g., Rogasch, Chassin 8 Sher, 1990; Zucker, 1992). Although the current study focused exclusively on parent marital conflict as the contextual variable which moderated (or mediated) differences in the parent/adolescent relationship, other contextual factors have been hypothesized as important moderators of the effects of alcoholism on family relationships. Most notably, many researchers have pointed to the importance of the comorbid psychopathology or personality traits of parents in alcoholic families in accounting for outcome risk in ACOAs (Rogasch et al., 1990; Zucker, 74 1992; Chassin et al., 1991). Co-existing psychOpathology in parents of subjects in this study were not ascertained and thus the "high risk" group here may, in fact, be quite heterogeneous in terms of parent characteristics. It is quite possible that certain parental characteristics (e.g., depressed, anti-social) which have been frequently associated with alcoholism (Zucker, 1987) may differentially impact parent/adolescent relationships. Moreover, the hypothesis that such personality traits are genetically transmitted to Offspring of alcoholics (Johnson, Sher 8 Rolf, 1991) suggests that parent/adolescent relationships may also differ as a function of the adolescent's own personality characteristics. Future studies should address such hypotheses by statistically controlling for parental psychopathology when observing the effects of alcoholism on offspring. ' The relationship catggories Analyses using the relationship categories offered a somewhat richer and more holistic picture of differences in parent/adolescent relationships between children of alcoholics and non-alcoholics. These analyses described patterns of self-reliance, emotional autonomy, and connectedness in participants' relationships with parents. Cluster analyses identified six relationship categories that were strikingly similar to those found in the Frank et al. (1988) study: pseudoautonomous, dependent, role reversed, identified, individuating/in progress, and individuated. Because Frank et a1. note that, "empirically derived typologies are potentially unstable from one sample to the next" (p. 736) it is especially noteworthy that almost all of the relationship typologies in the Frank et a1. study were replicated in the current one. The one exception was that an "individuating" cluster 75 emerged in this sample of late adolescents but was not apparent in the Frank et a1. sample of young adults. Undoubtably, this category describes a developmentally specific style of relating to parents that is no longer salient as children move into adulthood. Interpretation of the results from analyses using the relationship categories provided a more complex understanding of qualitative differences in parent/adolescent relationships in alcoholic and non- alcoholic families in that they took all three dimensions of the parent/adolescent relationship into account. However, like the dimensional analyses, these results reiterated the importance of considering contextual factors (i.e., parent conflict) when describing these differences. For example, stylistic differences between ACOAs and comparison subjects in their relationships with their mothers emerged in the initial chi square analyses, but proved to be a function of differences associated with parent marital conflict. In addition, PPCR appeared to moderate differences between ACOAs' and comparison subjects' relationships with fathers, and the clearest drinking group differences emerged after accounting for the degree of parent marital conflict. Regardless of parents' ability to resolve their conflicts, ACOAs described "identified" relationships with fathers more often than comparison subjects. Very little has been written about ACOAs' experience of "identified" relationships with their fathers. However, common descriptions of denial as a primary family defensive stance (see Seixas, 1982 and Straussner et al., 1979) help explain this phenomenon. Specifically, denial may serve to maintain idealized views of the alcoholic parent and, as such, may support identifications with a 76 parent who otherwise would be devalued as a role model and source of advice and support. In other words, those ACOAs who idealized their fathers (as found in the identified group) appeared to avoid conflictual feelings in regard to the father by denying, rather than taking responsibility for, the father's deficits. Parent conflict also appeared to be unrelated to drinking group differences in adolescents' experiences of "individuating" relationships with fathers. Specifically, ACOAs were less likely than comparison subjects to describe individuating relationships, regardless of the degree of parents' marital conflict. This finding is not a surprising one since adolescents in the individuating category were able to use parents for advice and assistance while making gains in separateness within the context of relatively harmonious relationships with the parent. The fact that many ACOAs continued to experience unresolved and overwhelming negative feelings toward the father apparently precluded them from making gains in individuation (at least from fathers), while adolescents from non-alcoholic families were better equipped to do so. Although drinking group differences were clear for the identified and individuating categories, other group differences emerged only in the context of either positive or negative parent conflict resolution. When parents were perceived as unable to adequately resolve their conflicts, ACOAs described relationships with fathers which were frequently characterized by pervasive role reversals. In fact, 211 subjects in the role reversed group were children of alcoholics. Clearly, this supports the contention of numerous writers in the field who have provided clinical evidence for children's tendency to become parentified or to assume responsibility for caring for their alcoholic 77 parent (e.g., Homonoff 8 Stephen, 1979; Bogdaniak 8 Piercy, 1987; Seixas, 1982). It was in the context of these role reversals that ACOAs exhibited the predicted relationship pattern of high self-reliance but low emotional autonomy, and an inordinate degree of responsibility for caring for the needs of the alcoholic parent. What was most striking about this finding was that pp£p_paternal alcoholism and poor parent conflict resolution were necessary for adolescents to describe role reversed relationships with fathers. In contrast, adolescents who described individuated relationships with fathers reported neither paternal alcoholism 22; poor parent conflict resolution. It appeared that adolescents reporting individuated relationships with fathers may, in fact, be develOpmentally advanced in their ability to experience themselves as both self-reliant and emotionally autonomous, and this advanced sense of separation from fathers was not likely to occur if the father/adolescent relationship was disrupted by either paternal alcoholism or marital conflict. Limitations pf the Study and Directions for Future Research Results of this study provide empirical validation of clinically- and theoretically-derived hypotheses concerning the impact of alcoholism on separation/individuation. Through empirical analyses of an in-depth interview, the study provides a relatively sophisticated and clinically sensitive way of evaluating adolescents' experiences of their relationships with their parents. In addition, the results point to contextual variables that moderate differences in these relationships between children from families with or without an alcoholic parent. Regardless, several limitations of the study are worth mention. 78 First, using a sample of college students inevitably limits the generalizability of the results. Research pointing to ACOAs' difficulties with academic achievement and failure to complete school (Deutsch, DiCicco 8 Mills, 1982) suggests that those ACOAs who go to college represent only a subsample of the larger ACOA pOpulation. Additionally, the choice to include only adolescents whose parents were married undoubtably further restricts the generalizability, particularly because marital dissolution is so frequently a casualty of alcoholism. Hence, as noted previously, the families of these ACOAs may be less dysfunctional, and the adolescents themselves may be relatively less impaired than a larger pOpulation of ACOAs. Second, the findings of this study are based exclusively on data provided by the adolescents. Several problems arise from reliance on self-report data. As noted previously, for example, adolescents' reports of their parents' drinking may be somewhat less accurate than obtaining parent reports of their own drinking. Additionally, the use of self-report data increases the likelihood of non-independent measures of family functioning. For example, it was noted that adolescents describing role reversed relationships with their parents also reported poor resolution of parent marital conflict. While this finding might suggest that parental conflict in some way "forces" children to assume parentified roles, it may also reflect a role-reversed adolescent's tendency to endorse such items as, "my parent asks me for advice." Future studies should minimize self-report bias by obtaining independent measures of family functioning, either through observational measures or by including reports from both children and parents. Third, the small sample size restricted the number and kind of 79 statistical analyses that could be performed and leads to a recommendation for replication. A larger sample size would also allow for more in-depth analyses of particular family dyads and would make it possible to more carefully examine the extent and nature of structural differences in the dimensions of parent/adolescent relationships among ACOAs versus adolescents from non-alcoholic families. Additionally, future studies may want to more narrowly define the parameters of alcoholism. As noted above, prOpositions that there are different types and/or manifestations of alcoholism (e.g., Zucker, 1987) should be considered, and the variable impact of different types of alcoholism on parent/adolescent relationships should be included in future research. It would also be interesting and important to look at differences in the parent/child relationship at different ages, including early adolescence and later adulthood. Results from the current study suggest some develOpmental differences in the structure of parent/adolescent and parent/young adult relationships. In this study of late adolescents, the factor structures of the relationship dimensions differed from those found by Frank et al. (1988) for young adults in slight, but meaningful, ways. For example, perceptions of self-assertion in relation to parents loaded on the self-reliance factor for late adolescents but not young adults, supporting the notion that in "healthy" autonomy development, self-reliance evolves out of relationships with parents that allow adolescents increasing opportunities for asserting their own beliefs and opinions (Grotevant 8 Cooper, 1986; Ryan 8 Lynch, 1989; Allen et al., 1990). Also in contrast to the Frank et al. (1988) study, the emotional 80 autonomy factor for late adolescents' relationships with their fathers (but not mothers) included a sizable negative loading for decision- making. Conceivably, when adolescents experience overwhelming negative feelings toward their fathers, they may distance themselves and abruptly disengage by making active denouncements of their fathers through denying their decision-making power. In contrast, adolescents who are more emotionally autonomous may be better equipped to accept their fathers as advisors. An obvious question is why decision-making did not also load on the emotional autonomy factor for mothers. Adolescents typically use mothers less frequently than fathers as advisors in major decisions (Youniss 8 Smollar, 1985); hence, they may not use denouncement of mothers' decision-making power to distance themselves from conflictual mother/adolescent relationships. Although the results from this study "hint" that late adolescents and young adults experience self-reliance and emotional autonomy in somewhat qualitatively different ways, cross-sectional data on diverse pOpulations can provide only limited information regarding whether certain relationship types develOpmentally precede and/or are "less mature" than others. It would be interesting and useful to employ longitudinal analyses to evaluate whether paternal alcoholism and parent marital conflict delay separation/individuation or alter the process in a more qualitative manner. Finally, the current study does not address the implications of variances in the parent/adolescent relationship for outcome in adolescents from alcoholic versus non-alcoholic families. It may be, for example, that certain types of parent/adolescent relationships are Imore "favorable" for ACOAs than for non-ACOAs (or vice versa). 81 Similarly, children of alcoholics may be at a greater risk for negative outcome if they describe certain types of relationships with their parents. One might suspect, for example, that adolescents who are identified with their alcoholic fathers may be at risk for greater psychOpathology (including alcoholism) than other children of alcoholics. Past research has suggested, for example, that when non- alcoholic mothers hold their alcoholic husbands in high esteem, sons may be at greater risk for alcoholism because of the increased likelihood of identifying with (and imitating) the alcoholic father (McCord, 1988). In sum, the current study has its limitations and may pose as many new questions as it has answered. Nonetheless, the results are compelling and suggest that, although many ACOAs describe relationships with their parents that are in many ways similar to relationships described by adolescent children of non-alcoholics, others disprOportionately experience pervasive and overwhelming negative feelings toward the parent and/or find it more difficult to individuate from parents within the context of positive relationships. More generally, the study makes two important contributions. First, these data provided compelling replication of earlier descriptions of parent/child relationships in young adulthood using in- depth interviews. The richness of the information obtained through the interview was particularly important for tapping into the sometimes more subtle aspects of separation/individuation (e.g., the unstated rage, disappointment, etc. characteristic of low emotional autonomy) and identifying holistic differences between ACOAs and non-ACOAs in the style of relationships experienced with parents. 82 Just as importantly, the results of this study underscore the importance of considering contextual factors when studying the effects of paternal alcoholism. Specifically, the impact of paternal alcoholism on late adolescent separation/individuation may not be evident without controlling for the effects of parent marital conflict. Further exploration of this and other contextual factors which may moderate the effects of paternal alcoholism on parent/adolescent relationships is an important task for future research. APPENDIX A 84 Part II. My mother drank most heavily when I was between the ages of years and years. FOR THAT PERIOD, please respond to the following questions. 1. On the average, how often did your mother drink any kind of alcoholic beverage during a typical month? Circle one. a. rarely b. about once a month c. 2 to 3 times a month d. 1 or 2 times a week e. 3 or 4 times a week f. nearly every day g. 2 times a day h. 3 or more times a day 2. When your mother drank any kind of alcoholic beverage, what was the average quantity she consumed; that is, how many drinks did she consume nearly every time or more than half the times she drank? A standard drink is defined as a 4-oz glass of wine, a 12-oz beer or a 1.5-oz drink of distilled spirits. Circle one. a. rarely drank b. 1-2 c. 3-4 d. 5-6 e. 7-8 f. 9-10 g. more than 10 3. Has your mother experienced any of the following problems because pf her drinking? Circle all that apply. a. marital separation or divorce f. treatment for alcohol- b. loss of employment related problems c. two or more arrest for g. economic distress drunken driving h. accidents d. two or more arrests for i. loss of friends public intoxication or j. shame of family drunken and disorderly k. belligerence, fighting, conduct or marital discord e. harm to health 1. a single alcohol related arrest 4. Does your mother presently drink 1) the same amount or b) less than she did when she was drinking most heavily? (Check one). If you checked "b", what accounted for the change? 85 Part III. Please answer the following about your parents' CURRENT drinking practices. 1. On the average, how often does your father drink any kind of alcoholic beverage during a typical month? Circle one. a. rarely b. about once a month c. 2 to 3 times a month d. 1 or 2 times a week e. 3 or 4 times a week f. nearly every day g. 2 times a day h. 3 or more times a day 2. When your father drinks any kind of alcoholic beverage, what is the average quantity he consumes; that is, how many drinks does he consume nearly every time or more than half the times he drinks? A standard drink is defined as a 4-oz glass of wine, a 12-oz beer or a 1.5-oz drink of distilled spirits. Circle one. a. rarely drank b. 1-2 c. 3-4 d. 5-6 e. 7-8 f. 9-10 g. more than 10 3. On the average, how often does your mother drink any kind of alcoholic beverage during a typical month? Circle one. a. rarely b. about once a month c. 2 to 3 times a month d. 1 or 2 times a week e. 3 or 4 times a week f. nearly every day g. 2 times a day h. 3 or more times a day 4. When your mother drinks any kind of alcoholic beverage, what is the average quantity she consumes; that is, how many drinks does she consume nearly every time or more than half the times she drinks? A standard drink is defined as a 4-oz glass of wine, a 12-oz beer or a l.5-oz drink of distilled spirits. Circle one. a. rarely drank b. 1-2 c. 3-4 d. 5-6 e. 7-8 f. 9-10 g. more than 10 APPENDIX B Appendix B Family Situation Checklist How true is each statement of your current situation? 1. My parents argue with each other in front of me. 2. I stay out of my parents' arguments. 3. I worry about my mother. 4. My parents hold grudges for a long time when fighting. 5. My mother asks me for advice. 6. My parents fight verbally with each other. 7. I take my mother's side when my parents fight. 8. I worry about my father. 9. I have a lot of responsibility in the family. 10. I enjoy being with my mother. 11. My mother tries to met me to side with her when she fights with my father. 12. My parents fight about money and possessions. 13. I am irritated when my parents fight. 14. My parents are able to resolve disagreements fairly quickly. 15. My parents fight physically with each other. 16. My father tries me to get to side with him when he fights with my mother. 17. I enjoy being with my father. 18. My mother says negative things to me about my father. 0‘ 87 19. I take my father's side when my parents fight. 4 3 2 20. My parents talk together about my future. 4 3 2 21. My parents are able to discuss and resolve their 4 3 2 disagreements. 22. My father asks me for advice. 4 3 2 23. My parents work together as parents. 4 3 2 24. I am embarrassed when my parents fight. 4 3 2 25. My parents never fight. 4 3 2 26. My father says negative things to me about my 4 3 2 mother. 27. I am upset by my parents' fighting. 4 3 2 28. Even if my parents fight about other things, 4 3 2 they respect each other as parents. 29. How well do you think your parents get along with each other? 1 2 3 4 5 very well badly 30. How often do your parents get angry with one another or disagree? 1 2 3 4 5 never all the time APPENDIX C Appendix C YOUNG ADULT-MOTHER TRANSITION INTERVIEW Name: Age: Interviewer: Code: Date: Parent Name: Introduction: We talked a little about this on the phone .... what kinds of contacts do you have with your parents now, such as visits, writing letters, telephone calls, and such? Let's start with your MOTHER. 1. What kinds of contacts do you have with your mother? (List each below). For each one mentioned, ask: "About how often do you do this?" CONTACT PER MONTH PER YEAR IF SUBJECT IS LIVING IN THE PARENTAL HOME, ASK: About how much time do you spend with your mother? 2. What kinds of things do you do when you're together with your mother? 3. What do you enjoy about your contacts with your mother? 4. What does your mother enjoy about these contacts? 89 YOUNG ADULT-MOTHER TRANSITION INTERVIEW 5. Of course, there's usually something we don't enjoy about others what might these things be about your mother? 6. What do you think might not be so enjoyable about these contacts for your mother? 7. Who initiates the contacts between you and your mother? Can you give me an example? 8. Do you wish you had more or less contact with your mother? IF MORE OR LESS: Why would you prefer (more/less) contact? IF SAME: What makes it seem about right? 9. 1In general, how much does your mother talk about her personal concerns with you? What kinds of things does she discuss? How do you feel about that? 90 YOUNG ADULT-MOTHER TRANSITION INTERVIEW 10. And how much do you talk about your personal concerns with her? What kinds of things do you discuss? (How do you feel about that?) 11a. Are there things that you avoid talking about with your mother? What kinds of things? b. What do you think makes you avoid these topics? (Can you give me one or two examples? How do you feel about that?) 91 YOUNG ADULT-MOTHER TRANSITION INTERVIEW 12. In what ways do you feel close to your mother? What kinds of things bring you together? And how do you and your mother express closeness? 13a. Are there any feelings of tenseness when you and your mother are together? Can you give me an example? b. How often does that occur? c. What do you think that feeling of tenseness is about? 92 YOUNG ADULT-MOTHER TRANSITION INTERVIEW 14. In what ways would you like to be closer to your mother? (Can you give me an example?) 15. What gets in the way of closeness between you and your mother? 16a. At what point were you closest to your mother? b. At what point were you the most distant from your mother? In what ways? c. (Take latest referent point, or age, whether close or distant and ask:) Your relationship has changed since (age or time ....) How do you account for the difference? 93 YOUNG ADULT-MOTHER TRANSITION INTERVIEW 17. In what ways are you like your mother? (What does that tell you about yourself?) 18. In what ways are you different from your mother? (What does that tell you about yourself?) 19a. In what ways does your mother show concern for your needs or welfare? (Can you give me a few examples? How do you feel about this?) b. Do you think she should be more or less responsive to your needs? (How would that affect your relationship?) 94 YOUNG ADULT—MOTHER TRANSITION INTERVIEW 20. In what ways do you show concern for your mother's needs or welfare? Do you think she gets the message? IF NO: What makes you say that? Do you wish it were different? IF YES: How do you know? 21. Can you think of a time recently when you felt conflicted about your mother's needs and your needs? What was that about? (How did you handle it? How did you feel about the way it was resolved?) 95 YOUNG ADULT-MOTHER TRANSITION INTERVIEW 223. Which parent do you feel closer to? (Probe "Lots of people feel closer to one parent than the other if says Neither What makes this so?) b. What makes you feel closer to than to (other parent)? c. How do you think it came about (that you're closer to )? 23a. To change the focus somewhat Can you think of a time you talked over an important decision with your mother? IF SAYS NEVER: Imagine what would happen if you did what might happen? IF SAYS CAN'T IMAGINE: "What makes it impossible to imagine?" b- What other kinds of decisions do you discuss with your mother for example, your financial affairs, personal relationships, work or career choices? 96 YOUNG ADULT-MOTHER TRANSITION INTERVIEW 24. Can you think of a time you and your mother disagreed about something very important? Tell me about that (How did you handle that, How did you feel about your response? How did it get resolved? Is that how disagreements usually get resolved between you and your mother?) 25a. In what ways are your values different from those of your mother? (Did you ever discuss that?) b. In what ways are your values similar to those of your mother? c. We've talked about things that you discuss with your mother. Besides advice, what other kinds of things does she help you with? (Do you ask for help, or does she always offer -- who initiates -- How do you think she feels about doing that for you? How do you feel about it?) 97 YOUNG ADULT-MOTHER TRANSITION INTERVIEW 27. What happens when your mother is not available? (If say Nothing, ask ... Imagine...) 28. Are there ways in which you wish you could rely more on your mother ... or maybe less? (How do you expect that to come about?) 29. What are some of your mother's weaknesses? What do you think makes her that way? 30. In what ways would you like her to change? Do you think she ever will? (How will that come about? If says I don't know ... What do you think you or she might do to change?) IF SAYS NO: What might prevent her from changing? 31. In what ways has she let you down? 32. What kinds of things about your mother make you feel proud? What are her strengths? APPENDIX D Appendix D Coding Manual for the Parent/Adolescent Interviews DECISION MAKING Definition: This dimension has to do with the adolescent's ability to make important decisions and life choices without undue influence from the parent. This includes the adolescent's ability to make decisions in accordance with his or her own values and preferences as well as the ability to make decisions as to the nature of those values and preferences. Note 1: Adolescents who are generally able to make their own decisions may occasionally seek advice from the parent; these adolescents should be rated as high on decision making if a) the parent's input does not appear to be essential (e.g., the adolescent indicates that he or she often makes decisions without consulting the parent) and b) the parent's advice is viewed as input for self-determined decisions rather than as the final word as to what the adolescent ought or ought not to do. These distinctions are made more explicit in the descriptions of the criteria for rating the various points on the scale. Note 2: If the adolescent does not consult the parent because of negative feelings towards the parent or because of a lack of respect for the parent ppd the adolescent makes his or her own decisions without relying on the parent's standards he or she should probably be coded as high on decision making. The negative feelings about the parent are coded elsewhere and should not be confused with the decision making dimension. Note 3: Some adolescents rely on a boyfriend/girlfriend, close friend, etc. rather than the parent in making decisions. This should pg; affect the rating. Only rate how much the adolescent relies on and is influenced by the parent. SPECIFIC RATING CRITERIA The adolescent's decision making abilities in relation to the parent are coded on a 5 point scale with 1 referring to "low" decision making abilities and 5 referring to "high" decision making abilities. The criteria for rating a protocol as 1,2,3,4, or 5 on the Decision Making Scale are described below. Indicatorsip£_Level.l: a) The adolescent generally relies on the parent in making decisions or in deciding his or her values; if there are any differences at all, the adolescent is extremely susceptible to the parent's pressures to conform to the parent's beliefs and/or he or she 98 99 feels very uncertain about his or her own views. b) If and when the adolescent makes decisions that are different from what the parent wants or thinks the adolescent should do, the adolescent feels guilty, ashamed, afraid, etc. c) The adolescent feels guilty, afraid, hurt, abandoned, etc. if the parent does not agree with and/or support the adolescent's decision. Indicators gf_Level 2; a) There is some evidence that the adolescent is at least trying to make his or her own decisions, but he or she has relatively little success or he or she is still heavily influenced by the parent and continues to experience self-doubts about his or her own decisions. b) The adolescent seems to feel like he or she makes his or her own decisions, but these claims are largely unsubstantiated and there is clear evidence to the contrary. c) The adolescent is very reliant on the parent for help in making decisions in some important areas of his or her life, but there are at least a few areas where they do not seek or feel that they need the parent's advice. d) At this level there is little or only minimal evidence that the adolescent's decisions are based on self-chosen values, standards, or preferences. Indicators 23 Level 3; a) The adolescent clearly makes his or her own decisions in some important areas of his or her life, but is still clearly influenced by the parent in others. b) There is clear evidence that they adolescent is attempting to make his or her own decisions, but at the same time he or she is susceptible to and at time swayed by the parent's influence; this experience is viewed negatively by the adolescent and sometimes leads him or her to act against his or her own wishes. c) The adolescent tries to avoid discussing his or her decisions with the parent in order no to be swayed, influenced, or coerced into taking the parent's point of view. Indicators pf Level 4: a) The adolescent clearly makes his or her own decisions in life but his or her values (or what he or she values) are less clearly differentiated from those of the parent than at Level 5. For example, the adolescent identifies with (and there is little or no evidence of having questioned) the parent's values,but he or she may implement these values in a different way. b) The adolescent's values are clearly differentiated from those of the parent but he or she tends to consult the parent on a wider range of issues than in Level 5. Indicators pf_Level 2; a) The adolescent has a clear sense of conviction about what he or she believes, values, etc., these standards or preferences are at least in part differentiated from those of the parent, and he or she uses these self-determined criteria to make his or her own decisions. At this level, the adolescent has identified areas where decisions differ 99 100 from the parent and is satisfied with these differences. In addition, he or she is able to evaluate and choose whether or not to accept the parent's advice. 101 INDEPENDENCE DEFINITION: This dimension focuses on the adolescent's experience of competence in the relationship with the parent, and on his or her belief thathe or she can COpe with the challenges and setbacks in his or her life without having to rely on the parent. Note: If the adolescent does not rely on the parent because of negative feelings towards the parent or because of a lack of respect for the parent he or she should probably be coded as high if it is clear that he or she is able to COpe with his or her own life without fear of repeating the parent's perceived failures. Those negative feelings about the parent that do not directly affect the adolescent's perception of his or her cOping abilities should not be confounded with the rating for this dimension; these feelings will be coded elsewhere. SPECIFIC RATING CRITERIA: The adolescent's independence from and competence in the relationship with the parent is coded on a 5 point scale with l referring to "low" independence and 5 referring to "high" independence. the criteria for rating a protocol as 1,2,3,4, or 5 on the Independence Scale are described below. Indicators pf Level_£: a) The adolescent generally relies on the parent to cape with challenges and difficulties in his or her own life. Adolescents at this level often experience themselves as helpless and unable to cOpe without the parent's support and may feel frustrated or angry when the parent is unable to help. b) The adolescent may realize and feel badly about his or her dependency but in spite of desires or wishful thinking about being more self-reliant he or she is unable or unwilling to give up this dependency and face the world on his or her own. c) The adolescent negatively identifies with the parent's major weaknesses and feels unable to avoid repeating the parent's failures or mistakes in important areas of his or her life. d) The adolescent feels inferior to or like a failure in the parent's eyes and either accepts or feels conflicted about that definition. The adolescent may have difficulty accepting the parent's help without feeling inferior, ashamed, angry or resentful. Indicators pf Level _2_: a) The adolescent provides some minimal evidence of competence, for example, he or she at times feels able to COpe with the world, but these feelings are often transitory, are based on thin denials of insecurities, and/or are accompanied by clear examples of strong dependencies on the parent. The adolescent may vacillate between feelings of competence and inferiority or may express serious conflict and self-doubts in spite of some minimal acknowledgment of his or her competence. 102 b) The adolescent has some sense that he or she can be of help to the parent, but retains strong doubts about his or her ability to cOpe with life in the world outside the family. c) The adolescent indicates some competencies that are not being used because of strong and pervasive dependencies on the parent. d) The adolescent makes some weak attempts to be more independent but is inhibited by strong ambivalences and fears of feeling helpless or deprived if the parent's help were not available. Indicators p£_Level 33 a) There are clearly some important areas in the adolescent's own life in which he or she feels able to cope without the parent's assistance, but there are just as clearly other areas in which they feel insecure and unable to cope without relying on the parent for advice and guidance. b) The adolescent feels like he or she is in a student relationship with the parent/teacher, although he or she has some abilities or expertise to offer the parent. The adolescent sees the parent as a positive role model and is working toward, but has not yet achieved that ideal. c) The adolescent describes some feelings of inferiority but these feelings are not all-pervasive or overwhelming; there is a definite sense that the adolescent in moving toward becoming more independent, and in some areas may even feel that he or she has strengths the parent does not have. Indicators _O_f_ Level 4: a) The adolescent clearly feels he or she can cape without the parent's help but this sense of competence is not as pronounced as in Level 5; there is greater emphasis on ESE being dependent than on active goal-oriented coping. b) There is no clear evidence that that adolescent is especially proud of his or her abilities nor direct evidence that he or she is seen by the parent as especially capable. c) In spite of an overall sense of competence the adolescent reveals some isolated but obvious areas of concern or self-doubt that may be or relatively minimal importance to their daily or overall functioning. Indicators pi Level _5_: a) The adolescent feels at least on equal par with the parent in evaluating his or her ability to cope in and deal effectively with the challenges and difficulties in his or her life. The adolescent may identify with the parent's strengths and/or sees differences in a positive light and feels able to pick and choose in emulating the parent's qualities. b) The adolescent has a strong sense of confidence in his or her ability to cOpe, a perception they often feel is shared by the parent (or an evaluation that clearly is unshaken in spite of the parent's skepticism). c) The adolescent can maintain a sense of competence even when accepting help from the parent and is able to draw on other sources when the parent is not available. 103 d) Ip_addition to being able to function independently the adolescent describes areas in which his or her expertise is helpful to the parent. 104 PERSONAL CONTROL DEFINITION: This dimension refers to the adolescent's ability to control his or her behavior and feelings in the relationship with the parent. NOTE: This dimension does not refer to and should not be confounded with how much the adolescent likes or does not like the parent (although the two may be correlated). The key issue rated in this dimension is how the adolescent handles his or her needs and feelings (including negative feelings) in the relationship with the parent. If the adolescent is overwhelmed by or has little control over his or her needs or feelings they would be identified as low in personal control; however, an adolescent who has some negative feelings about a parent but who is undistressed by, has some psychological distance from, and is able to keep those feelings from spilling into inappropriate behaviors could conceivably be rated as high on personal control. SPECIFIC RATING CRITERIA The adolescent's degree of personal control in the relationship with the parent is coded on a 5 point scale with l referring to "low" personal control and 5 referring to "high" personal control. The criteria for rating a protocol as 1,2,3,4, or 5 on the Personal Control Scale are described below. Indicators p£_Level 1; a) The adolescent behaves in very inapprOpriate ways (e.g., yells, flees the scene, loses his or her temper, argues incessantly, say very hostile things to the parent that may be regretted later, cried uncontrollably, etc.) in the relationship with the parent because he or she is unable to control intense feelings of anger, frustration, or shame. Indicators pf Level 2: a) The adolescent is overwhelmed by and/or has to be constantly on guard against intense feelings of unresolved ambivalence, hatred, rage, shame or guilt that are often distressful to the adolescent. b) The adolescent expresses highly intense feelings of rage, dependency, shame, etc.; although the adolescent appears to be unashamed or untroubled by these feelings, their intensity indicates a lack of control and/or constricted rage that is serious enough to be coded at a Level 2. c) The adolescent appears to have very little tolerance for frustration in the relationship with the parent. For example, the young adult acknowledges extreme dependency needs that are inapprOpriate in late adolescence; although the parent may be meeting these needs, the adolescent indicates that if his or her needs were not met by the parent this would result in feelings of anger, deprivation, or extreme frustration. Indicators pf Level 3: a) The adolescent is irritated by often seemingly minor conflicts or behaviors on the part of the parent. 105 b) The adolescent occasionally gets into mile arguments with, is sarcastic towards, or feels mildly ashamed or guilty in the relationship with the parent; these negative feelings toward the parent and the ways in which they are expressed are less intense and more controlled than at the previous levels. c) The adolescent is disturbed or ashamed by the parent's weaknesses, relationships, or behavior even when these do not directly affect the adolescent. d) The adolescent acknowledges mild ambivalence toward the parent that creates conflict for the adolescent, but is (potentially) resolvable or at least sufficiently under control so as not to be overly distressing. Indicators pf_Level 4; a) The adolescent experiences minor irritation or discomfort in the relationship with the parent; although these feelings are mostly under control or of relatively little importance to the adolescent, they are notable because they indicate that the adolescent has not altogether resolved parent/child issues. For example, these feelings may be stimulated by situations that would not result in conflicts if the other person involved were not a parent (e.g., the adolescent feels infantalized when the parent attempts to give advice that probably would be well received from someone else). Indicators p£_Level 2; a) The adolescent experiences little or no tension, distress, anger, etc. in the relationship with the parent; in some cases the relationship is described in a positive way and may be characterized by mutual enjoyment and pleasure. b) The adolescent has developed effective cOping strategies for dealing with potentially tense interactions or negative feelings; these ways of caping allow the adolescent to master negative perceptions of the parent or to deal successfully with difficult situations that might otherwise create discomfort in the relationship. 106 SELF-ASSERTION DEFINITION: This dimension refers to the extent to which the adolescent's behavior in the relationship with the parent is inhibited by feelings of shame or guilt. At the low end of the scale adolescents implicitly or explicitly us the parent as a superego or ego ideal; these adolescents' standards for evaluating and monitoring their behaviors vis a vis the parent are insufficiently differentiated or confused with parental expectations. At the high end of the scale adolescents are uninhibited by feelings of shame or guilt, are able to rely on their own standard in relating to the parent, and are able to assert themselves in the relationship when failure to do so would compromise these standards. SPECIFIC RATING CRITERIA The adolescent's self assertion in the relationship with the parent is coded on a 5 point scale with 1 referring to "low" self assertion and 5 referring to "high" self assertion. The criteria for rating a protocol as 1,2,3,4, or 5 on the Self-Assertion Scale are described below. Indicators_pf Level 1; a) The adolescent is clearly inhibited by feelings of shame and guilt in the relationship with the parent and/or monitors his or her behavior in order to avoid the parent's negative evaluation, disapproval, contempt, anger, or retribution. If the use of the parent as an ego ideal or superego figure is mostly ego syntonic, the adolescent's excessive self-abnegation may be relatively untroublesome to the adolescent even though it is painfully evident from the perspective of an outsider. b) The adolescent's self-abnegation in the relationship with the parent is so glaring that even minimal attempts at self-assertion appear futile or ineffective in view of the larger context of the relationship. Indicators _o_f_ Level _2_: a) The adolescent, as in Level 1, is constrained in the relationship by feelings of shame or guilt, perceives the parent as an appropriate evaluator of his or her behavior and/or is inhibited by the parent's judgment or disapproval of his or her actions; however, there is some evidence that the adolescent is beginning to assert him or herself in the relationship with the parent and is questioning or re- evaluating his or her perceptions of the parent's power or authority. b) The adolescent is able to assert him or herself in a few isolated areas in the relationship with the parent, although overall the adolescent is still inhibited by shame or guilt and/or uses the parent to evaluate and monitor his or her behavior. c) The adolescent projects his or her own discomfort about being him or herself onto the parent (e.g., "he can not take it when I let him know who I really am") and/or denies or does not acknowledge that he or she is concerned about or fears the parent's disapproval; however, this denial is difficult to believe in the face of obvious clues to the contrary. For example, the adolescent has tremendous tension or anxiety about opening up with the parent or is still emotionally involved with and experiences guilt or shame in response to memories of past 107 conflicts. Alternatively, the adolescent may be constrained in the relationship because he or she is intensely ashamed of or embarrassed by the parent; the parent functions as a negative ego ideal, detracting from and constraining the adolescent's ability to be him or herself in the relationship with the parent and implicitly or explicitly diminishing the adolescent's own feelings of self-worth. Indicators p£_Level 3; a) In some areas the adolescent appears to be uninhibited by shame or guilt in the relationship with the parent (e.g., he or she can discuss potentially shameful experiences or assert and maintain his or her own standards when these differ from those of the parent); However, in other important areas the adolescent appears to be constrained in the relationship by feelings of shame or guilt (be it shame about his or her own or the parent's characteristics) and these areas compromise the adolescent's feelings of self-worth. b) In some significant areas the adolescent implicitly or explicitly uses the parent to set standard fro his or her behavior in the relationship and/or indicates that he or she continues to need the parent's approval; in other areas, however, he or she is more able to be a self-evaluator. Indicators _o_f_ Level 3: a) The adolescent generally is uninhibited by feelings of shame and uses his or her own standards to monitor his or her behavior in the relationship with the parent. However, there is a fairly isolated area in which he or she continues to be inhibited and implicitly or explicitly seeks or desires the parent's approval. b) Although the adolescent generally is comfortable being him or herself in the relationship with the parent he or she admits to an important but isolated area in the relationship in which he or she is unable to be him or herself. Although the constraints are mostly outside of the adolescent's control the tensions in some way detract form the adolescent's feelings of self-worth; implicitly if not explicitly, the adolescent appears to need the parent's acknowledgment or approval to alleviate feelings of self-doubt, rejection, etc. Alternatively, the adolescent may be ashamed pf the parent and while these feelings are relatively isolated they detract in some way from the adolescent's feelings self-worth and from a generally high level of self-assertion in the relationship with the parent. c) While it is evident that the adolescent has developed his or her own standards and is not inhibited by guilt or shame in the relationship with the parent, repeated statements about pp£_needing the parent's approval suggest that his or her sense of being a self-evaluator is not on as firm ground as at Level 5. gdicators o_f Level 2: a) The adolescent is able to assert him or herself and express his or her needs, values, and interests in the relationship with the parent even when these needs or values clash with those of the parent. The adolescent gives no evidence that he or she views the parent as an anthority figure or as an apprOpriate judge of the adolescent's self- WC>1'th or behavior. 108 b) The adolescent clearly has his or her own standards for evaluating his or her behavior and these can be clearly distinguished from those of the parent. The adolescent does not sacrifice these standards in the relationship with the parent although at times he or she may avoid certain issues that might lead to unnecessary tensions. At Level 5, not bringing up these issues does not in any way detract from the adolescent's feelings of self-worth and is rather an indication that he or she does not need the parent's approval. c) The adolescent's relationship with the parent is limited more by the parent's conflicts, inhibitions, etc. than by the adolescent's lack of assertiveness. Although the adolescent may Openly confront the parent in an attempt to change the relationship, he or she is able to acknowledge and accept that, given the parent's limitation, he or she will "never" be him or herself with the parent. This knowledge does not compromise the adolescent's standards nor detract from his or her feelings of self-worth. 109 SELF/OTHER RESPONSIBILITY DEFINITION: This dimension describes how the adolescent resolves conflicts and pulls between obligations and needs created by his or her own life circumstances and the needs of and the obligations to the parent. Conflicts about where and with whom to spend vacation often provide information on this dimension. Self/other responsibility is similar to what other theorists have described as the parent's ability to allow the child to develop relationships and interests outside of the parent/child dyad; however, here it is coded from the adolescent's perspective and describes his or her ability to make attachments and investments outside of the relationship with the parent. Note: Most adolescents who are high on this dimension should be able to make new investments without totally denying the needs of or breaking contact with the parent. However, an adolescent may have a very negative relationship with the parent and still score high on self/other responsibility ithe or she has develOped new sources outside of the parent/adolescent relationship for finding love and affection, fulfillment, stimulation, efficacy, etc. An adolescent who continues to rely heavily on the parent for these psychological resources p£_who fails to make psychological investments outside of the dyad because he or she continues to be embroiled in unresolved conflicts with the parent would be rated as low on self/other responsibility. The quality of the parent/adolescent relationship that to a greater or lesser extent is left behind is irrelevant in coding this dimension and will be rated elsewhere. SPECIFIC RATING CRITERIA The adolescent's self/other responsibility in the relationship with the parent is coded on a 5 point scale with 1 referring to "low" self/other responsibility and 5 referring to "high" self/other responsibility. The criteria for rating a protocol as 1,2,3,4, or 5 are described below. Indicators‘p£_Level_l: a) The adolescent is unable to leave home in the psycholggical sense. The adolescent has not transferred his or her loyalties and investments to new sources of affection, stimulation, fulfillment, etc.; rather, he or she relies heavily on the relationship with the parent for stimulation, affection, fulfillment, etc. and feels lonely or abandoned without constant contact. b) The parent is involved in the minute details of the adolescent's life. The adolescent may have excessive responsibility for or involvement in the parent's life or concerns and/or is responsible for involving the parent in his or her own life concerns. c) The adolescent is often triangulated in the parent's marital or family relationships and/or is unable to keep the parent from interfering in his or her own relationships or efforts to make an independent life for him or herself. Indicators pf_Level 2; 110 a) As in Level 1, there are clear indications that the parent in overinvolved in significant areas of the adolescent's life and vice versa, but the enmeshment is not as intense, emotionally compelling, or all-pervasive as in a Level 1 protocol. The adolescent may view the enmeshment positively, may not exert any effort to separate, or else might make weak but mostly ineffective attempts to disinvest in the dyad. b) The adolescent resists pressures to disinvest in the dyad or yearns to increase his or her involvements at the expense of separation even though these efforts may be resisted by the parent. Indicators pf_Level.3: a) The adolescent makes clear attempts to resist the parent's intrusiveness, but constantly has to be on guard against inner pulls (e.g., guilt of dependency needs) and/or parental pressures that may interfere with or encroach on involvements and relationships outside of the dyad or family of origin. b) The adolescent's overinvolvement with in the dyadic relationship with the parent and/or inner pulls towards excessive involvement are moderated by investments in other important areas (e.g., other close relationships, career, etc.); these other investments are relevant or rating self/other responsibility if they serve as alternative sources of support affection, fulfillment, etc. 222 help to deintensify involvements with the parent. Indicators _o_f Level 4: a) The adolescent describes minor instances of overinvolvement or minimal conflicts about separating from the parent, but for the most part the adolescent has made satisfying or fulfilling investments outside of the parent/adolescent relationship. b) Although the parent attempts to triangulate or remain excessively involved in the adolescent's life, the adolescent for the most part actively and successfully fends off these attempts by the parent. c) There is some evidence that the adolescent has in some ways given in to his or her own or to the parent's desires for triangulation, overinvolvement, etc., but these compromises only minimally affect the adolescent's generally successful attempts to make satisfying investments outside of the dyadic relationship. Indicators gf_ Level 2: a) The adolescent clearly has made satisfying, stable, and fulfilling investments outside of the dyadic relationship with the parent and is able to regulate the relationship so that contacts with the parent do not intrude on his or her relationships outside of the dyad. b) The adolescent is able to share in the parent's life or to share their life with the parent (if this is viewed as desirable) without fear of triangulation, overinvolvement or enmeshment. c) If the adolescent does not directly address the issue of self/other responsibility but there is not indication of overinvolvement, triangulation, or enmeshment, code the protocol as a 5. 111 CLOSENESS DEFINITION: This dimension refers to the adolescent's perceptions of the degree of emotional connected versus estrangement and distance in his or her relationship with the parent. Note: Relationships with a good deal of conflict and negative feelings may nevertheless be close; the major issue in rating closeness is the depth and breadth of emotional connectedness, not the positive or negative tone of the relationship. Although it is unlikely that a relationship with a great deal of conflict would have sufficient closeness to be rated as a 5, it is not unusual for such a relationship to meet the criteria for a level 4 rating. SPECIFIC RATING CRITERIA The adolescent's experience of closeness in the relationship with the parent is coded on a 5 point scale with 1 referring to "low" closeness and 5 referring to "high" closeness. The criteria for rating a protocol as 1,2,3,4, or 5 on the Closeness scale are described below. Indicators 2: Level _1_: a) The adolescent reports feelings of being emotionally out of touch or estranged in the relationship with the parent; the relationship may appear cold and mechanistic and if the adolescent feels any sense of closeness it is more out of obligation than affection, and/of non- consequential given the overall level of estrangement. b) The adolescent describes closeness only in the negative, i.e., as the absence of tension or conflict rather than as a sense of connectedness. c) The adolescent indicates little desire for greater connectedness and/or actively sets up barriers against or indicates that he or she does not want a sense of closeness. d) The adolescent shies away from the parent's attempts at greater emotional connectedness and experiences these attempts as intrusions. Indicators p£_Level g; a) The adolescent describes some minimal sense of connectedness, e.g., the relationship remains quite distant but there is at least a noticeable improvement over the past. b) Although the adolescent clearly desires a greater sense of emotional connectedness, this is prevented by psychological barriers and prevailing feelings of estrangement. c) The adolescent's sense of connectedness to the parent is more intellectual than emotional although this does allow for some minimal sense of bondedness. _Indicators‘p£_Level.3: a) The adolescent reports a definite sense of emotional connectedness to the parent but this is based largely on common interests, a sense of family, grandchildren, share activities, etc. ‘rather than strong and deep emotional ties. b) The adolescent's sense of closeness is based on conventional and 112 stereotyped ways of sharing and relating often because the adolescent feels a need to avoid greater intimacy due to conflicts, disagreements, etc. Indicators pf Level 4; a) The protocol combines elements of Level 3 (closeness based on more stereotyped modes or sharing) and Level 5 (closeness based on deeper feelings of intimacy). b) The adolescent has a deep sense of emotional connectedness to the parent or experiences a growing sense of closeness, but there are some notable areas in which the adolescent feels distant from the parent, e.g., because of some past events, lack of shared values, or other obstacle. The adolescent may report as much depth in the relationship as at Level 5 but some limitations in breadth. Indicators pf_Level 5; a) The adolescent's description of deep emotional connectedness to the parent is supported by multiple indicators, for example, feelings of deep affection, understanding, support, shared experience, mutual enjoyment, etc. 113 COMMUNICATION DEFINITION: This dimension refers to the degree of Openness (i.e., both depth and breadth) in the communication between the parent and adolescent. SPECIFIC RATING CRITERIA The adolescent's communication with the parent is coded on a 5 point scale with a 1 referring to "low" communication and a 5 referring to "high" communication. The criteria for rating a protocol as 1,2,3,4, or 5 on the Closeness Scale are described below. Indicators Lf Level 1: a) The adolescent indicates that the lines of communication with the parent ar shut off and reports that he or she has no interest in communicating with the parent. Indicators Lf Level 2: a) The adolescent communicates with the parent around superficial, non-controversial subjects but there are clear limits on what can be talked about; e.g., there may be a sense of discomfort or, distrust in communicating about more personal or meaningful issues. b) The adolescent expressed a yearning to be able to communicate with the parent about important issues that are somehow designated as off limits. c) The adolescent's communication with the parent is very narrowly focused, often on a single issue. d) The adolescent reports that he or she communicates with the parent only when necessary or convenient without any evidence that he or she is interested in seeking out greater communication, e.g., he or she indicates that there "is not a lot to talk about." Indicators Lf Level 3: a) The— adolescent reports greater depth in his or her communication with the parent than at Level 2 and clearly values this communication; however, the exchanges with the parent generally revolved around stereotypical or "current" concerns, global attitudes, shared activities or interests, family matters, etc. b) The adolescent may indicate that the level of communication with the parent is satisfactory or may even prefer somewhat deeper communication, but in either case he or she avoids more controversial or intimate subjects. Edicators 93 Level 4: a) The adolescent describes Open and deep communication with the parent much like that found at Level 5, but the communication is qualified by some specific area that is "off limits"; the adolescent Perceives a "block" in communication that has a noticeable, and in the adolescent's eyes, a regrettable effects on the quality of an otherwise °Pen relationship. Indicators _oi Level _5_: 114 a) The adolescent describes a great deal of Open communication with the parent characterized by an ability to express differences in views, values, and conflicts. b) The adolescent indicates that his or her communication with the parent is characterized by a mutual sharing of deep and meaningful issues and personal issues, although in some cases this may not include very private matters that are outside of the parent/adolescent relationship (e.g., fights with a boy/girlfriend). Even at this level, the adolescent may hold off talking about some circumscribed issues he or she feels are better off not shared; withholding these issues does not detract from the quality of the relationship. 115 CONCERN DEFINITION: This dimension has to do with the extent to which the adolescent shows evidence of concern about the parent's well-being and interest in the parent's needs and feelings. SPECIFIC RATING CRITERIA The adolescent's concern for the parent is coded on a 5 point scale with 1 referring to "low" concern and 5 referring to "high" concern. The criteria for rating a protocol as 1,2,3,4, or 5 on the Concern Scale are described below. Indicators pf Level_£: a) The adolescent shows no interest in the parent's well-being, needs, or concerns and denies any sense of responsibility for feeling or showing concern. The adolescent is "actively unconcerned", denies the parent has real needs, or is simply indifferent to those needs. Indicators pf Level 3; a) The adolescent shows only minimal concern and this concern is expressed primarily in highly concrete, behavioral ways. b) The adolescent's concerns for the parent are focused on a very limited issue. c) The adolescent does not demonstrate much effort or interest in showing concern for the parent except out of an obligation or to get something in return; in the latter instance the adolescent's concern for the parent revolves around his or her own needs and is often displayed as a kind of concrete exchange. Indicators pg Level 3: a) The adolescent's concerns for the parent extend over a range of different areas but his or her concern is largely expressed by "helpful" behaviors, i.e. "doing things for the parent" rather than deeper psychological feelings. b) There is evidence that the adolescent longs to express feeling of concern for the parent but these desires are frustrated, qualified, or cut off by forces largely outside of the adolescent's voluntary control (e.g., resentments, parent's inaccessibility, etc.). The adolescent may feign an "I don't care anymore" or "I've given up" attitude, but his or her longing and caring still come through. Indicators 2; Level 4: The adolescent's concern for the parent at this level is more than just behavioral. There is clear evidence that the adolescent is interested in the parent's psychological well-being and EITHER: a) a strong sense of responsibility for being responsive to the parent's needs or to ensure their welfare; OR b) evidence of concern about the parent's fate outside of the dyadic relationship or the family setting, interest in the parent's effect on other people, or a concern about the ability to cope with stresses, etc. The adolescent's interest is in the parent's welfare rather than in the direct or indirect effect of the parent's well-being 116 on the adolescent or the adolescent/parent relationship. Indicators‘p£_Level.§: a) At this level, the adolescent must show clear evidence of criteria for both a) and b) from Level 4; i.e., they indicate an immediate responsiveness to the parent's difficulties and concerns and clear interest in the parent's welfare in areas that are outside of the dyadic relationship. If there is some doubt as to whether criteria from both a) and b) are present, code the protocol as a Level 4. 117 EMPATHY DEFINITION: This dimension has to do with the adolescent's ability to understand the parent as a complex person and to appreciate the parent's feelings and perspective, even when these differ from those of the adolescent. Note: Concern or sympathy do not enter into the rating of empathy. An adolescent may have a fairly complex understanding of the parent (empathy) but also acknowledge difficulties in the relationship that interfere with concern, positive feelings, etc. SPECIFIC RATING CRITERIA The adolescent's degree of understanding of the parent is coded on a 5 point scale with 1 referring to "low" empathy and 5 referring to "high" empathy. The criteria for rating a protocol as 1,2,3,4, or 5 on the Empathy Scale are described below. Indicators o_f_ Level A: a) The adolescent describes the parent primarily in concrete behavioral or black and white (extremely bad or extremely good) terms. b) The adolescent describes the parent in terms of a single overriding issue or trait that directly or indirectly has to do with the parent's relationship with the adolescent. c) The adolescent is unable to or has little interest in understanding the parent's own issues or concerns, or in viewing the parent as having a life outside of the dyadic relationship. For example, the adolescent often sees the parent's weaknesses as barriers to the relationship (e.g., the parent is nosy, pries into the adolescent's private business) without any attempt to understand these weaknesses from the parent's point of view. Indicators g£_Level g: a) Although the adolescent's description of the parent is mostly behavioral and concrete and tends to portray the parent in black and white terms (as in Level 1), the adolescent makes a minimal attempt to understand the parent as a psychological being. For example, the adolescent might comment on the parent's needs, attitudes or motive, but only in a simplistic, shallow, and mostly behavioral way (e.g., he is really concerned about money so he is not home very much, that's how he was brought up (unelaborated)1. b) The adolescent's description of the parent combines elements of Level 1 (concrete, behavioral descriptions) and Level 3 (descriptions of the parent in terms of global stereotyped traits and attitudes). fidicators Lf Level _3: a) The adolescent describes the parent in terms of stereotypical traits and relatively simplistic attitudes and feelings. The adolescent attempts to take the parent's perspective; however, he or she has difficulty seeing the parent's perspective from multiple points of view and is unable to recognize the parent as a diverse complex person. Although the parent may be viewed as having both positive and negative 118 traits, the adolescent is unable to integrate or understand the relationship between the parent's strengths and weaknesses. b) The adolescent may seek to understand the parent outside of the dyadic relationship, in relationship to the larger social context, or in relation to the parent's past experiences; however, the description of the parent is simplistic or stereotyped (e.g., an elaboration of "that's the way he was raised", "he has an unhappy marriage", "he was very close to his family" etc.) c) The adolescent's attempts to understand the parent are colored by projection of the adolescent's own attitudes and feelings or by uncertainty or idealization. d) The adolescent yearns to understand the parent but, largely because of limitations in the parent or other forces beyond the adolescent's control, these desires are thwarted. As a result, the adolescent has only a stereotyped or minimal understanding of the parent, even though he or she might be capable of more. Indicators p£_Level 4: a) The adolescent's description of the parent combines elements of Level 3 (stereotyped descriptions of the parent) and Level 5 (description of the parent as a complex person). b) The adolescent describes the parent in complex psychological terms (e.g., how he or she cOpes with difficulties, complex motives, psychological causation), as at Level 5, but this description primarily focuses on the adolescent's relationship with the parent rather than on the parent's attachments or concerns outside of the dyad. For example, the adolescent has a less in-depth appreciation of the parent's motives, needs, or concerns when these are not of direct relevance for the adolescent/parent relationship. Indicators p£_Level 2: Both of two characteristics are essential at Level 5: a) The adolescent views the parent as a person in his or her own right, and attempts to understand them in relation to the larger social context and/or relationships outside the parent/adolescent dyad, AND b) The adolescent describes the parent in process rather than behavioral terms and focuses on complex motive, feelings, ways of cOping with the world, ideologies, values, beliefs, etc., For example, the adolescent attempts to integrate the parent's positive and negative attributed into a complex psychological portrait; the parent is described in terms of contrasts and conflicts rather than in terms of simple stereotypical attitudes and traits. In addition, the adolescent may attempt to contrast and compare his or her own perspective with the parent's perspective. The parent's perspective is seen as multidimensional and overdetermined, and the adolescent may recognize how his or her own behavior indirectly (rather than indirectly) contributes to the parent's reaction to or views of the adolescent. LIST OF REFERENCES LIST OF REFERENCES Ackerman, R.J. (1983). Children pf_Alcoholics: ézguidebook for educators, therapists, and parents(2nd ed.). Holmes Beach, FL: Learning Publications. Adelson, J. 8 Doerhman, M.J. (1980). The psychodynamic approach to adolescence. In J. Adelson (Ed.) Handbook.pf_Adolescent Psychology, New York: Wiley-Interscience Publications. Adler, R. 8 Raphael, B. (1983). Children of Alcoholics. Australian and New Zealand Journal o£_Psychiatry, 17, 3-8. Allen, J.P., Aber, J.L. 8 Leadbeater, B.J. (1990). Adolescent problem behaviors: The influence of attachment and autonomy. Psychiatric Clinics p£_North America, 12(3), 455-465. Arentzen, W. (1978). Impact of alcohol misuse on family life. Alcoholism: Clinical and Experimental Research, 23 349-351. Bennett, L.A. 8 Wolin, S.J. (1986). Daughters and sons of alcoholics: Developmental paths in transmission. Alcoholism, 22(1), 3-15. Benson, C.S. 8 Heller, K. (1987). Factors in the current adjustment of young adult daughters of alcoholic and problem drinking fathers. Journal p£_Abnormal Psychologyy_gp(4), 305-312. Berlin, R., Davis, R.B., 8 Orenstein, A. (1988) Adaptive and reactive distancing among adolescents from alcoholic families. Adolescence, _2_3_, 577-584. 119 120 Black, C. (1979). Children of alcoholics. Alcohol, Health, and Research World, 4, 23-27. Blos, P. (1967). The second individuation process of adolescence. The Psychoanalytic Study of_the Child, Vol 29, New York: International Universities Press. Blos, P. (1979). The adolescent passage: Developmental issues. New York: International Universities Press. Bogdaniak, R.C. 8 Piercy, F.P. (1987). Therapeutic issues of adolescent children of alcoholics (AdCA) groups. International Journal o£_Group Psychotherapy. 21(4), 569-587. Bosma, W. (1972). Children of alcoholics -- a hidden tragedy. Maryland State Medical Journal; £l(1), 34-36. Brown, S. (1988). Treating Adult Children pf_Alcoholics: :4 Developmental Perspective. New York: Wiley 8 Sons. Burk, J.P., 8 Sher, K.J. (1988). The "forgotten children" revisited: Neglected areas of COA research. Clinical Psychology Review,8(3), 285-302. Cahalan, D. 8 Cisin, I. (1968). American drinking practices: Summary of findings from a national probability sample. Quarterly Journal Cermak, R. 8 Brown, S. (1982). Interactional group therapy with the adult children of alcoholics. International Journal p£_Group Psychotherapy,.§g(3), 375-389. Chassin, L., Rogasch, F. 8 Berrera, M. (1991). Substance use and symptomotology among adolescent children of alcoholics. Journal 2: Abnormal Psychology, 100(4), 449-463. 121 Chodorow, N. (1978) The reproduction p£_mothering: Psychoanalysis and the sociology 2; gender. Berkely: University of California Press. Christenson, A. 8 Margolin, G. (1988). Conflict and alliance in distressed and non-distressed families. In R.A. Hinde 8 J.S. Hinde (Eds.), Relationships within Families: Mutual Influences. Oxford: Clarendon Press. Clair, D. 8 Genest, M. (1987). Variables associated with the adjustment of offspring of alcoholic fathers. Journal o§_Studies op_ Alcohol, 4§(4), 345-355. Cork, R.M. (1969). The forgotten children: A study of children with alcoholic parents. Toronto: Addiction Research Foundation. Davis, P., Stern, D., 8 Vandeusen, J. (1978). Enmeshment-disengagement in the alcoholic family. In F.A. Seixas (Ed.) Currents ip_ alcoholism (Vol.4, pp. 15-27). New York: Grune 8 Stratton. Deutsch, C., DiCicco, L., 8 Mills,D.J. (1982). Services for children of alcoholic parents. In Alcohol and health monograph no. 2i Prevention, intervention and treatment: Concerns and models. Rockville, MD: NIAAA. Devereux, E. (1970). The role of peer group experience in moral development. In J. Hill (Ed.) Minnesota Symposium pp 95114 Psychology, 221 4, Minneapolis: University of Minnesota Press. Douvan, E., 8 Adelson, J. (1966). The adolescent experience. New York: Wiley. Elder, G.H. 8 Caspi, A. (1990). Studying lives in a changing society: Sociological and personological explorations. In A.I. Rabin, R.A. Zucker, R.A. Emmons, 8 S.J. Frank (Eds.) Studying_Persons and Lives. New York: Springer Publishing Company. 122 el-Guebaly, N. 8 Offord, D.R. (1977). The offspring of alcoholics: A critical review. American Journal 2! Psychiatgy, 134, 357-365. Fairbairn, W.R.D. (1981). Schizoid factors in the personality. In W.R.D. Fairbairn (Ed.) Psychoanalytic studies p£_the personality, 7th edition. London: Routledge and Kegan Paul. (Original work published 1943). Filstead, W., McElfresh, 0., 8 Anderson, C. (1981). Comparing the family environments of alcoholic and "normal" families. Journal pf Alcohol and Drug Education, 293 24-31. Frank, S.J., Avery, C., 8 Laman, M. (1988). Young adults' perceptions of their relationships with their parents: Individual differences in connectedness, competence and emotional autonomy. Developmental Psychology,g4, 729-737. Frank, S.J. 8 Burke, L. (1992). Deidealization and autonomy in late adolescence: Replications and extensions of earlier studies. Paper presented at the biannual meetings of the Society for Research on Adolescence, Washington, D.C. ’— Frank, S.J., Jacobson, S. , 8 Hole, C.B. (1988). The Parenting Alliance: Bridging the gap between marriagp_and parenting. Unpublished manuscript. Frank, S.J., Pirsch, L.A., 8 Wright, V.C. (1990). Late adolescents' perceptions of their relationships with their parents: Relationships among deidealization, autonomy, relatedness, and insecurity and implications for adolescent adjustment and ego identity status. Journal prYouth and Adolescence, 12(6), 571-588. 123 Frank, S.J., Pirsch, L.A., 8 Wright, V.C. (1992). Autonomy from parents and competence during early adulthood. Paper presented at the annual meetings of the American Psychological Association, Washington, D.C. Grotevant, H.D., 8 Cooper, C.R. (1986). Individuation in family relationships. Human Development, 22, 82-100. Grych, J.H. 8 Fincham, F.D. (1990). Marital conflict and children's adjustment: A cognitive-contextual framework. Psychological Bulletin, 14, 157-162. Hecht, M. (1973). Children of alcoholics are children at risk. American Journal p£_Nursing, 12(10), 1764-1767. Heller, K., Sher, K., 8 Benson, C. (1982). Problems associated with risk overprediction in studies of offspring of alcoholics: Implications for prevention. Clinical Psychology Review,2, 183-200. Herjanic, B., Herjanic, M., Penick, E., Tomelleri, C., 8 Armbruster, R. (1977). Children of alcoholics. In F.A. Seixas (Ed.), Currents ip alcoholism(Vol.2, pp. 445-455). New York: Grune 8 Stratton. Hill, J.P., 8 Holmbeck, G.N. (1986). Attachment and autonomy during adolescence. Annals p§_Child Developmentti, 145-189. Hollingshead, A.B. (1957). The two-factor index o£_social position. Unpublished manuscript, Yale University, New Haven. Homonoff, E. and Stephen, A. (1979). Alcohol education for children of alcoholics in a Boston neighborhood. Journal pf Studies pp Alcohol, 49, 923-926. Hughes, J.M. (1977). Adolescent children of alcoholic parents and the relationship of Alateen to these children. Psychology Bulletin, 8g, 33-65. 124 Jacob, T. (1986). Alcoholism: A family interaction perspective. In C. Rivers (Ed.) Nebraska Symposium pp_Motivation, Vol 34, Lincoln Nebraska: University of Nebraska Press. Jacob, T., Krahn, C., 8 Leonard, K. (1991). Parent-child interactions in families with alcoholic fathers. Journal 2; Consulting and Clinical Psychology, 22, 176-182. Jacob, T. 8 Leonard, D. (1986). Psychosocial functioning in children of alcoholic fathers, depressed fathers and control fathers. Josselson, R.L. (1980). Ego develOpment in adolescence. In J. Adelson (Ed.) Handbook p£_adolescent psychology, New York: Wiley. Kandel, D. 8 Lesser, G.S. (1972). Youth ip_two worlds. San Francisco: Jossey-Bass. Lund, C., 8 Landesman-Dwyer, S. (1979). Pre-delinquent and disturbed adolescents: The role of parental alcoholism. In M. Galanter (Ed.) Currents ip alcoholism (Vol. 5, pp.339-348). New York: Grune 8 Stratton. McCord, J. (1988). Identifying develOpmental paradigms leading to alcoholism. Journal p£_Studies pp Alcohol, 42(4), 357-362. Mitchell, H.E. (1958). The interrelatedness Of alcoholism and marital conflict. American Journal p£_0rthopsychiatgy,.g§, 888-891. Molholm. L.H. 8 Dinitz, S. (1972). Female mental patients and their normal controls. Archives pf_General Psychiatry, 21, 606-610. Moore, D.R. (1982). Alcohol and family interaction: Child adjustment issues. Unpublished manuscript. 125 Morehouse, E., 8 Richards, T. (1982). An examination of dysfunctional latency age children of alcoholic parents and problems in intervention. Journal of Children 32 Contemporary Society, 15, 21- 33. Murphey, E.B., Silber, E., Coelho, G.V., Hamburg, D.A., 8 Greenberg, I. (1963). Development of autonomy and parent-child interaction in late adolescence. American Journal p£_0rth0p§ychiatry, 22, 643-652. Nardi, P.M. (1981) Children of alcoholics: A role theoretical perspective. Journal of Social Psychology, iii, 237-245. Obuchowska, I. (1974). Emotional contact with the mother as a social compensatory factor in children of alcoholics. International Mental Health Research Newsletter, 14, 2-4. Offer, D., Ostrov, E., 8 Howard, K. (1981). The adolescent: :4 psychological self:portrait. New York: Basic Books. O'Malley, 8., Carey, K., 8 Maisto, S. (1986). The validity of young adult's reports of parental drinking practices. Journal of_Studies .op_Alcohol, 413 433-435. Parker, D.A. 8 Harford, T.C. (1988). Alcohol-related problems, marital disruption and depressive symptoms among Adult Children of Alcohol Abusers in the United States. Journal 2; Studies op Alcohol, 42(4), 306-313. Rimmer, J. (1982). The children of alcoholics: An exploratory study. Children and Youth Services Review, 4, 365-373. Robins, L., West, P., Ratcliff, K., 8 Herjanic, B. (1978). Father's alcoholism and children's outcomes. In F.A. Seixas (Ed.), 92553353 12 alcoholism (Vol.4, pp. 313-327). New York: Grune 8 Stratton. 126 Rogasch, F., Chassin, L. 8 Sher, K.J. (1990). Personality variables as mediators and moderators of family history risk for alcoholism: Conceptual and methodological issues. Journal g£_Studies pp. Alcohol, §l(4), 310-318. Ryan, R.M. 8 Lynch, J.H. (1989). Emotional autonomy versus detachment: Revisiting the vicissitudes of adolescence and young adulthood. Child Development, 49) 340-356. Schuckit, M.A. (1980). Self-rating of alcohol intoxication by young men with and without family histories of alcoholism. Journal 2: Studies pp Alcohol, 413 242-249. Schuckit, M.A. 8 Sweeney, 8. (1987). Substance use and mental health problems among sons of alcoholics and controls. Journal pf Studies pp_Alcohol,‘4§(6), 528-534. Searles,J.S. 8 Windle, M. (1990). Introduction and overview: Salient issues in the children of alcoholics literature. In M. Windle 8 J.S. Searles (Eds.) Children p£_Alcoholics: Critical Perspectives. New York: Guilford Press. Seilhamer, R.A., 8 Jacob, T. (1989). Family factors and adjustment of children of alcoholics. In M. Windle 8 J.S. Searles (Eds.) Children .2; Alcoholic: Critical Perspectives. New York: Guilford Press. Seixas, J. (1982). Children from alcoholic homes. In N. Estes 8 E. Heinemann (Eds.) Alcoholism: Development,_consequences and interventions (pp.193-201). St. Louis, MO: Mosby. Sessa, F.M. 8 Steinberg, L. (1991). Family structure and the development of autonomy during adolescence. Journal p£_Early Adolescence, 14(1), 38-55. 127 Sher, R.J., Walitzer, K.S., Wood, P.K. 8 Brent, E.E. (1991). Characteristics of children of alcoholics: Putative risk factors, substance use and abuse, and psychopathology. Journal 2; Abnormal Psychology, 199(4), 427-448. Spanier, G. 8 Cole, C. (1974). Toward clarification and investigation of marital adjustment. International Journal oi Sociology p£_the Family, 4) 121-146. Steinhausen, H., Gobel, D., 8 Nestle, V. (1984). Psychopathology in the offspring of alcoholic parents. Journal 2; the American Academy p£_Child Poychiatry, 3g, 465-471. Straussner, S., Weinstein, D., 8 Hernandex, R. (1979). Effects of alcoholism on the family system. Health and Social Work, 4) 111- 127. Wallerstein, J.S. 8 Kelly, J.B. (1980). Surviving the Break-up: How children and parents cope with divorce. New York: Basic Books. Weiss, R.S. (1982). Attachment in adult life. In C. Parke 8 J. Stevenson-Hinde (Eds.) The Place p£_Attachment‘ip_Human Behavior. New York: Basic Books. Werner, E.E. (1986) Resilient offspring of alcoholics: A longitudinal study from birth to age 18. Journal pf Studies 22 Alcohol, 41(1), 34-40. West, M.O., 8 Prinz, R.J. (1987). Parental alcoholism and childhood psychopathology. Psychological Bulletin, 102, 204-218. 128 White, R.M., Speisman, J.C., 8 Costos, D. (1983). Young adults and their parents: Individuation to mutuality. In H.D. Grotevant 8 C.R. COOper (Eds.) Adolescent Development ip_the Family: New Directions for Child Development, No. 22. San Francisco: Jossey-Bass. Woititz, J. (1978). Alcoholism and the family: A survey of the literature. Journal p£_Alcohol and Drug_Education, 22, 18-23. Woititz, J.G. (1983). Adult Children of Alcoholics. Deerfield Beech, FL: Health Communications. Wolin, S., Bennett, L., 8 Noonan, D. (1979). Family rituals and the recurrence of alcoholism over generations. American Journal of Psychiatry, 136, 589-593. Wood, B.L. (1987). Children pf_Alcoholism: The Struggle for Self and Intimacy ip_Adult Life. New York: New York University Press. Woodruff, R.A., Guze, S.B., 8 Clayton, P.J. (1972). Divorce among psychiatric outpatients. British Journal pf_Psychiatry, 121, 289- 292. Wright, V.C., Frank, S.J., 8 Pirsch, L.A. Late adolescents' separation/individuation from their parents: The implications of parental alcoholism. Paper in progress. Wright, V.C., Frank, S.J., 8 Pirsch, L.A. (1990). Mediators in the relationship between parent alcoholism and late adolescent separation-individuation pathology. Paper Presented at the meeting of the Society for Research on AdOlescence, Atlanta, GA. YOUDISB, J. 8 Smollar, J. (1985). Adolescent relations with mothers, fatflners, and friends. Chicago: University of Chicago Press. 129 Zucker, R. (1987). The four alcoholisms: A developmental account of the etiologic process. In P.C. Rivers (Ed.) Nebraska symposium op motivation, Vol. 24, Lincoln, Nebraska: University of Nebraska Press. Zucker, R. (in press). Pathways to alcohol problems and alcoholism: A developmental account of the evidence for multiple alcoholisms and for contextual contributions to risk. R. Zucker 8 G.M. Boyd (Eds.) The development 22 alcohol problems: Exploring p bi0psychosocial matrix p£_risk. Rockville, MD: National Institute on Alcohol Abuse and Alcoholism. "IIIIIIIIIIIIIIIII