Control, attachment style, and relationship satisfaction among adult children of alcoholics. (Research).

This study investigated possible differences in need for control, attachment style, and relationship satisfaction between a sample of adult children of alcoholics (ACOAs, 18 males and 22 females) and a sample of adult children of nonalcoholics (ACONAs, 10 males and 30 females). Preliminary analyses revealed that need for control, attachment style, and relationship satisfaction were significantly correlated across the board for both the ACOA and ACONA groups. A multivariate analysis of variance revealed that the two groups differed significantly on the dependent variables of need for control and relationship satisfaction, with the ACOAs reporting a significantly higher need for control and significantly lower relationship satisfaction. A discussion of the theoretical, clinical, and research implications for mental health counselors is also included.

**********

Although alcoholics in the United States number some 8 to 10 million, in reality the disease of alcoholism affects an estimated 30 million others, including family members and offspring (Woodside, 1988). However, it is only in the past decade or so that this peripheral population has been targeted for study by the mental health profession. In particular, several identifiable patterns of interpersonal discomfort and intrapsychic conflict among adult children of alcoholics (ACOAs) have been explored in the existing literature (Domenico & Windle, 1993; Wright & Heppner, 1993). In fact, the past 10 years have yielded a growing number of studies exploring the potential connection between control issues and relationship problems among adult children of alcoholics (Brown, 1988; Knoblauch, 1990).

The literature is consistent in defining ACOAs as adults from a family with alcoholic parent(s), grandparent(s), and/or other family member (Kritzberg, 1990). One of the most reasonable theories for conceptualizing the problem of treating this population views alcoholism as a systemic process in which the entire family is affected. Family systems theory is based on the notion that a feedback mechanism exists that continuously monitors the family state. Thus, children who grow up in alcoholic homes learn to monitor the family climate and engage in behaviors designed to minimize the conflict and chaos that are such a part of the alcoholic family environment. Unfortunately, the ensuing maladaptive cycle keeps family members from confronting significant issues, and the persistent pressure to constantly reassess the balance in the family system, in turn, leads to rigid, controlling behaviors that interfere with individual growth and differentiation and the formation of healthy relationships (Bepko & Krestan, 1985).

Publications by Woititz (1989) and Cermak and Brown (1982) have echoed this theme by suggesting that it is the inconsistency and chaos in alcoholic families that leads to the perpetuation of maladaptive behaviors into adulthood as ACOAs begin to settle into families of their own. As a result of growing up in a dysfunctional environment, the interpersonal functioning of ACOAs is often characterized by a dependence on the approval of others, thereby circumventing the development of a stable sense of self and personal control. In order to fill this interpersonal vacuum, ACOAs may exhibit behaviors shaped by a need for control, denial, rationalization, and compulsivity (Friel, 1988).

Recently, articles have begun to address more directly the issue of control among ACOAs and the ensuing negative effects on the formation of intimate relationships (Shapiro, Weatherford, Kaufman, & Broenen, 1994; Sheridan & Green, 1993). However, no empirical studies have focused on the possible contribution of attachment style in illuminating and clarifying the dynamics of control and relationship issues among this population. For example, attachment theorists assert that the sense of security in any attachment relationship depends on the quality of responsiveness between the relationship partners (El-Guebaly, West, Maticka-Tyndale, & Pool, 1993). Bowlby (1969) and others (Ainsworth, Blehar, Waters, & Wall, 1978; Main, Kaplan, & Cassidy, 1985) have long espoused the belief that the attachment system functions as the cornerstone for adaptive developmental processes. Using this framework to examine how the dynamics in the alcoholic family of origin interfere with the development of the child's personal identity helps to illuminate how the child's sense of self becomes distorted in an effort to maintain the attachment to the alcoholic parent(s). Just as the alcoholic in the family strives to control his or her drinking behaviors, the child endeavors to mediate the family chaos by attempting to control himself or herself and others. In this context, the interpersonal problems of ACOAs may mirror the rigidity and pervasiveness of relational difficulties within the alcoholic family of origin across the developmental years. The failure of the alcoholic caregiver to respond interferes with the child's creation of a secure representational model of self in relation to others; a model of insecurity that is transported into adulthood (El-Guebaly et al.). This continued pattern of insecurity in relationships is then reinforced and reinterpreted by the ACOA in the context of adult attachment relationships as well. Thus, since attachment styles have also been shown to influence adult interpersonal relationships, and dysfunctional interpersonal relationships have been identified as another salient feature of ACOAs, attachment theory may provide an effective tool for exploring control-related issues among this segment of the population.

CONTROL AND RELATIONSHIPS AMONG ACOAS

According to the literature, the control issues of ACOAs are played out individually in the need to dominate and control the environment, primarily the self and relationships in particular, and are highlighted by difficulties with intimacy and trust (Knoblauch & Bowers, 1989). In particular, the need for control seems related to the dynamics in the dysfunctional family of origin. Because the control of self, environment, and others is an all-consuming task, ACOAs experience persistent frustration and an inability to relax. In a study conducted by Cermak and Rosenfeld (1987), ACOAs exhibited symptoms similar to individuals with Post Traumatic Stress Disorder or histories of physical, emotional, and sexual abuse. This tendency to over control and manipulate behavior may put ACOAs at risk for a variety of physiological and psychological disorders that may include highly defensive, self-protective behaviors, inability to deal with feelings, lack of trust and problems with intimacy, dependency on the approval of others, and an inordinate need for control of self, others, and the surrounding environment.

According to Woititz (1989), many ACOAs transport their need for control into adult relationships, developing coping skills along the way in order to mediate the chaos and unpredictability in their lives. Although in the short term these seem adaptive maneuvers, in the long run when carried into adult relationships they become maladaptive (Ackerman, 1987). She also suggests that ACOAs are at a distinct disadvantage in that they have limited access to models for healthy relationships and an even more limited understanding of the key elements necessary for establishing and maintaining adaptive relationships: intimacy, vulnerability, trust, honesty, and mutual sharing. Their inability to express their needs and feelings are based on ingrained patterns of distrust, secretiveness, and fear of intimacy and abandonment. El-Guebaly et al. (1993) add that the interpersonal difficulties of ACOAs are reflective of the relational patterns in the alcoholic family of origin. Over time these patterns become the foundation for negative expectations about establishing and maintaining secure relationships.

Several studies have focused on this need for control as an obstacle to establishing close relationships. In a study on intimacy in ACOA populations, Latham (1988) found that married female ACOAs reported experiencing more difficulty with intimacy, more family of origin dysfunction, and a greater need to take a controlling role in their marriages than married female adult children of nonalcoholics (ACONAs). Heinemann (1989) suggests that the relationships of ACOAs suffer because they experience difficulty dealing with emotions and vulnerability, the precursors of intimacy. For ACOAs, the unpredictable behavior of others leads to stress in the relationship and a feeling of loss of control that harkens back to childhood fears of abandonment and isolation. In Ackerman's (1989) study, daughters of alcoholics reported a significantly higher need for control, overreaction to change, and feelings of over responsibility for others. They also rated themselves higher in difficulty with intimacy, approval and affirmation, and judging themselves harshly as compared to daughters of nonalcoholics. In an open-ended, follow-up survey, 33% of the adult daughters of alcoholics (versus 9% of the adult children of non-alcoholics) reported the greatest parenting issue for them as parents was their "need for control" (Ackerman). They reported taking on too many responsibilities for their children by overprotecting them, had extremely high expectations of their children, and felt responsible for making sure everything in the family was under control. Finally, the results of Knoblauch and Bowers' (1989) research revealed that the higher the need for control reported by ACOAs, the lower the level of satisfaction with the relationship.

ATTACHMENT AND ADULT CHILDREN OF ALCOHOLICS

Attachment theory as conceptualized by John Bowlby (1969) is a homeostatic process, the purpose of which is to ensure the survival of the species. This system is activated when the infant feels threatened or insecure. When this occurs, the child engages in behaviors designed to re-establish contact with the attachment figure. Once contact is regained, the child modifies his or her behavior to ensure and maintain the proximity of the caregiver.

Berman and Sperling (1994) build on Bowlby's (1969) original definition by adding that adult attachment is "the stable tendency of an individual to make substantial efforts to seek and maintain proximity to and contact with one or a few specific individuals who provide the subjective potential for physical and/or psychological safety and security" (p.8). Attachment style refers to "particular internal working models of attachment that determine people's behavioral responses to real or imagined separation and reunion from their attachment figures" (p.11). This tendency, which begins in early childhood, provides the potential for secure relationships throughout the life span and seems to be directly connected to initial attachment experiences. According to attachment theory, early childhood interactions with caregivers lead to the development and accumulation of knowledge and experiences concerning the self, others, and relationships. The infant develops internal working models (IWMs) of the attachment figure(s) based on shared interaction experiences (Bowlby, 1988). These models, in turn, provide the infant with an awareness of the caregiver's absence, determine how the child will respond when the attachment figure returns, and ultimately comprise the foundation for anticipating and interpreting interactions with others across the life span.

Adult attachment styles are characterized by specific variations in internal working models of attachment that theoretically emanate directly from early attachment experiences. The activation of these internal working models with their underlying schemata has been shown to affect behavioral responses by creating a complicated interplay of behaviors between the relationship partners. For instance, recent studies suggest that securely attached adults, much like children, make more realistic attributions for their partners' behaviors, seem more emotionally positive and trusting, and view relationships as supportive and comforting (Feeney, 1999; Miller & Noirot, 1999), whereas insecurely attached adults appear conflicted and perceive relationships as ranging from desirable, but unpredictable, to overtly threatening (Lopez & Brennan, 2000). As the list of parallels between childhood and adulthood attachment dynamics grows so do researchers' beliefs that attachment styles are indeed similar in children and adults. Thus, certain expectations for adult attachment styles may be hypothesized based on preliminary research with children.

It is noteworthy that recent attachment research appears to support the idea that attachment styles are predictive of future social adjustment. In fact, the secure/insecure continuum seems to be a fairly accurate reflection of competency in relationships. For example, individuals who have grown up in homes where the caretaker was inconsistent, unreliable, and/or unresponsive appear to have more interpersonal problems and difficulties with anxiety and depression and, according to one study, scored significantly higher than their securely attached counterparts on almost every personality disorder scale (Brennan & Shaver, 1998). In light of this, attachment theory may serve as a useful paradigm for understanding both adult personality and relationship dynamics.

Furthermore, the notion that internal working models of attachment develop in early childhood and are carried forward into adulthood to color relationships across the life span sets the stage for attachment theory to be used as a valuable tool for exploring relationship problems among ACOAs. Viewed through an attachment-theory lens, the difficulties ACOAs experience are likely reflective of the lack of responsiveness and availability of the alcoholic parent as the child begins to create and assimilate a representational model of the self in relation to others. Accordingly, research utilizing attachment theory may provide a more empirically grounded framework within which to comprehend and address the impact of growing up in an alcoholic family (Brennan, Shaver, & Tobey, 1991; Brown, 1991) and may provide mental health counselors with additional avenues through which to identify, view, and harness possible sources of strength and resilience for ACOAs. This study was designed to explore possible differences among ACOAs and ACONAs in reported need for control, in attachment style, and in satisfaction with relationships in order to investigate the possibility that attachment style affects need for control and relationship satisfaction in ACOAs. To this end, the following specific research hypotheses were explored: (a) that adult children of alcoholics would report a significantly higher need for control, (b) that ACOAs would report a significantly more insecure attachment style, and (c) that ACOAs would report significantly less satisfaction with intimate relationships than adult children of nonalcoholics.

METHOD

Participants

The 80 participants (40 ACOAs, 40 ACONAs) in this study were selected from a convenience sample of 203 volunteer participants from an undergraduate psychology pool at a major Midwestern university. The ACOA group consisted of 18 males and 22 females, ranging in age from 18 to 59 (M = 31.8). The ACONA group consisted of 10 males and 30 females ranging in age from 18 to 57 (M = 30.2). The ethnic breakdown per group was as follows: ACOAs = 85% Caucasian, 10% African American, 2.5% Asian American, and 2.5% Native American. The breakdown of the ACONA group was 80% Caucasian, 5% African American, 7.5% Asian American, 5% Hispanic/Latino, and 2.5% Native American.

Instruments

ACOA status. The participants' ACOA status was determined using two measures. The first, the Children of Alcoholics Screening Test (CAST; Jones 1991) is a widely used instrument for identifying adults with at least one alcoholic parent. It also assesses the severity and impact of parental problem drinking on the individual. For this study, the CAST was utilized only to help delineate group membership. The instrument contains 30 yes/no items such as "Did you ever wish a parent would stop drinking?" A minimum score of 6 is generally considered indicative of an individual who has grown up in an alcoholic environment. A Spearman-Brown split-half reliability coefficient of .98 was computed with various samples of latency-age, adolescent, and adult children of alcoholics (Jones). A study by Clair & Genest (1992) reported a test-retest reliability coefficient of .88 over a 2-month interval. Several validity studies on the CAST have revealed that self-identified children of alcoholics consistently scored six or greater on the instrument (Carroll, 1991; Jones, 1991; Pilat & Jones, 1984/1985). In correlating ACOA versus ACONA group membership with total CAST scores, Pilat & Jones reported a validity coefficient of .78 (p < .0001).

Several studies in the literature have demonstrated that ACOAs can also be accurately identified by using a single, objective question (Berkowitz & Perkins, 1988; DiCicco, Davis, Travis, & Orenstein, 1983/1984; DiCicco, Davis, & Orenstein, 1984). Thus, the second measure of ACOA status in this study consisted of a single question ("Do you believe that either one or both of your parents have/had a drinking problem?") used in conjunction with the CAST to confirm parental problem drinking.

An additional questionnaire was included to provide basic demographic data. This data included age, gender, ethnicity, family income, and so forth.

The Desirability of Control Scale (DC). The Desirability of Control Scale (Burger & Cooper, 1979) is a 20-item instrument designed to measure individual differences in desire for control over life events. A sample item from this scale reads: "I prefer situations where I have a lot of control over what I do and when I do it." Subjects are asked to respond to each item using a 7-point Likert rating ranging from "This statement doesn't apply to me at all" to "This statement always applies to me." Scores are summed to reflect overall level of desire for control, with higher scores indicating a greater need for control. Kuder-Richardson 20 reliabilities for the DC scale in studies with students from a college population ranged from .80 to .81, and test-retest reliability was .75 (Burger & Cooper). A study of discriminant validity compared the DC to Rotter's Internal-External Locus of Control Scale (Rotter, 1966) and revealed a low negative correlation of -.19 suggesting that the two scales measure different control constructs.

Attachment Style Questionnaire (ASQ). The Attachment Style Questionnaire is a 40-item instrument that provides scores on 5 subscales: Confidence in Self & Others, Need for Approval, Preoccupation with Relationships, Discomfort with Closeness, and Relationships as Secondary. The score on the Confidence subscale provides the measure of secure attachment, while the sum of scores on the remaining four subscales constitutes the measure of insecure attachment. The instrument was developed and tested by Feeney, Noller, and Hanrahan (1994) to provide a broader-based measure of the underlying dimensions of adult attachment and to be used with adolescents and individuals who have had more limited experiences with romantic relationships. One of the sample items from the secure subscale reads: "Overall, I am a worthwhile person." Sample items from the subscales measuring insecure attachment include: "I find it hard to trust other people" and "I prefer to keep to myself." The underlying structure of the ASQ is based on Hazan and Shaver's (1987) secure, anxious/ambivalent, and avoidant attachment styles and uses a 6-point Likert scale from "Totally disagree" to "Totally agree." Coefficient alphas for the five subscales ranged from .76 to .84. Test-retest reliability coefficients over a 10-week period ranged from .67 to .78. In order to test the validity of the ASQ, Feeney et al. correlated scales from their new measure with the three Likert ratings based on Hazan and Shaver's original forced-choice measure. The Confidence subscale of the ASQ was positively correlated with the Hazan and Shaver secure scale and negatively correlated with the four subscales measuring insecurity (i.e., Discomfort with Closeness, Need for Approval, Preoccupation with Relationships, and Relationships as Secondary). These results suggest that the ASQ taps constructs similar to those assessed by the Hazan and Shaver measure. As a further validity check, they ran analyses of variance (ANOVAs) using the original forced-choice measure as the independent variable. Again, their findings indicate that the ASQ measures attachment styles similar to those conceptualized by Hazan and Shaver's original measure.

Research on the Attachment Style Questionnaire has shown that the most consistent and useful distinction between attachment styles looks at the secure (Confidence subscale) measure and the combination of the four subscales that comprise the insecure measure (Discomfort with Closeness, Need for Approval, Preoccupation with Relationships, and Relationships as Secondary; Brennan et al., 1991). The insecure measure can be further broken down into two separate subscales: Anxious/ambivalent and Avoidant. In conjunction with previous research on the ASQ and given our hypothesis that ACOAs would report being more insecurely attached than ACONAs, the data were analyzed based on the two primary (secure/insecure) factors of the ASQ. For this study, coefficient alphas for the secure and insecure subscales were .83 and .84 respectively.

The Relationship Satisfaction Questionnaire (RSAT). The Relationship Satisfaction Questionnaire is a 13-item instrument developed by Burns and Sayers (1992) to measure satisfaction in a variety of relationship areas (e.g., Communication and Openness, Intimacy and Closeness, Overall Satisfaction with the Relationship, etc.). Subjects indicate their degree of relationship satisfaction in each area using a 7-point Likert rating from 0 (very dissatisfied) to 6 (very satisfied). Scores are then summed to indicate total satisfaction, with higher scores corresponding to reports of greater relationship satisfaction. The RSAT has demonstrated good internal consistency with a coefficient alpha of .97, and test-retest correlations at 6 weeks were .72 (Heyman, Sayers, & Bellack, 1994). Convergent validity was also established with the Marital Adjustment Scale (r = .80; Burns & Sayers, 1988) and the Quality of Marriage Index (r = .91; Heyman et al.). The RSAT also demonstrated discriminant validity (r = -.31 to -.51) with the psychopathology subscales of the Symptom Checklist-90-Revised (SCL-90-R). Factor analysis further indicates that the RSAT measures a single factor of relationship satisfaction among both men and women.

Procedure

Approximately 200 students from a large Midwestern university were administered the CAST and a demographics questionnaire containing a single-question item as to whether or not either or both of their parents were alcoholic. This administration was conducted as part of a group-testing program for screening purposes that allows students to receive class research credits in exchange for completion of specified instruments.

The following criteria were utilized in order to operationalize the ACOA variable. As per previous research, a score of 6 or higher on the 30-item CAST was used to indicate the presence of parental alcoholism. In order to correct for possible false positives on the CAST participants were also asked to respond to a single question asking whether one or both of their parents were alcoholic. This CAST score combined with the affirmative endorsement of the single-item demographic question as to whether or not one or both parents were alcoholic was used to indicate the presence of parental alcoholism (ACOA group). Conversely, a score of less than 6 on the CAST and an accompanying negative endorsement of parental problem drinking was used to indicate the absence of parental alcoholism (ACONA group). It should be noted that all participants in this study who scored a 6 or greater on the CAST also endorsed the confirmatory single-question item about parental alcoholism in the affirmative. Forty-two potential ACOA participants were identified via group testing. When contacted about further participation, 40 individuals agreed to complete further research for course credit. Since there were considerably more than the required number of individuals comprising the ACONA group, 40 participants were randomly selected and contacted as to the time and place to meet to complete further research. The above-mentioned informed consent procedures were repeated for participants who elected to complete the additional research protocol. During this subsequent research phase participants were administered three other inventories, including the Attachment Style Questionnaire, the Desirability of Control Scale, and the Relationship Satisfaction Questionnaire.

RESULTS

Descriptive statistics for both the ACOA and ACONA groups were calculated and preliminary analyses were conducted to examine possible differences between ACOAs and ACONAs on selected demographic variables. Subsequent t-tests and chi-square analyses revealed the two groups did not differ significantly on gender, age, family income, parental education level, or ethnicity.

Preliminary correlations were run in order to examine whether need for control, attachment style, and relationship satisfaction would be significantly correlated. The results showed that need for control, attachment style, and relationship satisfaction were significantly correlated across the board for both the ACOA and the ACONA groups (see Table 1).

A MANOVA was used to investigate possible differences between ACOAs and ACONAs on the variables of need for control, attachment style, and relationship satisfaction. A conventional statistical significance level of p < .05 was utilized for this process. Results from this analysis revealed a significant group membership difference [Wilks' lambda = 0.72, F (4, 75) = 7.13, p < .0001, [[eta].sub.[rho].sup.2] = .28]. Separate univariate F-tests on the individual variables of need for control, attachment style, and relationship satisfaction revealed significant results only on the variables of need for control and relationship satisfaction [F (1, 78) = 11.12, p < .001, [[eta].sub.[rho].sup.2] = .13 and F (1,78) = 14.69, p < .0001, [[eta].sub.[rho].sup.2] = .16 respectively]. Means, standard deviations, and F and p values for both groups on these dependent measures appear in Table 2. There was no significant group membership difference for the variable of attachment style [F (1,78) = 2.10, p = .15, [[eta].sub.[rho].sup.2] = .03].

DISCUSSION

The primary purpose of this investigation was to explore differences between adult children of alcoholics and adult children of nonalcoholics on the variables of need for control, attachment style, and relationship satisfaction. It was hypothesized that ACOAs would exhibit a significantly higher need for control, a significantly more insecure attachment style, and significantly less relationship satisfaction than ACONAs. Analyses revealed support for the hypotheses that ACOAs would report significantly higher need for control and significantly less relationship satisfaction than ACONAs. It is interesting to note that while ACOAs reported being more insecurely attached than their ACONA counterparts, the mean difference between the two groups was not statistically significant. Possible explanations for the lack of significance among ACOAs on the variable of attachment style may lie in the fact that, as previous literature has suggested (e.g., Sher, Walitzer, Wood, & Brent, 1991), college ACOAs may represent a more resilient subset of the ACOA population in that they may have developed particular qualities that have helped ameliorate some of the maladaptive effects of parental alcoholism. Another possibility may be the confounding variable of family dysfunction other than alcoholism. Other research (e.g., Jenkins, Fisher, & Harrison, 1993) has demonstrated that individuals from dysfunctional families often present with issues very similar to ACOAs. Thus, it is a distinct possibility that at least some of the ACONA participants came from dysfunctional, although not alcoholic, homes. This fact may have masked or at least minimized important significant differences between the two groups. Finally, the Attachment Style Questionnaire, a relatively new instrument, may lack discriminatory power, especially since some researchers have suggested that attachment styles may not be distinctly manifested without some precipitating event being perceived as threatening to personal security.

Limitations

It is obvious that the use of self-report measures as well as the lack of collateral informants (i. e., other family members who could corroborate the report of family alcoholism) may well have contributed to bias in this research. Also, this study used a convenience sample of college students which no doubt makes generalizability to other populations tentative at best.

While the collection of data for the ACONA (control) group allowed for randomization of participants, the difficulty in obtaining a sufficient number of ACOA participants did not permit such a procedure. Thus, the hypotheses suggested herein must be viewed as tentative and interpreted with caution. They do, however, provide new food for thought where ACOAs are concerned as well as a new platform from which to launch further confirmatory research.

Strengths

This investigation, while limited in large part by the exploratory nature of its design, does build on and contribute to previous research in the area in several important ways. First, while a substantial body of literature exists that explores a variety of ACOA issues, only a small fraction attempts to explore the implications or value of existing theory in an effort to develop new lenses through which to examine how growing up in an alcoholic environment may manifest itself differentially among ACOAs. Second, there has been very limited research investigating connections between attachment style and ACOA functioning, and there has been no published empirical research exploring the possible synergistic effects of need for control, attachment style, and relationship satisfaction among this population. Third, this study was designed to fill in some gaps in previous research by using multiple measures to screen for and confirm the existence of parental alcohol usage and attempted to collect information to add to the existing research base by including specific questions on family functioning. Fourth, the use of a newer, more continuous measure of attachment was utilized in lieu of one of the older, categorical measures. Finally, Black (1981) and others have suggested that the consequences of parental alcoholism may not manifest themselves until ACOAs are in their mid-20s, perhaps explaining why several previous studies designed to compare the functioning of ACOAs to their non-ACOA counterparts using college student populations revealed no significant differences between the two groups. In contrast, the present study is based on data collected from a nontraditional university population ranging in age from 18 to 59 with an average age of approximately 30 years (Median age = 30.25).

Theoretical and Clinical Implications

The premise of examining attachment theory's potential contribution to clarifying the ACOA issues of need for control and relationship dissatisfaction may have paved the way for a variety of new theoretical and clinical possibilities. When working with ACOAs, mental health counselors may benefit from an increased awareness of how attachment style influences the way these clients' perceive and respond in relationships. For the most part, we learn to relate to others via our family of origin. Those family dynamics, combined with individual proclivities, determine how we navigate relationships. It is obvious that growing up in an alcoholic, or for that matter any dysfunctional family environment, may do much to set in motion certain problematic patterns or styles of interacting that are carried forth into adulthood to color relationships. Thus, a three-tiered approach beginning with a primary prevention program based on a recognition of these problematic patterns should begin as early as possible in order to mediate and reverse the maladaptive sequence. In using psychoeducational programs, mental health counselors should involve individual, family, and group skill-building exercises. This type of program would consist of community, school, medical and mental health, recreational and social, and peer-based prevention settings and would ideally be part of a larger wellness-based approach emphasizing healthy overall lifestyles. Secondary treatment planning would focus on mental health counselors educating participants about problematic behaviors and promote the development of more adaptive coping skills. Individual and group therapy settings should provide affirmation of the experience of growing up in an alcoholic home, offer alcohol education, and emphasize the identification and expression of feelings and the development of healthy social interactions, and teach problem-solving and coping skills. Since alcoholism permeates family functioning, mental health counselors' tertiary treatment would focus on family systems-related therapy and should provide family-skills training as well as opportunities for working through unresolved family issues.

For college populations, mental health counselors could offer workshops and seminars to educate students about ACOA issues, attachment and communication styles, and relationship dynamics and to raise self-awareness, increase coping skills, and reduce anxiety and insecurity.

Attachment theory also offers potential for assisting mental health counselors with pre-marital and couples' counseling. Specifically, it can be used to enhance couples' awareness of relationship dynamics and to help them to recognize maladaptive processes and construct more open lines of communication, which have been shown to appreciably increase relationship satisfaction (Johnson & Greenberg, 1995).

Future Research

As for the direction of future research, a sophisticated qualitative investigation could do much to tease out possible moderating and confounding variables among ACOA populations and might provide a refined understanding of adult attachment theory and its relationship to ACOA functioning. An examination of the influence of social and cultural underpinnings of attachment utilizing perhaps multiple measures of attachment and relationship satisfaction might provide even more clues about the role this construct plays in ACOA functioning and relationship dynamics. Future studies could be designed to measure self-perceptions of attachment combined with enacted attachment scenarios to investigate how adults react to a variety of processes in romantic relationships. This type of study would be invaluable in that it would provide information on how and when the attachment system is eventually activated and whether, as according to theory, individual experiences periodically and incrementally modify system functioning.

Future longitudinal research designed to utilize multivariate statistical models could examine the obvious multiple causes and effects of parental alcoholism. Refined methods could tease out individual factors contributing to both psychological distress and resilience, which are critical for operationally defining parental alcoholism as well as for developing and streamlining therapeutic interventions that can be individually tailored to target respective client needs. This task of teasing out individual contributory factors will no doubt be complicated by the reality that it is likely the summative effects of alcoholism precipitate long-term adjustment difficulties in ACOAs. In this respect, developing and utilizing lifespan developmental models for viewing the continued process may help identify critical periods in development when ACOAs are more vulnerable to parental alcoholism effects.

In addition, studies designed to look at within-group differences among ACOAs as well as individuals from other dysfunctional environments would be helpful in order to shed more light on whether it is the dysfunctional environment per se or more specifically the alcoholic experience that wields the greatest impact on issues of control and relationship satisfaction among ACOAs. Identifying, categorizing, and comparing subgroups of ACOAs based on such criteria as gender of the alcoholic parent, whether or not the primary caregiver was alcoholic, and length of exposure to problematic drinking behaviors may also provide valuable information as to what factors in the alcoholic environment may either buffer or contribute to increased risk among members of this population. Finally, longitudinal research examining developmental issues would also add clarity to this literature.

In spite of any implications suggested by this study, we cannot by any means assume that all ACOAs translate their developmental experiences in an alcoholic environment in the same ways. It is obvious that individuals are differentially affected by their experiences and that their psychological adjustment must be multi-determined and not merely seen as the product of family alcoholism, bus, continued identification and classification of the contributing effects of control, attachment, and relationship dynamics with an emphasis on quality longitudinal research as to how these experiences contribute to developing positive coping skills may help mental health counselors to develop and utilize treatment strategies designed to target the specific risk factors more characteristic of this particular population.

Table 1. Correlations of Dependent Variables for ACOAs (top) and
ACONAs (bottom)

Variable         ASQ(secure)   ASQ(insecure)   RSAT

DC                 -.50 **         .58 **      -.44 **

ASQ (secure)                      -.58 **       .73 **

ASQ (insecure)                                 -.61 **

DC                 -.37 **         .31 *       -.34 *

ASQ (secure)                       .36 **       .76 **

ASQ (insecure)                                 -.60 **

Note. DC = Desirability of Control Scale; ASQ = Attachment Style
Questionnaire; RSAT = Relationship Satisfaction Questionnaire.

* p < .05. ** p <.01.

Table 2. Univariate Comparisons of Dependent Variables of Interest

                          ACOAs                       ACONAs

Variable            M      SD        M      SD       F       p

DC                95.93   11.51    88.58    8.71   11.12   <.001

ASQ (secure)      32.25    4.74    33.88    5.10    2.18    <.14

ASQ (insecure)   107.08   19.96   101.38   14.84    2.10    <.15

RSAT              51.25   12.35    61.30   11.07   14.69   <.000

Variable         [[eta].sub.[rho].sup.2]

DC                         .13

ASQ (secure)               .03

ASQ (insecure)             .03

RSAT                       .16

Note. DC = Desirability of Control Scale; ASQ = Attachment Style
Questionnaire; RSAT = Relationship Satisfaction Questionnaire.

REFERENCES

Ackerman, R. J. (1987). Same house, different home: Why adult children of alcoholics are not all the same. Pompano Beach, FL: Health Communications.

Ackerman, R. J. (1989). Perfect daughters. Deerfield Beach, FL: Health Communications.

Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment: A psychological study of the Strange Situation. Hillsdale, NJ: Erlbaum.

Bepko, C., & Krestan, J. (1985). The responsibility trap: A blueprint for treating the alcoholic family. New York: The Free Press.

Berkowitz, A., & Perkins, H. W. (1988). Personality characteristics of children of alcoholics. Journal of Consulting and Clinical Psychology, 56, 206-209.

Berman, W. H., & Sperling, M. B. (1994). The structure and function of adult attachment. In M. D. Sperling & W. H. Berman (Eds.), Attachment in adults: Clinical and developmental perspectives (pp. 1-30). New York: Guilford.

Black, C. (1981). It will never happen to me! Denver: M. A. C. Printing & Publications Division.

Bowlby, J. (1969). Attachment and loss (Vol. 1). New York: Basic.

Bowlby, J. (1988). Developmental psychiatry comes of age. American Journal of Psychiatry, 145, 1-10.

Brennan, K. A., & Shaver, P. R. (1998). Attachment styles and personality disorders: Their connections to each other and to parental divorce, parental death, and perceptions of parental caregiving. Journal of Personality, 66, 835-878.

Brennan, K. A., Shaver, P. R., & Tobey, A. E. (1991). Attachment styles, gender and parental problem drinking. Journal of Social and Personal Relationships, 8, 451-466.

Brown, S. (1988). Treating adult children of alcoholics: A developmental perspective. New York: Wiley.

Brown, S. (1991). Treating adult children of alcoholics: A developmental perspective. Co-dependence and ACAs: Theoretical expansion from an interpersonal perspective, Paper presented at National Consensus Symposium on Children of Alcoholics and Co-dependency, Airlie, VA.

Burger, J. M., & Cooper, H. M. (1979). The desirability of control.

Motivation and Emotion, 3, 381-393.

Burns, D. D., & Sayers, S. L. (1988). Cognitive and affective components of marital satisfaction: Development and validation of a brief relationship satisfaction scale. Unpublished manuscript.

Burns, D. D., & Sayers, S. L. (1992). Development and validation of a brief relationship satisfaction scale. Unpublished manuscript.

Carroll, S. (1991). Personality characteristics of adult children of alcoholics (Doctoral dissertation, Southern Illinois University at Carbondale, 1989). Dissertation Abstracts International, 51, 3556B.

Cermak, T. L., & Brown, S. (1982). Interactional group therapy with adult children of alcoholics. International Journal of Group Psychotherapy, 30, 375-389.

Cermak, T. L., & Rosenfeld, A. (1987). Therapeutic considerations with adult children of alcoholics. Advances in Alcohol and Substance Abuse, 6(4), 17-32.

Clair, D. J., & Genest, M. (1992). The Children of Alcoholics Screening Test: Reliability and relationship to family environment, adjustment, and alcohol-related stressors of adolescent offspring of alcoholics. Journal of Clinical Psychology, 48, 414-420.

DiCicco, L., Davis, R., & Orenstein, A. (1984). Identifying the children of alcoholics from survey responses. Journal of Alcohol and Drug Education, 30, 1-17.

DiCicco, L., Davis, R., Travis, J., & Orenstein, A. (1983/4). Recruiting children from alcoholic families into a peer education program. Alcohol Health and Research World, 8, 28-34.

Domenico, D., & Windle, M. (1993). Intrapersonal and interpersonal functioning among middle-aged female adult children of alcoholics. Journal of Consulting and Clinical Psychology, 61, 659-666.

El-Guebaly, N., West, M., Maticka-Tyndale, E., & Pool, M. (1993). Attachment among adult children of alcoholics. Addiction, 88, 1405-1411.

Feeney, J. A. (1999). Adult romantic attachment and couple relationships. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (pp. 355-377). New York: Guilford.

Feeney, J. A., Noller, P., & Hanrahan, M. (1994). Assessing adult attachment. In M. D. Sperling & W. H. Berman (Eds.), Attachment in adults: Clinical and developmental perspectives. New York, Guilford.

Friel, J. C. (1988). Adult children: The secrets of dysfunctional families. Pompano Beach, FL: Health Communications.

Hazan, C., & Shaver, P. R. (1987). Romantic love conceptualized as an attachment process, Journal of Personality and Social Psychology, 59, 270-280.

Heinemann, M. A. (1989). Perceived levels of autonomy and intimacy in the family of origin and current functioning of adult children of alcoholics. Unpublished masters thesis. Texas Women's University, Denton.

Heyman, R. E., Sayers, S. L., & Bellack, A. S. (1994). Global marital satisfaction versus marital adjustment: An empirical comparison of three measures. Journal of Family Psychology, 4, 432-446.

Johnson, S. M., & Greenberg, L. S. (1995). The emotionally focused approach to problems in adult attachment. In N. S. Jacobson & A. S. Gurman (Eds.), Clinical handbook of couple therapy (pp. 121-141). New York: Guilford.

Jenkins, S. J., Fisher, G. L., & Harrison, T. C. (1993). Adult children of dysfunctional families: Childhood roles. Journal of Mental Health Counseling, 15, 310-319.

Jones, J. W. (1991). The children of alcoholics screening test manual. Chicago: Camelot.

Knoblauch, D. L. (1990). The theoretical and treatment implications of high-risk variables for adult children of alcoholics. Journal of College Student Psychotherapy, 5(1), 87-101.

Knoblauch, D. L., & Bowers, N. D. (1989). A therapeutic conceptualization of adult children of alcoholics. Journal of College Student Psychotherapy, 4(1), 37-52.

Kritzberg, W. (1990). Healing together: A guide to intimacy and recovery for co-dependent couples. Deerfield Beach, FL: Health Communications.

Latham, M. S. (1988). Relationship patterns of female offspring of alcoholics: An examination of intimacy and individuation in marriage. Dissertation Abstracts International, 49(8), 3446.

Lopez, F. G., & Brennan, K. A. (2000). Dynamic processes underlying adult attachment organization: Toward an attachment theoretical perspective on the healthy and effective self. Journal of Counseling Psychology, 47, 283-300.

Main, M., Kaplan, N., & Cassidy, J. (1985). Security in infancy, childhood, and adulthood: A move to the level of representation. Monographs of the Society for Research in Child Development, 50(1&2), 66-104.

Miller, J. B., & Noirot, M. (1999). Attachment memories, models, and information processing. Journal of Social and Personal Relationships, 16, 1147-173.

Pilat, J. M., & Jones, J. W. (1984/1985). Identification of children of alcoholics: Two empirical studies. Alcohol Health and Research World, 9, 27-33, 36.

Rotter, J. B. (1966). Generalized expectations for internal versus external control of reinforcement. Psychological Monographs, 80(1), (Whole No. 609).

Shapiro, D. H., Weatherford, V., Kaufman, E., & Broenen, R. E. (1994). A control profile of adult children of alcoholics: A preliminary investigation. American Journal of Alcohol Abuse, 20, 247-262.

Sher, K. J., Walitzer, K. S., Wood, P. K., & Brent, E. E. (1991). Characteristics of children of alcoholics: Putative risk factors for substance use and abuse, and psychopathology. Journal of Abnormal Psychology, 100, 427-448.

Sheridan, M. J., & Green, R. G. (1993). Family dynamics and individual characteristics of adult children of alcoholics: An empirical analysis. Journal of Social Service Research, 17(1-2), 73-97.

Woititz, J. G. (1989). The self-sabotage syndrome: Adult children in the workplace. Deerfield Beach, FL: Health Communications.

Woodside, M. (1988). Research on children of alcoholics: Past and future, British Journal of Addiction, 83, 785-792.

Wright, D. M., & Heppner, P. P. (1993). Examining the well-being of nonclinical college students: Is knowledge of the presence of parental alcoholism useful? Journal of Counseling Psychology, 40, 324-334.

Denise Beesley, Ph.D., is an assistant professor, and Cal D. Stoltenberg, Ph.D., is a professor. Both are with the Department of Educational Psychology at the University of Oklahoma, Norman. E-mail: [e-mail])