Addictive Behaviors. Vol. I 7, pp. I9 I - 199. 1992 Printed in the USA. All rights reserved.
0306-460319’ $5.00 + .@I Copyright D 1992 Pergamon Press Ltd.
PERSONAL AND PARENTAL PROBLEM PROBLEILI-SOLVING PERFORMANCE
DRINKING: EFFECTS ON AND SELF-APPRAISAL
STACEY L. SLAVKIN, RICHARD G. HEIMBERG, CHERYL D. WINNING, and ROBERT J. McCAFFREY University at Albany, State University of New York Abstract - This study examined the problem-solving performances and self-appraisals of problem-solving ability of college-age subjects with and without parental history of problem drinking. Contrary to our predictions, children of problem drinkers (COPDs) were rated as somewhat tnore effective in their problem-solving skills than non-COPDs. undermining prevailing assumptions about offspring from alcoholic households. While this difference was not large and was qualified by other variables. subjects’ own alcohol abuse did exert a detrimental effect on problem-solving performance, regardless of parental history of problem drinking. However. a different pattern was evident for problem-solving self-appraisals. Alcohol-abusing non-COPDs saw themselves as eficlire problem-solvers while alcohol-abusing COPDs appraised themselves as poor problem-solvers. In addition, the self-appraisals of alcohol-abusing COPDs were consistent with objective ratings ofsolution effectiveness (i.e., they were both negative) while alcohol-abusing non-COPDs were overly positive in their appraisals. opposing the judgments of trained raters. This finding suggests that the relationship between personal alcohol abuse and self-appraised problem-solving abilities may differ as a function of parental history of problem drinking. Limitations on the generalizability of findings are addressed.
The ability to resolve personal problems has been related to a variety of indices of psychological health (Tisdell & Lawrence, 1986). Among maladjusted children, for example, deficiencies in the quantity and quality of generated problem solutions have been related to clinical ratings ofadaptive functioning (Shure & Spivack, 1972). Similar findings have been reported for emotionally disturbed adolescents (Platt, Spivack, Altman, Altman, & Peizer, 1974) and adolescent and adult psychiatric patients (Platt & Spivack, 1972). In addition, proficient problem-solving skills may serve a stress-buffering role. Nezu and Ronan’s (1988) recent study revealed that effective problem-solving skills moderate depressive symptoms for individuals experiencing negative life events. Though the literature consistently documents poor problem-solving skills in emotionally disturbed and pathological populations, exactly how these deficits contribute to emotional and behavioral maladjustment remains unclear. In an effort to better understand this process, investigators have begun to examine the role of higher-order or meta-cognitive variables which may influence the execution of effective problemsolving skills. One such variable is the self-appraisal of one’s problem-solving effectiveness (Heppner & Petersen, 1982). Problem-solving self-appraisals have been related to behavioral. cognitive, and affective components of the coping process (Heppner & Anderson, 1985; Heppner, Baumgardner, &Jackson, 1985). Self-appraised effective problem-solvers (i.e., those who perceive their problem-solving efforts as effective in modifying problems) are more motivated to solve problems, more systematic and persistent, have higher expectations Portions of this paper were presented at the annual meeting ofthe Association for Advancement of Behavior Therapy, New York, NY, November 1988. Requests for reprints should be sent to Richard Heimberg, Psychology Department, University at Albany, State University of New York, 1400 Washington Avenue, Albany, NY 12222. 191
ofsuccess, and generate more alternative solutions than self-appraised ineffective problem-solvers. This line of research on problem-solving self-appraisals has involved functional populations classified as positive or negative in their self-appraisals by Heppner and Petersen’s (1982) Problem Solvings Inventory (PSI). An equally informative strategy might involve studying individuals who have been less effective at negotiating life’s problems and daily hassles. For example. it is documented that adult alcoholics are deficient in problem-solving skills (Intagliata, 1978). Alcoholics’ self-reports ofstrategies ty.pically employed in interpersonal problem-solving situations have been scored as significantly less competent than those of nonalcoholics (Patterson, Parsons, Schaeffer, 6c Errico, 1988). It has been suggested, however, that alcoholics’ self-appraisals may not mirror their problem-solving deficits (Larson, 1986). In fact, in a recent study of recovering alcoholics (Larson & Heppner, 1989), problem-solvings self-appraisals overestimated adjustment and alcohol use as measured by clinical ratings. In a similar vein, the present study examined the problem-solving performances and self-appraisals of children of problem drinkers, another population with suspected problem-solving deficiencies. Support for this conjecture is derived from both clinical and empirical literature on children of alcoholics (COAs). Related studies suggest that being a COA confers risk for a variety of unfavorable outcomes on offspring including alcoholism, impaired cognitive/intellectual functioning, anxiety, and depression (Parker & Harford. 1988; West & Prinz, 1987). While problem-solving performance and self-appraisal have not been evaluated directly in the offspring of problem drinkers. research on the preferred coping styles of COAs suggests that they see problems as less under their control and rely more on emotion-focused coping than subjects without fami1y:histot-y of alcoholism (Clair & Genest, 1987). Thus. COAs may be impoverished in thetr repertoire of potential problem-solving behaviors and rely more heavily on emotional remedies (e.g., seeking social support, blaming self, avoidance) to problem situations. In addition to influencing the development of problem-solving skills, the alcoholic family environment may affect COAs’ problem-solving self-appraisals. Alcoholic parents may be inconsistent in their evaluations of children’s efforts (Woititz, 1983). creating confusion in offspring. While drinking, parents may be belittling and unable to offer constructive criticism or advice. These speculations are supported by Clair and Genest’s ( 1987) finding that COAs receive less informational support (which may facilitate effective problem-solving) than non-COAs. Alternatively, during sober periods, when attempts to reconcile with family members are likely, alcoholic parents may encourage children’s problem-solving. Thus, COAs may integrate contradictory evaluations of their problem-solving abilities when forming self-appraisals. Such inconsistencies may manifest themselves as problem-solving performance/appraisal discrepancies. In Larson and Heppner’s (1989) study, problem-solving performance quality was operationalized as the number of items endorsed on a list of personal problems, with low numbers taken as an index of problem-solving. The present study included a more direct evaluation of problem-solving performance. Subjects generated possible solutions to hypothetical problems, and objective raters judged their quantity and quality. While others have noted that hypothetical problem-solving may not correspond exactly with “real-life” behavior (Heppner & Krauskopf, 1987) vve believe that this strategy provides an adequate analogue for the purposes of this investigation. as well as a more precise examination of components of the problem-solving process (i.e., generation of
Personal
and parental
problem
drinking
alternative solutions, appraisal of solution suitability) (D’Zurilla & Goldfried, than that provided by Larson and Heppner’s (1989) methodology. hl E T H 0
I93
197 1)
D
Experimental design The present study was a 2 (Positive Parental History of Problem Drinking vs. Negative Parental History of Problem Drinking) X 2 (Alcohol-abusing vs. Non-alcohol abusing) X 2 (Male vs. Female) X 2 (Firstborn vs. Laterborn) factorial design. Previous research suggests that having alcoholic parents places sons at greater risk for dependent problem drinking than daughters (Goodwin, Schulsinger, Knop, Mednick, & Guze. 1977: Parker & Harford, 1988). Furthermore, clinical writings on the role structure of alcoholic families (Wegscheider & Wegscheider, 1978; Woititz, 1983) imply that firstborn COAs typically assume caretaking responsibilities, presumably defined by effective problem-solving. Given these suggestions, the effects of gender and birth order were also considered in the present study. All subjects completed the assessment battery described below. Subjects Subjects were 85 women and 67 men enrolled in introductory psychology courses at the University at Albany, State University of New York, who earned credit toward fulfillment of course requirements for their participation. Informed consent was given for participation in the present study. Subject classification. Of 772 students who completed initial screening procedures, 152 were recontacted and agreed to participate in the present study. An attempt was made to recontact as many children of problem drinkers (COPDs) as possible because they constituted only 10% of the sample. Children of nonproblem drinkers were randomly selected from the contact list. All students completed the following classification measures: The .\lichigan Alcoholism Screening Test (MAST) (Selzer, 1971). The MAST is a brief screening test for the detection of alcoholic drinking and is the standard measure for this purpose in the field. A cutoff score of 5 is traditionally interpreted to represent alcoholic drinking and was required for classification as an alcohol abuser in the current study. A total of 69 subjects (45.4%) were classified as “abusing” and the remaining 83 subjects (54.6%) were classified as “nonabusing.” The Short MASTs for Mother and Father (F-SMAST & M-SMAST) (Sher & Descutner, 1986). These scales are abbreviated versions of the MAST reworded to inquire about parental drinking history and related family issues. Reliability data have been reported by Sher and Descutner (1986) and several additional studies provide validational support for these measures (Benson & Heller, 1987; Levenson, Oyama, & Meek, 1987; Newlin, 1985). In each of these studies, the parental MASTS were used for subject classification. In addition, Sher (personal communication, November 4, 1989) reports that scores on these measures correspond well to diagnostic information derived from selected sections of the Family History-Research Diagnostic Criteria interview (FHRDC; Endicott, Andreasen, & Spitzer, 1978), a structured clinical interview concerning family history of alcohol and drug abuse, antisocial personality, and depression, as well as other forms of psychopathology. In the present study, a cutoff score of 5 was used to
194
STACEY
L. SLAVKIS
et al
identify COPDs, a conservative criterion relative to studies where this demarcation point was used to classify subjects as offspring of alcoholics (Sher & Descutner. 1986). In our sample, 61 (40%: 18 alcohol-abusing; 43 nonabusing) of the subjects were COPDs and 9 1 (60%; 5 1 abusing; 40 nonabusing) were not. We prefer the label of COPD to that of COA because we were unable to conduct diagnostic studies with parents. Demographic informafion. All subjects completed a demographic sheet requesting their birth order (i.e., first or laterborn). Ofsubjects studied, 6 1 (40%) were firstborn and 9 1 (60%) were laterborn. Dependent measures The Problem Assessment Schedule (PAS). There are few descriptions of behavioral assessment in the problem-solving literature. Krasnor and Rubin (198 1) suggest that behavioral problem-solving assessment should involve obsenation of actual problems as they arise in the individual’s daily social interactions. Clearly, such in vivo assessment would be desirable and appropriate in clinical settings. Hovvever, given the problems inherent in uncontrolled observational environments, we chose to provide analogue situations deemed relevant to college students. The PAS (developed for the present study) yields both a subject-generated quantitative measure (i.e., the number of solutions generated) and a qualitative measure (i.e., observer ratings of solution adequacy), two central dimensions of effective problem-solving behavior (D’Zurilla & Goldfried, 1971). The PAS presents four problem situations germane to college students, two interpersonal and two noninterpersonal situations. Interpersonal situations include: 1. You watch your closest friend become increasingly involved with someone who doesn’t treat him/her very vvell. How do you respond? 2. Your boyfriend’s/girlfriend’s roommate expresses an interest in you. How do you respond? Noninterpersonal
situations include:
3. You are home alone and find that the basement of your apartment has six inches of water in it. How do you respond? 4. The end of the month arrives with bills to pay and the cassette tape recorder you’ve wanted is on sale. How do you respond? The order of presentation of interpersonal and noninterpersonal situations was counterbalanced. For each situation, subjects were asked to list all solutions they could think of for each problem. To establish the objective adequacy of subjects’ problem solutions, a panel of five undergraduate and two graduate raters provided solution effectiveness ratings following procedures established in the literature (D’Zurilla & Nezu, 1980; Nezu & D’Zurilla, 1979). Raters were given definitions ofproblem and effective solution from which their decisions were made. Problem was defined as a “real life situation or set of circumstances to which an individual is required to respond if he is to function effectively within his environment, but for which no effective response or solution is immediately available or apparent to the individual” (Nezu & D’Zurilla. 1979, p. 27 1). &Fxtive solution was defined as a “course of action which, if properly implemented, would alter
Personal and parental problem drinking
195
the problematic situation so that it is no longer a problem to the individual, while maximizing positive consequences and minimizing negative consequences, long-term as well as short-term, and social as well as personal consequences” (Nezu & D’Zurilla, 1979, p. 27 1). All raters completed three weeks of training where ten inventories collected from pilot subjects were scored by all raters and consensual rating criteria were derived. Each generated solution was rated on a 9-point Likert-type scale of effectiveness where ratings of 1,2, and 3 were designated as “low effectiveness,” ratings of 4,5, and 6 as “moderate effectiveness,” and ratings of 7, 8, and 9 as “high effectiveness.” The intraclass correlation coefficient (Winer, 1982) was calculated as a measure of interrater agreement for the seven raters, who each rated 100% of subjects’ PAS responses in this study. Inter-rater agreement was high, ranging from r = .88 to r = .94 across the four situations. The scores for the seven raters on each response were averaged, and mean consensus ratings were used in analyses for each stimulus situation. The Problem Solving Inventory (PSI) (Heppner & Peterson, 1982). The PSI was administered to assess self-perceptions of personal problem-solving. It consists of 32 sixpoint, Likert-type items and contains three subscales based on factor analytic studies: Problem-Solving Confidence ( 11 items), Approach-Avoidance Style ( 16 items), and Personal Control (5 items). In addition, the total score is viewed as a single, general index of problem-solving self-appraisal. Initial reliability estimates (Heppner & Petersen, 1982) suggested that the factors are internally consistent (.72 to .90, n = 150) and stable over a two-week period (.83 to .89, n = 3 1). A broad range of validity estimates are provided in the research manual (Heppner, 1990). RESULTS
A series of 2 (Positive Parental History of Problem Drinking vs. Negative Parental History of Problem Drinking) X 2 (Alcohol-abusing vs. Nonabusing) X 2 (Male vs. Female) X 2 (Firstborn vs. Laterborn) analyses of variance (ANOVAs) was conducted on all subject classification and dependent measures. Duncan’s Multiple Range Tests were used to examine significant interaction effects. Because of the study’s focus on the impact of parental problem drinking, only significant interactions involving parental problem drinking are presented. Subject class$cation measures Classification measures identified unique subsamples (i.e., effectively produced nonoverlapping distributions) as alcohol-abusers had higher MAST scores than nonabusers (abuser mean = 7.93, SD = 3.76; nonabuser mean = 1.8 1, SD = 1.69, F( 1, 136) = 178.74, p < .OOl), and COPDs had higher F-SMAST and M-SMAST scores than nonCOPDs (F-SMAST: COPD mean = 6.77, SD = 3.39; non-COPD mean = 0.87, SD = 1.79; F(1, 136) = 198.2, p < .OOl; M-SMAST: COPD mean = 1.64, SD = 2.38, non-COPD mean = 0.49, SD = 0.84; F(1, 136) = 25.2,~ < .OOl). Dependent measures The Problem Assessment Schedule (PAS). Preliminary analyses on the four stimulus situations revealed no between-situation differences for the number of alternatives generated, and subjects’ scores were summed across situation for further analyses. No sig-
nificant between-group differences were found for the total number of solutions generated. Analyses of observer ratings of solution effectiveness revealed a main effect for stimulus situation (F(3,384) = 9.3 1, p < .OO1). However, situation did not interact systematically with parental problem drinking or any other grouping variable, suggesting that observer ratings did not vary systematically as a function of group membership. Therefore. observer rating measures were also collapsed across situation for additional analyses. Analyses were conducted on mean observer ratings of the first three solutions generated across situations (examination of larger numbers of solutions was impaired by increasing amount of missing data). A five-way repeated measures ANOVA (Parental History of Problem Drinking X Alcohol Abuse X Gender X Birth Order X Solution Number) revealed a significant effect for the repeated measure (solution number). (F(2,230) = 6.23, p < .002). Mean effectiveness ratings decreased significantly from first to third solution for all subjects. Analyses also produced a main effect for alcohol abuse (F( 1,130) = 4.18. p < .05). Subjects who abused alcohol received poorer ratings ofsolution effectiveness than those who did not. There was also a significant main effect for parental history of problem drinking (J’( 1,130) = 7.38, p < .OOS).Contrary to our predictions, COPDs’ solutions were rated as ln01-e effective than those of non-COPDs. This main effect was qualified by significant interaction among parental history of problem drinking, birth order, and solution number (F(2,230) = 3.58, p < .03). To clarify this interaction. 2 (Parental History of Problem Drinking) X 2 (Birth Order) ANOVAs were conducted separately for each solution. The interaction was significant for the first solution only, (F( 1,130) = 5.63. p < .03). Laterborn COPDs received significantly higher effectiveness ratings on their first solutions than all other subjects, but this difference was no longer evident on second and third solutions. The Problem Solving Inventory (PSI). Preliminary analyses revealed that PSI subscales and the total score were highly intercorrelated for all subjects (IS = .73-.93). Analyses of subscales and total scores were virtually identical, so we report only the analysis of total scores. A 2 (Parental History of Problem Drinking) X 2 (Alcohol Abuse) X 2 (Gender) X (Birth Order) ANOVA revealed a significant interaction between parental history of problem drinking and alcohol abuse (F( 1.134) = 6.80. p < .O1). Alcohol-abusing COPDs were significantly less positive in their problem-solving self-appraisals than alcohol-abusing non-COPDs, who were the most favorable (lower scores mean more positive appraisals). The nonabuser groups fell in between. nonsignificantly different from either alcohol-abusing group (see Figure 1). DISCUSSION
The present study examined the problem-solving performances and self-appraisals of problem-solving ability of college-age subjects with and without parental history of problem-drinking. Contrary to our prediction that children of problem drinkers would demonstrate problem-solving performance deficiencies, their solutions vvere judged as significantly more effective than those of children of nonproblem drinkers. While this difference was not large and absent after the first solution generated, it suggests that some COPDs become competent problem-solvers despite their family experience. This finding adds to a growing body of research on “resilient offspring,” or individuals who
197
Personal and parental problem drinking
COPD
Non-COPD
100 98 -
-
96 -
z P
88 86 84 82 80 Abusing
NonAbusing
Abusing
Non.Abusing
Fig. I. Effects of parental history of problem drinking and alcohol abuse on Problem Solving Inventor) (PSI) scores. COPD = Child of Problem Drinker: Non-COPD = Child of Non-Problem Drinker. Lower PSI scores represent higher self-perceived problem-solving effectiveness.
function adequately (in some cases exceptionally), despite the adversity of their early family experience (Rutter, 1983; Werner, 1986). This finding also contradicts prevailing assumptions from clinical writings on children of alcoholics (COAs), which suggest that all COAs are vulnerable to problem-solving difficulties by virtue of their parental drinking history alone (Woititz, 1983). More generally, the adequacy of COPDs’ problem-solving performance highlights the value of empirically validating assumptions from the clinical literature because a variety of factors, other than those directly influenced by the drinking parent, may affect offspring development and problem-solving competence. How then, do we account for the apparent adequacy of COPDs problem-solving skills? Perhaps the performance of the present sample reflects a tendency for signs of maladjustment (i.e., problem-solving skills deficiencies) to attenuate when offspring leave the problem-drinking household (Benson & Heller, 1987). Alternatively, COPDs with adequate problem-solving skills may be more likely than skill deficient COPDs to attend college. Indeed, paternal criminality and childhood disorder are among the most potent predictors of problem drinking and maladjustment in COAs (West & Prinz, 1987). and these factors may predispose the child against going to college.’ Thus, our sample may be comprised of highly functional COPDs, by virtue of their university attendance. Longitudinal studies of offspring of problem drinkers (e.g., from ages 6 to 18) could examine these hypotheses by studying developmental trends in COPDs’ prob‘R’e wish to thank one ofthe anonymous
reviewers for this comment
198
STACEY
L. SL.4VKIN
et al
lem-solving abilities, comparing subjects’ performance and self-appraisals to parent and teacher ratings of problem-solving efficacy and measures of adjustment. The present study also revealed a main effect for subjects’ own alcohol abuse on problem-solving performance. Regardless of parental drinking history, the problem solutions of current alcohol abusers were significantly less effective than those of nonabusers. This finding is consistent with earlier research documenting problem-solving skill deficiencies among alcoholic populations (Intagliata, 1978: Patterson et al., 1988). While recent studies suggest that on-campus alcohol abuse is remarkably common (Winning, Heimberg, & Slavkin, 1990), it may be argued that such a period of heightened alcohol consumption is transitory and dictated by the social norms of college life. If this is true, the relative problem-solving skill deficiencies of alcohol-abusing subjects may not prevail after college. Alternatively, for asubset ofcurrent alcohol abusers, especially COPDs, this period may be “prealcoholic,” or a time when excessive drinking habits are acquired and begin to shape the individual’s behavioral repertoire. Larson and Heppner (1989) suggest that congruence between self-appraised abilities and problem-solving performance may be an index ofpsychological adjustment worthy of investigation. In our study discrepancies between self-appraised problem-solving ability (PSI) and demonstrated problem solving skill (PAS) distinguished between abusing subjects with and without parental history of problem drinking. Both COPD and non-COPD alcohol abusers fell outside the normative range of PSI problem-solving self-appraisals (Heppner, 1990), but in different directions. COPD abusers expressed the least confidence in their abilities while alcohol-abusing non-COPDs expressed the most confidence. Given the negative effect of alcohol abuse on problem-solving performance, the self-evaluations of alcohol-abusing COPDs were negative though relatively accurate while those of alcohol abusing non-COPDs appeared unrealistically positive. This discrepancy for alcohol abusing non-COPDs suggests that they will overestimate their problem solving skills and make poor problem solving decisions. However, discrepancy between perceived and actual problem solving skills does not tell the whole story when both indices are low, as in the case of abusing COPDs. Intervention with abusing COPDs and non-COPDs should both focus on increasing problem solving skill. However, abusing COPDs may require additional interventions targeted at self-esteem and self-efficacy, while abusing non-COPDs may require a focus on reducing appraisal/ performance discrepancy. Our findings suggest that a variety of factors may influence problem-solving performance and self-appraisal, highlighting the utility of designs which consider both parental history (i.e., of problem drinking) and current adjustment (i.e., personal alcohol abuse) variables simultaneously. They may only generalize, however, to other college populations, and replication studies are needed with representative community samples and with offspring ofclinically diagnosed alcoholic parents. The results ofthis study may also be limited by the use of hypothetical problem situations in the PAS (Heppner & Krauskopf, 1987). Because the present study was a preliminary assessment, however, it was deemed pragmatic to first examine problem-solving performance in this manner and then, ideally, to replicate the procedure using observational measures of problemsolving performance (Krasnor & Rubin, 198 1). REFERENCES Benson, C. S.. & ;Ieller, K. (1987). Factors in the current adjustment of young adult daughters of alcoholic and problem-drinking fathers. Jow~alo~Abnorrnal Psychology. 96, 305-J 12. Clair. D.. & Genest, M. (1987). Variables associated with the adjustment of offspring of alcoholic fathers.
Journal oJ.Sictdies on Alcohol. 48, 345-355.
Personal and parental problem drinking
199
D’Zurilla. T. J., & Goldfried, M. R. (197 I). Problem-solving and behavior modification. Ju~natof~-lbnormal Psychology, 78, 107- 128. D’ZuriIla, T. J., & Nezu. A. (I 980). A study ofthe generation ofalternatives process in social problem-solving. Cogni[ive Therap! and Research, 4,67-72. Endicott, J., Andreasen. N., & Spitzer, R. L. (1978). Family Hislory - Research Diagnosric Crireria (FHRDC). New York: New York State Psychiatric Institute. Goodwin, D. W., Schulsinger, F., Knop, J., Mednick, S.. & Guze, S. B. (1977). Alcoholism and depression in adopted-out daughters of alcoholics. Archives of General Psychiarry, 34, 75 l-755. Heppner, P. P. (1990). The Problem Solving Invenrory: Manual. Palo Alto, CA: Consulting Psychologists Press. Heppner, P. P., & Anderson, W. P. (1985). The relationship between problem-solving self-appraisal and psychologjcal adjustment. Co,qnilive Therapv and Research, 9.4 15-427. Heppner, p. P., gaumgardner,-A. H., & Jackson, J. (1985). Problem-solving, depression and attributional style: Are they related? Cognirive Therapy and Researc.h, 9, 105-I 13. Heppner, P. P., & Krauskopf, C. J. (1987). An information-processingapproach to personal problem-solving. The Counseling Psychologist. 15, 37 l-447. Heppner, P. P., & Petersen, C. H. (1982). The development and implications of a personal problem-solving inventory. Journal of Counseling Psychology, 29,66-75. Intagliata, J. C. (1978). Increasing the interpersonal problem-solvingskillsofan alcoholic population. Journal of Consulfing and Clinical Psychology, 46,489-498. Krasnor, L. R., & Rubin, K. H. (198 I). Assessment of social problem-solving skills in young children. In T. Merluzzi, C. R. Glass, & M. Genest (Eds.), Cognirive assessment (pp. 452-476). New York: Guilford Press. Larson, L. M. (1986). Problem-solving appraisal with male alcoholics. Unpublished doctoral dissertation, University of Missouri, Columbia. Larson, L. M., & Heppner, P. P. (1989). Problem-solving appraisal in an alcoholic population. Journal of Counseling Psychology. 36,73-78. Levenson, R. W., Oyama, 0. N., & Meek, P. S. (I 987). Greater reinforcement from alcohol for those at risk: Parental risk, personality risk, and sex. Journal ofdbnormal Psvchology. 96,242-253. Newlin, D. B. ( 1985). Offspring of alcoholics have enhanced antagonistic placebo response. Journal of&dies on Alcohol, 46,490-494. Nezu, A. M., & D’Zurilla, T. J. (1979). An experimental evaluation of the decision-making process in social problem-solving. Connirive Therapy and Research. 3.169-277. Nez;, A. M.. & Ron&. G1 F. ( 1988). Social problem-solving as a moderator ofstress-related depressive symptoms: A prospective analysis. Journal of Counseling Psychology, 35, I34- 138. Parker, D. A., & Hartford, T. C. (1988). Alcohol-related problems, marital disruption, and depressive symptoms among adult children of alcohoi abusers in the United States. Journal of Sutdies on .-llcohol. 49, 306-313. Patterson. B. W.. Parsons, 0. A., Schaeffer, K. W., & Errico, A. L. (1988). Interpersonal problem solving in alcoholics. Journal y/‘lVervotrs and Menral Disease, 176, 707-7 13. Platt. J. J.. & Spivack. G. ( 1972). Problem-solving thinking of psychiatric patients. Journal qj’Consul!ing and Clinical Psychology.. 39, I48- 15 I. Platt. J. J.. Spivack. G.. Altman, N., Altman, D., & Peizer. S. B. (1974). Adolescent problem-solving thinking. Journalof Consulring and Clinical Psyct1olog.v.42, 787-793. Rutter, M. (1983). Stress, coping and development: Some issues and some questions. In N. Garmezy & M. Rutter (Eds.), .%ess. coping and developmen in children (pp. l-43). New York: McGraw-Hill. Selzer, M. L. (197 I ). The Michigan Alcoholism Screening Test: The quest for a new diagnostic instrument. American Journal o/Psjrhiatq: 127, 1653-1658. Sher, K. J., & Descutner, C. (1986). Reports of paternal alcoholism: Reliability across siblings. .-lddictive Behaviars. 11, 25-30, Shure, M., & Spivack. G. ( 1972). Means-ends thinking. adjustment and social class among elementary schoolaged children. Journal of Consulting and Clinical Psl,chology, 38, 348-353. Tisdell, D. A., & Lawrence. J. S. (I 986).-Interpersonal prbblem&olving competency: Review and critique of the literature. Clinical Psvcholonv Rerierc: 6,337-356. Wegscheider, D., & \\‘egscheiher, S-(1978). Family illness: Chemical dependency. Crystal, MN: Surturing Networks. Werner, E. E. (1986). Resilient offspring of alcoholics: A longitudinal study from birth to age 18. Journal y/ Studies
on Acohol.
47, 34-40.
West, M. 0.. & Prinz. R. J. (1987). Parental alcoholism and childhood psychopathology. Psychological Burllerin. 102,204-2 18. Winer, B. J. (1982). Sralisticalprinciples in e.rperimenral design (2nd ed.). New York: hlcGraw-Hill. Winning, C. D., Heimbern. R. G., & Slavkin, S. L. (1990). Use ofllze Michiean Alcoholism Screeninp ” Tc.!~ wiri college sruden!s:%‘ormalive data andjacror analysis. Unpublished manuscript. Woititz, J. G. (1983). .4dulr children ofalcoholics. Pompano Beach, FL: Health Communications, Inc.
0306-460319’ $5.00 + .@I Copyright D 1992 Pergamon Press Ltd.
PERSONAL AND PARENTAL PROBLEM PROBLEILI-SOLVING PERFORMANCE
DRINKING: EFFECTS ON AND SELF-APPRAISAL
STACEY L. SLAVKIN, RICHARD G. HEIMBERG, CHERYL D. WINNING, and ROBERT J. McCAFFREY University at Albany, State University of New York Abstract - This study examined the problem-solving performances and self-appraisals of problem-solving ability of college-age subjects with and without parental history of problem drinking. Contrary to our predictions, children of problem drinkers (COPDs) were rated as somewhat tnore effective in their problem-solving skills than non-COPDs. undermining prevailing assumptions about offspring from alcoholic households. While this difference was not large and was qualified by other variables. subjects’ own alcohol abuse did exert a detrimental effect on problem-solving performance, regardless of parental history of problem drinking. However. a different pattern was evident for problem-solving self-appraisals. Alcohol-abusing non-COPDs saw themselves as eficlire problem-solvers while alcohol-abusing COPDs appraised themselves as poor problem-solvers. In addition, the self-appraisals of alcohol-abusing COPDs were consistent with objective ratings ofsolution effectiveness (i.e., they were both negative) while alcohol-abusing non-COPDs were overly positive in their appraisals. opposing the judgments of trained raters. This finding suggests that the relationship between personal alcohol abuse and self-appraised problem-solving abilities may differ as a function of parental history of problem drinking. Limitations on the generalizability of findings are addressed.
The ability to resolve personal problems has been related to a variety of indices of psychological health (Tisdell & Lawrence, 1986). Among maladjusted children, for example, deficiencies in the quantity and quality of generated problem solutions have been related to clinical ratings ofadaptive functioning (Shure & Spivack, 1972). Similar findings have been reported for emotionally disturbed adolescents (Platt, Spivack, Altman, Altman, & Peizer, 1974) and adolescent and adult psychiatric patients (Platt & Spivack, 1972). In addition, proficient problem-solving skills may serve a stress-buffering role. Nezu and Ronan’s (1988) recent study revealed that effective problem-solving skills moderate depressive symptoms for individuals experiencing negative life events. Though the literature consistently documents poor problem-solving skills in emotionally disturbed and pathological populations, exactly how these deficits contribute to emotional and behavioral maladjustment remains unclear. In an effort to better understand this process, investigators have begun to examine the role of higher-order or meta-cognitive variables which may influence the execution of effective problemsolving skills. One such variable is the self-appraisal of one’s problem-solving effectiveness (Heppner & Petersen, 1982). Problem-solving self-appraisals have been related to behavioral. cognitive, and affective components of the coping process (Heppner & Anderson, 1985; Heppner, Baumgardner, &Jackson, 1985). Self-appraised effective problem-solvers (i.e., those who perceive their problem-solving efforts as effective in modifying problems) are more motivated to solve problems, more systematic and persistent, have higher expectations Portions of this paper were presented at the annual meeting ofthe Association for Advancement of Behavior Therapy, New York, NY, November 1988. Requests for reprints should be sent to Richard Heimberg, Psychology Department, University at Albany, State University of New York, 1400 Washington Avenue, Albany, NY 12222. 191
ofsuccess, and generate more alternative solutions than self-appraised ineffective problem-solvers. This line of research on problem-solving self-appraisals has involved functional populations classified as positive or negative in their self-appraisals by Heppner and Petersen’s (1982) Problem Solvings Inventory (PSI). An equally informative strategy might involve studying individuals who have been less effective at negotiating life’s problems and daily hassles. For example. it is documented that adult alcoholics are deficient in problem-solving skills (Intagliata, 1978). Alcoholics’ self-reports ofstrategies ty.pically employed in interpersonal problem-solving situations have been scored as significantly less competent than those of nonalcoholics (Patterson, Parsons, Schaeffer, 6c Errico, 1988). It has been suggested, however, that alcoholics’ self-appraisals may not mirror their problem-solving deficits (Larson, 1986). In fact, in a recent study of recovering alcoholics (Larson & Heppner, 1989), problem-solvings self-appraisals overestimated adjustment and alcohol use as measured by clinical ratings. In a similar vein, the present study examined the problem-solving performances and self-appraisals of children of problem drinkers, another population with suspected problem-solving deficiencies. Support for this conjecture is derived from both clinical and empirical literature on children of alcoholics (COAs). Related studies suggest that being a COA confers risk for a variety of unfavorable outcomes on offspring including alcoholism, impaired cognitive/intellectual functioning, anxiety, and depression (Parker & Harford. 1988; West & Prinz, 1987). While problem-solving performance and self-appraisal have not been evaluated directly in the offspring of problem drinkers. research on the preferred coping styles of COAs suggests that they see problems as less under their control and rely more on emotion-focused coping than subjects without fami1y:histot-y of alcoholism (Clair & Genest, 1987). Thus. COAs may be impoverished in thetr repertoire of potential problem-solving behaviors and rely more heavily on emotional remedies (e.g., seeking social support, blaming self, avoidance) to problem situations. In addition to influencing the development of problem-solving skills, the alcoholic family environment may affect COAs’ problem-solving self-appraisals. Alcoholic parents may be inconsistent in their evaluations of children’s efforts (Woititz, 1983). creating confusion in offspring. While drinking, parents may be belittling and unable to offer constructive criticism or advice. These speculations are supported by Clair and Genest’s ( 1987) finding that COAs receive less informational support (which may facilitate effective problem-solving) than non-COAs. Alternatively, during sober periods, when attempts to reconcile with family members are likely, alcoholic parents may encourage children’s problem-solving. Thus, COAs may integrate contradictory evaluations of their problem-solving abilities when forming self-appraisals. Such inconsistencies may manifest themselves as problem-solving performance/appraisal discrepancies. In Larson and Heppner’s (1989) study, problem-solving performance quality was operationalized as the number of items endorsed on a list of personal problems, with low numbers taken as an index of problem-solving. The present study included a more direct evaluation of problem-solving performance. Subjects generated possible solutions to hypothetical problems, and objective raters judged their quantity and quality. While others have noted that hypothetical problem-solving may not correspond exactly with “real-life” behavior (Heppner & Krauskopf, 1987) vve believe that this strategy provides an adequate analogue for the purposes of this investigation. as well as a more precise examination of components of the problem-solving process (i.e., generation of
Personal
and parental
problem
drinking
alternative solutions, appraisal of solution suitability) (D’Zurilla & Goldfried, than that provided by Larson and Heppner’s (1989) methodology. hl E T H 0
I93
197 1)
D
Experimental design The present study was a 2 (Positive Parental History of Problem Drinking vs. Negative Parental History of Problem Drinking) X 2 (Alcohol-abusing vs. Non-alcohol abusing) X 2 (Male vs. Female) X 2 (Firstborn vs. Laterborn) factorial design. Previous research suggests that having alcoholic parents places sons at greater risk for dependent problem drinking than daughters (Goodwin, Schulsinger, Knop, Mednick, & Guze. 1977: Parker & Harford, 1988). Furthermore, clinical writings on the role structure of alcoholic families (Wegscheider & Wegscheider, 1978; Woititz, 1983) imply that firstborn COAs typically assume caretaking responsibilities, presumably defined by effective problem-solving. Given these suggestions, the effects of gender and birth order were also considered in the present study. All subjects completed the assessment battery described below. Subjects Subjects were 85 women and 67 men enrolled in introductory psychology courses at the University at Albany, State University of New York, who earned credit toward fulfillment of course requirements for their participation. Informed consent was given for participation in the present study. Subject classification. Of 772 students who completed initial screening procedures, 152 were recontacted and agreed to participate in the present study. An attempt was made to recontact as many children of problem drinkers (COPDs) as possible because they constituted only 10% of the sample. Children of nonproblem drinkers were randomly selected from the contact list. All students completed the following classification measures: The .\lichigan Alcoholism Screening Test (MAST) (Selzer, 1971). The MAST is a brief screening test for the detection of alcoholic drinking and is the standard measure for this purpose in the field. A cutoff score of 5 is traditionally interpreted to represent alcoholic drinking and was required for classification as an alcohol abuser in the current study. A total of 69 subjects (45.4%) were classified as “abusing” and the remaining 83 subjects (54.6%) were classified as “nonabusing.” The Short MASTs for Mother and Father (F-SMAST & M-SMAST) (Sher & Descutner, 1986). These scales are abbreviated versions of the MAST reworded to inquire about parental drinking history and related family issues. Reliability data have been reported by Sher and Descutner (1986) and several additional studies provide validational support for these measures (Benson & Heller, 1987; Levenson, Oyama, & Meek, 1987; Newlin, 1985). In each of these studies, the parental MASTS were used for subject classification. In addition, Sher (personal communication, November 4, 1989) reports that scores on these measures correspond well to diagnostic information derived from selected sections of the Family History-Research Diagnostic Criteria interview (FHRDC; Endicott, Andreasen, & Spitzer, 1978), a structured clinical interview concerning family history of alcohol and drug abuse, antisocial personality, and depression, as well as other forms of psychopathology. In the present study, a cutoff score of 5 was used to
194
STACEY
L. SLAVKIS
et al
identify COPDs, a conservative criterion relative to studies where this demarcation point was used to classify subjects as offspring of alcoholics (Sher & Descutner. 1986). In our sample, 61 (40%: 18 alcohol-abusing; 43 nonabusing) of the subjects were COPDs and 9 1 (60%; 5 1 abusing; 40 nonabusing) were not. We prefer the label of COPD to that of COA because we were unable to conduct diagnostic studies with parents. Demographic informafion. All subjects completed a demographic sheet requesting their birth order (i.e., first or laterborn). Ofsubjects studied, 6 1 (40%) were firstborn and 9 1 (60%) were laterborn. Dependent measures The Problem Assessment Schedule (PAS). There are few descriptions of behavioral assessment in the problem-solving literature. Krasnor and Rubin (198 1) suggest that behavioral problem-solving assessment should involve obsenation of actual problems as they arise in the individual’s daily social interactions. Clearly, such in vivo assessment would be desirable and appropriate in clinical settings. Hovvever, given the problems inherent in uncontrolled observational environments, we chose to provide analogue situations deemed relevant to college students. The PAS (developed for the present study) yields both a subject-generated quantitative measure (i.e., the number of solutions generated) and a qualitative measure (i.e., observer ratings of solution adequacy), two central dimensions of effective problem-solving behavior (D’Zurilla & Goldfried, 1971). The PAS presents four problem situations germane to college students, two interpersonal and two noninterpersonal situations. Interpersonal situations include: 1. You watch your closest friend become increasingly involved with someone who doesn’t treat him/her very vvell. How do you respond? 2. Your boyfriend’s/girlfriend’s roommate expresses an interest in you. How do you respond? Noninterpersonal
situations include:
3. You are home alone and find that the basement of your apartment has six inches of water in it. How do you respond? 4. The end of the month arrives with bills to pay and the cassette tape recorder you’ve wanted is on sale. How do you respond? The order of presentation of interpersonal and noninterpersonal situations was counterbalanced. For each situation, subjects were asked to list all solutions they could think of for each problem. To establish the objective adequacy of subjects’ problem solutions, a panel of five undergraduate and two graduate raters provided solution effectiveness ratings following procedures established in the literature (D’Zurilla & Nezu, 1980; Nezu & D’Zurilla, 1979). Raters were given definitions ofproblem and effective solution from which their decisions were made. Problem was defined as a “real life situation or set of circumstances to which an individual is required to respond if he is to function effectively within his environment, but for which no effective response or solution is immediately available or apparent to the individual” (Nezu & D’Zurilla. 1979, p. 27 1). &Fxtive solution was defined as a “course of action which, if properly implemented, would alter
Personal and parental problem drinking
195
the problematic situation so that it is no longer a problem to the individual, while maximizing positive consequences and minimizing negative consequences, long-term as well as short-term, and social as well as personal consequences” (Nezu & D’Zurilla, 1979, p. 27 1). All raters completed three weeks of training where ten inventories collected from pilot subjects were scored by all raters and consensual rating criteria were derived. Each generated solution was rated on a 9-point Likert-type scale of effectiveness where ratings of 1,2, and 3 were designated as “low effectiveness,” ratings of 4,5, and 6 as “moderate effectiveness,” and ratings of 7, 8, and 9 as “high effectiveness.” The intraclass correlation coefficient (Winer, 1982) was calculated as a measure of interrater agreement for the seven raters, who each rated 100% of subjects’ PAS responses in this study. Inter-rater agreement was high, ranging from r = .88 to r = .94 across the four situations. The scores for the seven raters on each response were averaged, and mean consensus ratings were used in analyses for each stimulus situation. The Problem Solving Inventory (PSI) (Heppner & Peterson, 1982). The PSI was administered to assess self-perceptions of personal problem-solving. It consists of 32 sixpoint, Likert-type items and contains three subscales based on factor analytic studies: Problem-Solving Confidence ( 11 items), Approach-Avoidance Style ( 16 items), and Personal Control (5 items). In addition, the total score is viewed as a single, general index of problem-solving self-appraisal. Initial reliability estimates (Heppner & Petersen, 1982) suggested that the factors are internally consistent (.72 to .90, n = 150) and stable over a two-week period (.83 to .89, n = 3 1). A broad range of validity estimates are provided in the research manual (Heppner, 1990). RESULTS
A series of 2 (Positive Parental History of Problem Drinking vs. Negative Parental History of Problem Drinking) X 2 (Alcohol-abusing vs. Nonabusing) X 2 (Male vs. Female) X 2 (Firstborn vs. Laterborn) analyses of variance (ANOVAs) was conducted on all subject classification and dependent measures. Duncan’s Multiple Range Tests were used to examine significant interaction effects. Because of the study’s focus on the impact of parental problem drinking, only significant interactions involving parental problem drinking are presented. Subject class$cation measures Classification measures identified unique subsamples (i.e., effectively produced nonoverlapping distributions) as alcohol-abusers had higher MAST scores than nonabusers (abuser mean = 7.93, SD = 3.76; nonabuser mean = 1.8 1, SD = 1.69, F( 1, 136) = 178.74, p < .OOl), and COPDs had higher F-SMAST and M-SMAST scores than nonCOPDs (F-SMAST: COPD mean = 6.77, SD = 3.39; non-COPD mean = 0.87, SD = 1.79; F(1, 136) = 198.2, p < .OOl; M-SMAST: COPD mean = 1.64, SD = 2.38, non-COPD mean = 0.49, SD = 0.84; F(1, 136) = 25.2,~ < .OOl). Dependent measures The Problem Assessment Schedule (PAS). Preliminary analyses on the four stimulus situations revealed no between-situation differences for the number of alternatives generated, and subjects’ scores were summed across situation for further analyses. No sig-
nificant between-group differences were found for the total number of solutions generated. Analyses of observer ratings of solution effectiveness revealed a main effect for stimulus situation (F(3,384) = 9.3 1, p < .OO1). However, situation did not interact systematically with parental problem drinking or any other grouping variable, suggesting that observer ratings did not vary systematically as a function of group membership. Therefore. observer rating measures were also collapsed across situation for additional analyses. Analyses were conducted on mean observer ratings of the first three solutions generated across situations (examination of larger numbers of solutions was impaired by increasing amount of missing data). A five-way repeated measures ANOVA (Parental History of Problem Drinking X Alcohol Abuse X Gender X Birth Order X Solution Number) revealed a significant effect for the repeated measure (solution number). (F(2,230) = 6.23, p < .002). Mean effectiveness ratings decreased significantly from first to third solution for all subjects. Analyses also produced a main effect for alcohol abuse (F( 1,130) = 4.18. p < .05). Subjects who abused alcohol received poorer ratings ofsolution effectiveness than those who did not. There was also a significant main effect for parental history of problem drinking (J’( 1,130) = 7.38, p < .OOS).Contrary to our predictions, COPDs’ solutions were rated as ln01-e effective than those of non-COPDs. This main effect was qualified by significant interaction among parental history of problem drinking, birth order, and solution number (F(2,230) = 3.58, p < .03). To clarify this interaction. 2 (Parental History of Problem Drinking) X 2 (Birth Order) ANOVAs were conducted separately for each solution. The interaction was significant for the first solution only, (F( 1,130) = 5.63. p < .03). Laterborn COPDs received significantly higher effectiveness ratings on their first solutions than all other subjects, but this difference was no longer evident on second and third solutions. The Problem Solving Inventory (PSI). Preliminary analyses revealed that PSI subscales and the total score were highly intercorrelated for all subjects (IS = .73-.93). Analyses of subscales and total scores were virtually identical, so we report only the analysis of total scores. A 2 (Parental History of Problem Drinking) X 2 (Alcohol Abuse) X 2 (Gender) X (Birth Order) ANOVA revealed a significant interaction between parental history of problem drinking and alcohol abuse (F( 1.134) = 6.80. p < .O1). Alcohol-abusing COPDs were significantly less positive in their problem-solving self-appraisals than alcohol-abusing non-COPDs, who were the most favorable (lower scores mean more positive appraisals). The nonabuser groups fell in between. nonsignificantly different from either alcohol-abusing group (see Figure 1). DISCUSSION
The present study examined the problem-solving performances and self-appraisals of problem-solving ability of college-age subjects with and without parental history of problem-drinking. Contrary to our prediction that children of problem drinkers would demonstrate problem-solving performance deficiencies, their solutions vvere judged as significantly more effective than those of children of nonproblem drinkers. While this difference was not large and absent after the first solution generated, it suggests that some COPDs become competent problem-solvers despite their family experience. This finding adds to a growing body of research on “resilient offspring,” or individuals who
197
Personal and parental problem drinking
COPD
Non-COPD
100 98 -
-
96 -
z P
88 86 84 82 80 Abusing
NonAbusing
Abusing
Non.Abusing
Fig. I. Effects of parental history of problem drinking and alcohol abuse on Problem Solving Inventor) (PSI) scores. COPD = Child of Problem Drinker: Non-COPD = Child of Non-Problem Drinker. Lower PSI scores represent higher self-perceived problem-solving effectiveness.
function adequately (in some cases exceptionally), despite the adversity of their early family experience (Rutter, 1983; Werner, 1986). This finding also contradicts prevailing assumptions from clinical writings on children of alcoholics (COAs), which suggest that all COAs are vulnerable to problem-solving difficulties by virtue of their parental drinking history alone (Woititz, 1983). More generally, the adequacy of COPDs’ problem-solving performance highlights the value of empirically validating assumptions from the clinical literature because a variety of factors, other than those directly influenced by the drinking parent, may affect offspring development and problem-solving competence. How then, do we account for the apparent adequacy of COPDs problem-solving skills? Perhaps the performance of the present sample reflects a tendency for signs of maladjustment (i.e., problem-solving skills deficiencies) to attenuate when offspring leave the problem-drinking household (Benson & Heller, 1987). Alternatively, COPDs with adequate problem-solving skills may be more likely than skill deficient COPDs to attend college. Indeed, paternal criminality and childhood disorder are among the most potent predictors of problem drinking and maladjustment in COAs (West & Prinz, 1987). and these factors may predispose the child against going to college.’ Thus, our sample may be comprised of highly functional COPDs, by virtue of their university attendance. Longitudinal studies of offspring of problem drinkers (e.g., from ages 6 to 18) could examine these hypotheses by studying developmental trends in COPDs’ prob‘R’e wish to thank one ofthe anonymous
reviewers for this comment
198
STACEY
L. SL.4VKIN
et al
lem-solving abilities, comparing subjects’ performance and self-appraisals to parent and teacher ratings of problem-solving efficacy and measures of adjustment. The present study also revealed a main effect for subjects’ own alcohol abuse on problem-solving performance. Regardless of parental drinking history, the problem solutions of current alcohol abusers were significantly less effective than those of nonabusers. This finding is consistent with earlier research documenting problem-solving skill deficiencies among alcoholic populations (Intagliata, 1978: Patterson et al., 1988). While recent studies suggest that on-campus alcohol abuse is remarkably common (Winning, Heimberg, & Slavkin, 1990), it may be argued that such a period of heightened alcohol consumption is transitory and dictated by the social norms of college life. If this is true, the relative problem-solving skill deficiencies of alcohol-abusing subjects may not prevail after college. Alternatively, for asubset ofcurrent alcohol abusers, especially COPDs, this period may be “prealcoholic,” or a time when excessive drinking habits are acquired and begin to shape the individual’s behavioral repertoire. Larson and Heppner (1989) suggest that congruence between self-appraised abilities and problem-solving performance may be an index ofpsychological adjustment worthy of investigation. In our study discrepancies between self-appraised problem-solving ability (PSI) and demonstrated problem solving skill (PAS) distinguished between abusing subjects with and without parental history of problem drinking. Both COPD and non-COPD alcohol abusers fell outside the normative range of PSI problem-solving self-appraisals (Heppner, 1990), but in different directions. COPD abusers expressed the least confidence in their abilities while alcohol-abusing non-COPDs expressed the most confidence. Given the negative effect of alcohol abuse on problem-solving performance, the self-evaluations of alcohol-abusing COPDs were negative though relatively accurate while those of alcohol abusing non-COPDs appeared unrealistically positive. This discrepancy for alcohol abusing non-COPDs suggests that they will overestimate their problem solving skills and make poor problem solving decisions. However, discrepancy between perceived and actual problem solving skills does not tell the whole story when both indices are low, as in the case of abusing COPDs. Intervention with abusing COPDs and non-COPDs should both focus on increasing problem solving skill. However, abusing COPDs may require additional interventions targeted at self-esteem and self-efficacy, while abusing non-COPDs may require a focus on reducing appraisal/ performance discrepancy. Our findings suggest that a variety of factors may influence problem-solving performance and self-appraisal, highlighting the utility of designs which consider both parental history (i.e., of problem drinking) and current adjustment (i.e., personal alcohol abuse) variables simultaneously. They may only generalize, however, to other college populations, and replication studies are needed with representative community samples and with offspring ofclinically diagnosed alcoholic parents. The results ofthis study may also be limited by the use of hypothetical problem situations in the PAS (Heppner & Krauskopf, 1987). Because the present study was a preliminary assessment, however, it was deemed pragmatic to first examine problem-solving performance in this manner and then, ideally, to replicate the procedure using observational measures of problemsolving performance (Krasnor & Rubin, 198 1). REFERENCES Benson, C. S.. & ;Ieller, K. (1987). Factors in the current adjustment of young adult daughters of alcoholic and problem-drinking fathers. Jow~alo~Abnorrnal Psychology. 96, 305-J 12. Clair. D.. & Genest, M. (1987). Variables associated with the adjustment of offspring of alcoholic fathers.
Journal oJ.Sictdies on Alcohol. 48, 345-355.
Personal and parental problem drinking
199
D’Zurilla. T. J., & Goldfried, M. R. (197 I). Problem-solving and behavior modification. Ju~natof~-lbnormal Psychology, 78, 107- 128. D’ZuriIla, T. J., & Nezu. A. (I 980). A study ofthe generation ofalternatives process in social problem-solving. Cogni[ive Therap! and Research, 4,67-72. Endicott, J., Andreasen. N., & Spitzer, R. L. (1978). Family Hislory - Research Diagnosric Crireria (FHRDC). New York: New York State Psychiatric Institute. Goodwin, D. W., Schulsinger, F., Knop, J., Mednick, S.. & Guze, S. B. (1977). Alcoholism and depression in adopted-out daughters of alcoholics. Archives of General Psychiarry, 34, 75 l-755. Heppner, P. P. (1990). The Problem Solving Invenrory: Manual. Palo Alto, CA: Consulting Psychologists Press. Heppner, P. P., & Anderson, W. P. (1985). The relationship between problem-solving self-appraisal and psychologjcal adjustment. Co,qnilive Therapv and Research, 9.4 15-427. Heppner, p. P., gaumgardner,-A. H., & Jackson, J. (1985). Problem-solving, depression and attributional style: Are they related? Cognirive Therapy and Researc.h, 9, 105-I 13. Heppner, P. P., & Krauskopf, C. J. (1987). An information-processingapproach to personal problem-solving. The Counseling Psychologist. 15, 37 l-447. Heppner, P. P., & Petersen, C. H. (1982). The development and implications of a personal problem-solving inventory. Journal of Counseling Psychology, 29,66-75. Intagliata, J. C. (1978). Increasing the interpersonal problem-solvingskillsofan alcoholic population. Journal of Consulfing and Clinical Psychology, 46,489-498. Krasnor, L. R., & Rubin, K. H. (198 I). Assessment of social problem-solving skills in young children. In T. Merluzzi, C. R. Glass, & M. Genest (Eds.), Cognirive assessment (pp. 452-476). New York: Guilford Press. Larson, L. M. (1986). Problem-solving appraisal with male alcoholics. Unpublished doctoral dissertation, University of Missouri, Columbia. Larson, L. M., & Heppner, P. P. (1989). Problem-solving appraisal in an alcoholic population. Journal of Counseling Psychology. 36,73-78. Levenson, R. W., Oyama, 0. N., & Meek, P. S. (I 987). Greater reinforcement from alcohol for those at risk: Parental risk, personality risk, and sex. Journal ofdbnormal Psvchology. 96,242-253. Newlin, D. B. ( 1985). Offspring of alcoholics have enhanced antagonistic placebo response. Journal of&dies on Alcohol, 46,490-494. Nezu, A. M., & D’Zurilla, T. J. (1979). An experimental evaluation of the decision-making process in social problem-solving. Connirive Therapy and Research. 3.169-277. Nez;, A. M.. & Ron&. G1 F. ( 1988). Social problem-solving as a moderator ofstress-related depressive symptoms: A prospective analysis. Journal of Counseling Psychology, 35, I34- 138. Parker, D. A., & Hartford, T. C. (1988). Alcohol-related problems, marital disruption, and depressive symptoms among adult children of alcohoi abusers in the United States. Journal of Sutdies on .-llcohol. 49, 306-313. Patterson. B. W.. Parsons, 0. A., Schaeffer, K. W., & Errico, A. L. (1988). Interpersonal problem solving in alcoholics. Journal y/‘lVervotrs and Menral Disease, 176, 707-7 13. Platt. J. J.. & Spivack. G. ( 1972). Problem-solving thinking of psychiatric patients. Journal qj’Consul!ing and Clinical Psychology.. 39, I48- 15 I. Platt. J. J.. Spivack. G.. Altman, N., Altman, D., & Peizer. S. B. (1974). Adolescent problem-solving thinking. Journalof Consulring and Clinical Psyct1olog.v.42, 787-793. Rutter, M. (1983). Stress, coping and development: Some issues and some questions. In N. Garmezy & M. Rutter (Eds.), .%ess. coping and developmen in children (pp. l-43). New York: McGraw-Hill. Selzer, M. L. (197 I ). The Michigan Alcoholism Screening Test: The quest for a new diagnostic instrument. American Journal o/Psjrhiatq: 127, 1653-1658. Sher, K. J., & Descutner, C. (1986). Reports of paternal alcoholism: Reliability across siblings. .-lddictive Behaviars. 11, 25-30, Shure, M., & Spivack. G. ( 1972). Means-ends thinking. adjustment and social class among elementary schoolaged children. Journal of Consulting and Clinical Psl,chology, 38, 348-353. Tisdell, D. A., & Lawrence. J. S. (I 986).-Interpersonal prbblem&olving competency: Review and critique of the literature. Clinical Psvcholonv Rerierc: 6,337-356. Wegscheider, D., & \\‘egscheiher, S-(1978). Family illness: Chemical dependency. Crystal, MN: Surturing Networks. Werner, E. E. (1986). Resilient offspring of alcoholics: A longitudinal study from birth to age 18. Journal y/ Studies
on Acohol.
47, 34-40.
West, M. 0.. & Prinz. R. J. (1987). Parental alcoholism and childhood psychopathology. Psychological Burllerin. 102,204-2 18. Winer, B. J. (1982). Sralisticalprinciples in e.rperimenral design (2nd ed.). New York: hlcGraw-Hill. Winning, C. D., Heimbern. R. G., & Slavkin, S. L. (1990). Use ofllze Michiean Alcoholism Screeninp ” Tc.!~ wiri college sruden!s:%‘ormalive data andjacror analysis. Unpublished manuscript. Woititz, J. G. (1983). .4dulr children ofalcoholics. Pompano Beach, FL: Health Communications, Inc.
