May 1990 * American Psychologist * 612
ABSTRACT: The relation between psychological characteristics and drug use was investigated in subjects studied longitudinally, from preschool through age 18. Adolescents who had engaged in some drug experimentation (primarily with marijuana) were the best-adjusted in the sample. Adolescents who used drugs frequently were maladjusted, showing a distinct personality syndrome marked by interpersonal alienation, poor impulse control, and manifest emotional distress. Adolescents who, by age 18, had never experimented with any drug were relatively anxious, emotionally constricted, and lacking in social skills.
Psychological differences between frequent drug users, experimenters, and
abstainers could be traced to the earliest years of childhood and related to the
quality of parenting received. The findings indicate that (a) problem drug use
is a symptom, not a cause, of personal and social maladjustment, and (b) the
meaning of drug use can be understood only in the context of an individual's
personality structure and developmental history. It is suggested that current
efforts at drug prevention are misguided to the extent that they focus on
symptoms, rather than on the psychological syndrome underlying drug abuse.
Drug abuse among young people is one of the greatest challenges of our time. Almost daily, we are besieged by media reports of drug-related tragedy, of shootings in our schools, gang warfare, and overdose-related deaths. Many see the drug problem as epidemic (Robins, 1984). As an increasing share of society's resources is diverted toward coping with the drug problem and its consequences, the need for sound, scientific information on the factors contributing to drug use is urgent.
Considerable research has already been directed toward studying the causes and correlates of drug use, and important recognitions have developed (for reviews, see Bush & Iannotti, 1985; Cox, 1985; Hawkins, Lishner, & Catalano, 1985; Jessor, 1979; Jones & Battjes, 1985; Kandel, 1980). Nevertheless, many studies to date have been interpretively constrained by various research-design or empirical limitations.
Large-scale epidemiological studies (e.g., Jessor, Chase, & Donovan, 1980; Johnston, O'Malley, & Bachmail, 1984, 1986; National Institute on Drug Abuse [NIDA], 1986) have provided much-needed information about the prevalence and patterns of drug Use, about the demographics of drug users, and about certain psychosocial characteristics of drug users. In general, however, these studies have been unable to provide the kind of in depth, psychologically rich, clinically oriented information needed to inform intervention efforts. And by their very nature, cross-sectional studies and panel studies of relatively brief duration can offer only limited or confounded understandings of the antecedents of drug use.
Recognizing the crucial importance of prospective inquiry into the psychological antecedents of drug use, a number of longitudinal studies of adolescent development have been undertaken and have deepened our understanding of the interplay of psychosocial forces during adolescence (e.g., Brook, Gordon, & Whiteman, 1985; Brook, Whiteman, Gordon. & Cohen, 1986; Jessor & Jessor, 1977, 1978; Smith & Fogg, 1978; see Kandel, 1978, for a review). In general, however, these studies have also been interpretively constrained because they have studied adolescents already well along in years (subjects have rarely been younger than age 13) and because they have tended to track these adolescents for no more than three or four years, from junior high school into high school or from high school into college. Also, these studies have tended to depend, perhaps too heavily, on self-administered, mailed, or impersonally offered questionnaires.
To date, only two truly long-term investigations into the childhood antecedents of drug use have appeared. The Woodlawn study of Kellam and his associates (Kellam, Branch, Agrawal, & Ensminger, 1975; Kellam, Brown, Rubin, & Ensminger, 1983) traced the development of a group of poor, Black, urban children beginning at ages 6 to 7. In the Woodlawn study, Kellam et al. found that psychological characteristics assessed at ages 6 to 7 foretold drug use at ages 16 to 17, a decade later. The longitudinal study initiated by Jeanne and Jack Block (see J. H. Block & J. Block, 1980) followed a group of San Francisco Bay area children from nursery school on and found numerous, theoretically coherent relations between psychological characteristics assessed in nursery school and subsequent drug use in early adolescence, at age 14 (Block, Block, & Keyes, 1988). These studies converge in demonstrating the existence of important psychological antecedents of drug use, antecedents dating to the earliest years of childhood. Conjointly, they suggest that early psychological factors may be central to an understanding of drug use, and they highlight the need for prospective research.
The present study further reports on the Block and Block sample, studied again in late adolescence when the
Copyright 1990 by the American Psychological Association Inc.
subjects had reached age 18. This later age represents a different developmental era, one in which the implications of drug use and nonuse can well take on psychological significance different from the significance of drug use and nonuse in early adolescence. The findings we report span 13 years, from preschool through age 18. By virtue of their prospective nature, these data allow inferences about the antecedents of drug use that cannot be made from retrospective, cross-sectional, or short-term panel studies.
Beyond the length of time spanned by the present investigation, the study differs from previous studies in two important ways. In most empirical studies, psychological descriptions are limited to a small number of variables that are selected by researchers on a priori grounds. In the present study, psychological descriptions are, for all practical purposes, comprehensive and open-ended. They are based on extensive evaluations of participants by panels of psychologists, and they encompass the full range of constructs subsumed by the California Adult Q-sort (CAQ; Block, 1961/1978) and the California Child Q-sort (CCQ; J. Block & J. H. Block, 1980) personality assessment instruments specifically designed to allow clinicians to provide in-depth, comprehensive psychological descriptions. The intention was to gather information psychologically rich enough to speak to clinical concerns and to inform intervention efforts.
The study also Differs from previous studies in its approach to data analysis. Previous investigators have tended to assume (and test for) linear relations between level of drug use and measures of psychosocial disturbance. In effect, such an approach assumes that occasional experimentation with drugs is psychologically problematic, if not quite as problematic as regular use, and that complete avoidance of drugs is psychologically optimal.
However, the majority of young adults in the United States, nearly two thirds, have experimented with marijuana at one time or another (Johnston et al., 1986; Johnston, Bachman, & O'Malley, 1981 a, 1981 b; Miller et al., 1983; NIDA, 1986), and the vast majority of these young people do not subsequently become drug abusers. Little is known about the relative psychosocial adjustment of adolescents who have experimented with drugs on an occasional basis and of adolescents who have avoided drugs entirely. Indeed, a number of researchers have suggested that occasional drug use among adolescents may be best understood as a manifestation of developmentally appropriate experimentation. Newcomb and Bentler (1988), for example, have observed that one defining feature of adolescence is a quest for or establishment of independence and autonomous identity and functioning. This may involve experimentation with a wide range of behaviors, attitudes, and activities before choosing a direction and way of life to call one's own. This process of testing attitudes and behavior may include drug use. In fact, experimental use of various drugs, both licit and illicit, may be considered a normative behavior among United States teenagers in terms of prevalence, and from a developmental task perspective. (p. 214, emphasis added)
These empirical and developmental considerations suggest that the relations between psychological variables and level of drug use may not be linear at all. To the extent that drug experimentation may represent normative behavior during the prolonged adolescent period, as individuals seek a sense of self and possibility, it may be wrong to pathologize adolescents who experiment with drugs by assuming that they fall between nonusers and drug abusers on a continuum of psychosocial adjustment. To evaluate this conceptual possibility in the present study, we identify and contrast discrete groups of nonusers, experimenters, and frequent drug users. Additionally, we employ quadratic regression methods to formally test for curvilinear relations, when the data indicate that such relations may exist. These approaches permit the emergence of findings not discernible through conventional correlational methods with their assumption of linearity.
Subjects were 101 18-year-olds, 49 boys and 52 girls, from an initial sample of 130 participating in a longitudinal study of ego and cognitive development. The subjects were initially recruited into the study at age 3, while attending either a university-run nursery school or a parent cooperative nursery school in the San Francisco Bay area. They were assessed on wide-ranging batteries of psychological measures at ages 3, 4, 5, 7, 11, 14, and 18 (see J. H. Block & J. Block, 1980, for an extended description of the study). Because so few subjects were lost over the years, there can be little influence of differential attrition.
The subjects live primarily in urban settings and are heterogeneous with respect to social class and parent education. About two thirds are White, one fourth are Black, and one twelfth are Asian. Not all subjects are used in all analyses to be reported, as will be discussed.
Measuring Drug Use
Information about drug use was collected at age 18 during individual interviews with the subjects. Skilled clinicians conducted these interviews, which ranged over a variety of topics including schoolwork, peer relations, family dynamics, personal interests, dating experiences, and so on. Total interview time was typically four hours per subject, and all interviews were videotaped.
The subjects were asked whether they "smoked pot or used it in another form." Their responses were coded from the interview videotapes as follows: (0) never used marijuana; (1) used once or twice; (2) used a few times; (3) used once a month; (4) used once a week; (5) used two or three times a week; and (6) used daily. The subjects
Correspondence concerning this article should be addressed to Jonathan
Shedler or Jack Block, Department of Psychology, University of California,
Berkeley, CA 94720.
were also given a list of other substances and were asked to check which (if any) they had used at least once on a "recreational" basis. The list included inhalants (e.g., glue, nitrous oxide), cocaine, hallucinogens, barbiturates, amphetamines, tranquilizers, heroin, and an open-ended category for "other" drugs not specifically listed.
Although self-report data on drug use are always subject to underreporting, the findings of a number of investigations indicate that such data have high validity (e.g., Block et al., 1988; Haberman, Josephson, Zanes, & Elinson, 1972; Jessor & Jessor, 1977; Perry, Killen, & Slinkard, 1980; Single, KandeI, & Johnston, 1975). Additionally, there is every reason to believe that the subjects in this investigation answered our questions honestly. The interviewers were skilled in gaining rapport and in eliciting information without inducing discomfort. Moreover, the subjects had been involved in the longitudinal study from earliest childhood; they not only had been assured that their individual responses would be held in confidence, but they knew from years of prior experience that this promise had been honored.
Age 18 assessment. At age 18, the personality characteristics of each subject were described by four psychologists, using the standard vocabulary of the California Adult Q-sort (Block, 1961/1978). The CAQ is a personality assessment instrument that allows psychologists to provide comprehensive personality descriptions in a conceptually systematic, quantifiable, and readily comparable form. The CAQ consists of 100 personality-descriptive statements, each printed on a separate index card. The psychologist sorts these statements into a fixed nine-step distribution, according to their evaluated salience vis-a-vis the person being described. Thus, the CAQ yields a score of 1 through 9 for each of 100 personality-descriptive statements; higher scores indicate that a statement is relatively characteristic of a person, and lower scores indicate that it is relatively uncharacteristic. The validity and usefulness of Q-sort personality descriptions has been demonstrated frequently (see, e.g., Bem & Funder, 1978; Block, 1961/1978; Block, 1971; Gjerde, Block, & Block, 1988; Mischel, Shoda, & Peake, 1988). 1
The psychologists based their CAQ descriptions of each subject on observations made while administering a variety of experimental procedures designed to tap various aspects of psychological functioning. These psychologists were not the interviewers who gathered information about drug use; they had no knowledge of subjects' drug use or of any other information elicited during the interviews. Each of the four psychologists who provided CAQ-based personality descriptions saw the subjects in a different assessment context, so that four entirely independent Q-sort descriptions were available per subject. The scores assigned to each Q sort item were then averaged across the four psychologists, to yield a final, composite Q-sort for each subject. These composite Q-sorts thus represent the consensual judgment of four independent assessors. The reliabilities of the composite Q-sorts differed somewhat from subject to subject, and were of the order of .70 to .90.
Childhood assessments. At ages 7 and 11, the personality characteristics of the subjects were described in a similar manner, each time by entirely different sets of psychologists, using the standard vocabulary of the California Child Q-sort (CCQ). The CCQ is an age-appropriate modification of the California Adult Q-sort, and consists of statements describing the personality, cognitive, and social characteristics of children (see J. Block & J. H. Block, 1980; J. H. Block and J. Block, 1980). At age 7, the standard 100-item CCQ was used; at age 11, an abridged 63-item version was used. Three psychologists observed the children at age 7, and five psychologists observed the children at age 11, while administering a variety of age appropriate experimental procedures. The scores assigned to the CCQ items were averaged across the psychologists to produce a composite Q sort for age 7 and a composite Q-sort for age 11. Again, the reliabilities of the composite Q-sorts were of the order of .70 to .90.
Measuring the Quality of Parenting
During each joint assessment session, the children were given a variety of age-appropriate tasks to perform, such as assembling objects from wooden blocks, arranging plastic pieces according to shape and color, solving mazes, and so on. The parents were instructed to respond to their child's eventual difficulties with the tasks and to provide whatever help they felt was needed. The tasks were designed to be of interest to parent and child, to be appropriately challenging to the child, and to be readily understandable to all parents. The order of the sessions and the order of tasks within sessions were counterbalanced. The joint assessment procedure has been described in more detail elsewhere (Block & Block, 1971; Gjerde, 1988).
Parent and child were left alone to work on the tasks while a trained observer watched the interaction through a one-way mirror. Additionally, the sessions were videotaped. After the session, the observer described the parent's manner of interacting with the child using a 49-item Parent Child Interaction Q-sort (PCIQ) specially developed for this purpose (Block & Block, 1971). A second observer provided
Rates of Drug Use
The primary purpose of this study was to investigate the relations between drug use and psychological characteristics. However, it is first useful to consider the rate of drug use in the sample in absolute terms.
Of the 101 subjects for whom information about drug use was available, 68% had tried marijuana (four years earlier, 51% of the subjects had used marijuana; see Keyes & Block, 1984). Thirty-nine percent of the subjects reported using marijuana once a month or more, and 21% reported using it weekly or more than weekly. These figures are comparable to figures obtained in nationwide probability samples of adolescents and young adults (Johnston et al., 1986; Johnston, Bachman & O'Malley, 1981 a, 1981 b; Miller et al., 1983; NIDA, 1986).
Approximately 37% of the subjects reported trying cocaine, and 25% reported trying hallucinogens. Approximately 10% of the subjects reported trying amphetamines, barbiturates, tranquilizers, or inhalants. Only one subject reported that she had used heroin.
Creation of Comparison Groups
Based on the drug use information collected at age 18, the subjects were divided into three nonoverlapping groups, as follows.
Abstainers were defined as subjects who had never tried marijuana or any other drug. This group contained 29 subjects, 14 boys and 15 girls. 2
Experimenters were defined as subjects who had used marijuana "once or twice," "a few times," or "once a month," and who had tried no more than one drug other than marijuana. This group contained 36 subjects, 16 boys and 20 girls. The mean number of other drugs tried by the subjects in this group was 0.31 (i.e., 11 of the 36 subjects had tried one drug other than marijuana).
Frequent users were defined as subjects who reported using marijuana frequently, that is, once a week or more, and who had tried at least one drug other than marijuana. This group contained 20 subjects, 11 boys and 9 girls. The mean number of other drugs tried by the subjects in this group was 2.70. 3
Sixteen subjects "fell between the cracks" of the classification scheme, and did not meet the definitional criteria for any of the groups. In general, these were subjects who were excluded from the abstainer and experimenter groups because of their use of drugs other than marijuana.
The basis for the groupings derives from conceptual considerations, as well as from some recognitions derived from prior evaluation of the subjects in early adolescence
(Block et al., 1988). Obviously, a degree of arbitrariness is unavoidable in any such classification scheme; however, the results to be reported are robust with respect to the various group definitions. That is, we considered both broader and narrower definitions for the various groups (e.g., excluding from the group of experimenters subjects who had tried any drug other than marijuana, or including subjects who had tried as many as two other drugs). As long as the sample was divided into three groups that could be broadly construed in terms of nonusers, experimenters, and regular users, the pattern of results we report emerged reliably.
The groups were first compared on the control variables of socioeconomic status, as assessed by both the Duncan (1961) and Warner, Meeker, and Eells (1949) indexes and IQ, as measured by the Wechsler Preschool and Primary Scale of Intelligence (WPPSI) at age 4, the Wechsler Intelligence Scale for Children (WISC) at age 11, and the Wechsler Adult Intelligence Scale (WAIS) at age 18. No associations approaching significance were observed; consequently, these variables cannot be readily invoked to explain subsequent findings.
Personality Concomitants of Drug Use
The major findings from the age 18 personality assessment are presented in Table I, which lists mean scores for the CAQ items differentiating between frequent drug users, experimenters, and abstainers.
Findings are presented using the experimenters as a reference group, and the personality characteristics associated with the other groups are elucidated through comparison with them. The experimenters are used as a frame of reference for two reasons: (a) they constitute the largest group and reflect the pattern of drug use most typical for this sample and most typical for adolescents in the nation as a whole; (b) the group of experimenters lies between the other groups on the continuum of frequency of drug use; therefore, its use as a reference group facilitates the discernment of possible curvilinear relations between drug use and personality measures. As will be seen, this second consideration takes on considerable importance.
Personality Characteristics of Frequent Users
The frequent users were compared with the experimenters on each of the 100 Q-sort items, by means of separate tests. The number of statistically significant differences between the groups is striking and far exceeds the number
3. The most frequently mentioned drugs were cocaine, hallucinogens, and
amphetamines, which had been used by 81%, 67%, and 43% of the frequent users,
to be expected by chance. Fully 51 of the 100 Q-sort items revealed differences at the .05 significance level (hypothesis tests are two-tailed unless otherwise noted). 4
The following set of Q-sort items, all of which discriminate beyond the .05 level, serve to characterize the frequent users. The items are grouped according to general conceptual similarity. Inspection of Table I will reveal additional items that supplement this summary characterization.
Relative to experimenters, frequent users are described as not dependable or responsible, not productive or able to get things done, guileful and deceitful, opportunistic, unpredictable and changeable in attitudes and behavior, unable to delay gratification, rebellious and nonconforming, prone to push and stretch limits, self - indulgent, not ethically consistent, not having high aspirations, and prone to express hostile feelings directly.
Relative to experimenters, frequent users are also described as critical, ungiving, not sympathetic or considerate, not liked and accepted by others, not having warmth or the capacity for close relationships, having
hostility toward others, prone to avoid close relationships, distrustful of people, not gregarious, not personally charming, and not socially at ease.
Finally, frequent users are described as relatively overreactive to minor frustrations, likely to think and associate to ideas in unusual ways, having brittle ego-defense systems, self-defeating, concerned about the adequacy of their bodily functioning, concerned about their adequacy as persons, prone to project their feelings and motives onto others, feeling cheated and victimized by life, and having fluctuating moods.
Consistent with the CAQ descriptions suggesting alienation and poor impulse control, the frequent users attain significantly lower high school grade point averages than the experimenters, 2.3 versus 3.0 (p <.01).
When the Q-sort descriptions are considered as a set, the picture of the frequent user that emerges is one of a troubled adolescent, an adolescent who is interpersonally alienated, emotionally withdrawn, and manifestly unhappy, and who expresses his or her maladjustment through undercontrolled, overtly antisocial behavior.
Personality Characteristics of Abstainers
The abstainers were compared with the experimenters on each of the 100 Q-sort items, by means of separate t-tests. Once again, the number of statistically significant CAQ items well exceeds chance, with 19 of the 100 CAQ items showing differences between the groups at the .05 level.
The following Q-sort items, all significant beyond the .05 level, serve to characterize the abstainers.
The abstainers and the experimenters achieve identical high school Grade Point Averages, 3.0 in both cases.
When the Q-sort items are considered as a set, the picture of the abstainer that emerges is of a relatively tense, overcontrolled, emotionally constricted individual who is somewhat socially isolated and lacking in interpersonal skills.
Personality Antecedents of Drug Use
An unusual feature of the present study is that psychological descriptions of subjects are available from early childhood on. Moreover, the psychological descriptions obtained at different ages are wholly independent of one another. We wish to emphasize this independence: The psychologists who saw the subjects at different ages were different people, they saw the subjects under different conditions, they saw the subjects only at the age at which they served as assessors, and they had no contact with one another. Because of the safeguards taken to ensure the independence of the data, relations between psychological characteristics observed at age 18 and psychological characteristics observed in early childhood must be attributed to continuities in psychological development over time (and not to artifacts of the research design).
On the basis of the CAQ descriptions obtained at age 18, a priori directional hypotheses were generated for virtually all of the age 11 and age 7 California Child Q-Sort items. Specifically, it was hypothesized that abstainers would show signs of impulse overcontrol, and frequent users would show signs of impulse undercontrol, relative to experimenters; and that abstainers and frequent users would both show signs of interpersonal alienation and psychological distress, relative to experimenters. 5 In view of the existence of directional hypotheses and the independence of the data collected at different ages, one-tailed statistical tests were employed in evaluating the childhood CCQ data. Table 2 lists the mean scores for the age 11 CCQ items discriminating between the groups (recall that an abridged 63-item Q-sort was used at the age 11 assessment, so fewer significant relationships can be expected), and Table 3 lists the mean scores for the age 7 CCQ items discriminating between groups.
The Childhood Personality of Frequent Users
The frequent users were compared with the experimenters on each of the CCQ items by means of separate t-tests. At age 11, frequent users were described (in comparison to experimenters) as visibly deviant from their peers, emotionally labile, inattentive and unable to concentrate, not involved in what they do, stubborn (preceding items significant at the .05 level), unhelpful and uncooperative, pushing and stretching limits, not eager to please, immobilized under stress, not curious and open to new experience, likely to give up easily, likely to withdraw under stress, not having high performance standards, suspicious and distrustful, and overreactive to minor frustrations (preceding items significant at the .10 level).
At age 7, the frequent users were described as not getting along well with other children, not showing concern for moral issues (e.g., reciprocity, fairness), having bodily symptoms from stress, tending to be indecisive and vacillating, not planful or likely to think ahead, not trustworthy or dependable, not able to admit to negative feelings, not self-reliant or confident (preceding items significant at the .05 level), preferring nonverbal methods of communication, not developing genuine and close relationships, not proud of their accomplishments, not vital or energetic or lively, not curious and open to new experience, not able to recoup after stress, afraid of being deprived, appearing to feel unworthy and "bad," not likely to identify with admired adults, inappropriate in emotive behavior, and easily victimized and scapegoated by other children (preceding items significant at the .10 level).
In short, the frequent users appear to be relatively maladjusted as children. As early as age 7, the picture that emerges is of a child unable to form good relationships, who is insecure, and who shows numerous signs of emotional distress. These data indicate that the relative social and psychological maladjustment of the frequent users predates adolescence, and predates initiation of drug use.
The Childhood Personality of Abstainers
The abstainers were compared with the experimenters on each of the CCQ items. At age 11, the abstainers were described as relatively fearful and anxious, using and responding to reason, not physically active, not vital or energetic or lively, inhibited and constricted, not liking to compete, not curious and open to new experiences, not interesting or arresting, physically cautious, neat and orderly (implies fussiness), anxious in unpredictable environments, not having a rapid personal tempo, looking to adults for help and direction, not responsive to humor, not self-assertive, not self-reliant or confident, shy and reserved (preceding items significant at the .05 level), cold and unresponsive, immobilized under stress, obedient and compliant, not calm or relaxed, planful and likely to think
6. The use of one-tailed tests may seem insufficiently conservative to Some. However, the thrust of the findings would be the same regardless of whether one- or two-tailed statistical tests were employed. Moreover, there is an inherent trade-off between the level of Type I and Type II errors. Given an exploratory study involving difficult-to-obtain data, it seems scientifically strategic to lessen the likelihood of Type II errors, rather than to overprotect against Type I errors (cf. Block, 1960).
At age 7, the abstainers were described as relatively eager to please, inhibited and constricted, conventional in thought, neat and orderly, planful and likely to think ahead, not verbally expressive, not seeking to be independent and autonomous (preceding items significant at the .05 level), not proud of their accomplishments, not physically active, immobilized under stress, obedient and compliant, not self-assertive, not competent and skillful, and not creative (preceding items significant at the .10 level).
These descriptions present a picture of a child who is relatively overcontrolled, timid, fearful, and morose.
While the characterizations of these children as "anxious," "inhibited," and "immobilized under stress" are telling, more telling, perhaps, may be the descriptions of what these children are not; relative to the reference group of experimenters, they are not warm and responsive, not curious and open to new experience, not active, not vital, and not cheerful.
Quality of Parenting
Table 4 lists the mean scores for the PCIQ items that discriminate between the mothers of frequent users, experimenters, and abstainers. These data reflect direct observations of mother-child interactions when the subjects were five years of age. Hypothesis tests are two tailed.
Quality of Parenting: Frequent Users
Compared with the mothers of the experimenters, the mothers of the frequent users are described as hostile, not spontaneous with their children, not responsive or sensitive to their children's needs, critical of their children and rejecting of their ideas and suggestions, not supportive and encouraging of their children, tending to dramatize their teaching, making the test situation grim and distasteful rather than fun, pressuring their children to work at the tasks, underprotective of their children, overly interested in and concerned with their children's performance, conductingthe session in such a way that their children do not enjoy it (preceding items significant at the .05 level), appearing to lack pride in and be ashamed of their children, seeming to be confused about what is expected of them in the test situation, conducting the session in unusual or atypical ways, not giving their children praise, and not having a clear and coherent teaching style (preceding items significant at the .10 level).
In brief the mothers of the frequent users are perceived as relatively
cold, unresponsive, and underprotective. They appear to give their children
little encouragement, while, conjointly, they are pressuring and overly
interested in their children's "performance. " The apparent net effect of this
double-bind is that they turn a potentially enjoyable interaction into a grim
and unpleasant one.
Quality of Parenting: Abstainers
Compared with the mothers of the experimenters, the mothers of the abstainers were described as hostile, not responsive or sensitive to their children's needs, critical of their children and rejecting of their ideas and suggestions, frustrated by an inability to find adequate strategies for teaching their children, not valuing their children's originality, not supportive and encouraging of their children, overly interested in, and concerned with, their children's performance, impatient with their children (preceding items significant at the .05 level), appearing to lack pride in and be ashamed of their children, seemingto be confused about what is expected of them in the test situation, conducting the session in unusual or atypical ways, not giving their children praise, not having a clear and coherent teaching style, pressuring their children to work at the tasks, and conducting the session in such a way that their children do not enjoy it (preceding items significant at the .10 level).
Like the mothers of the frequent users, these mothers are perceived as relatively cold and unresponsive. They give their children little encouragement, while, conjointly, they are pressuring and overly interested in their children's performance. Again, the apparent net effect is that they make the interaction grim and unenjoyable.
A variety of PCIQ items discriminated between fathers of abstainers and fathers of experimenters (these items are listed here and are not presented in a separate
table). Compared with the fathers of the experimenters, the fathers of the abstainers were described as relatively attentive to the cognitive elements in the test situation, not responsive or sensitive to their children's needs, not allowing open disagreement between parent and child, maintaining tight control of the session, critical of their children and rejecting of their ideas and suggestions, appearing to lack pride in and be ashamed of their children, not encouraging their children to proceed independently, not valuing their children's originality, using physical means (e.g., body language and facial expression) to communicate with their children, overly interested in, and concerned with, their children's performance, impatient with their children, conducting the session in such a way that the children do not enjoy it, not deriving pleasure from being with their children, intruding physically into their children's activities (preceding items significant at the .05 level), setting too fast a pace for their children, seeming confused about what is expected in the situation, not easy and relaxed, and pressuring their children to work at the tasks (preceding items significant at the .10 level).
The picture that emerges is of an authoritarian and domineering father who squelches spontaneity and creativity and who demands that things be done his way. He does not appear to enjoy being with his child, and he ensures that his child does not enjoy being with him.
Underlying Personality Factors: Linear and Curvilinear Relations
Up to this point, we have taken a person-centered approach in presenting our findings. That is, we focused on discrete groups of subjects (e.g., frequent users) and attempted to provide comprehensive and psychologically rich characterizations of these subjects. Such a person centered approach is congruent with the orientation of clinical practitioners and facilitates the often difficult task of translating empirical findings into usable clinical insights.
The more common analytical approach is nomothetic or variable-centered. Such an approach emphasizes variables rather than persons. In the context of the present study, drug use is treated as a continuum, and the research inquiries become: "What are the major personality variables associated with drug use?" and "What is the form of the relations between these variables and drug use?"
Person-centered and variable-centered approaches can inform and complement
one another, each illuminating different facets of the problem at hand. The
variable-centered analyses that follow examine concomitant relations only (i.e.,
relations between drug use and personality characteristics, both assessed at age
Impressionistic content analysis of the age 18 Q-sort items that discriminate between frequent users, experimenters, and abstainers suggests that these items cluster around three broad themes, having to do with (a) interpersonal relations, (b) subjectively experienced emotional distress and (c) impulse regulation. To formally evaluate the importance of these themes, the Q-sort items listed in Table I were subjected to factor analysis (principle facto method). This factor analysis yielded three conceptually interpretable factors after varimax rotation, which together accounted for 64% of the variance in the Q-sort descriptions. The items that best mark the factors are listed in Table 5. It can be seen from this table that the factors correspond to the three themes identified here.
Factor 1, which we have labeled Quality of Interpersonal Relations, reflects warm interpersonal relation versus interpersonal alienation and distrust. Factor 2, labeled subjective distress, reflects self-devaluation and emotional distress versus a sense of personal well-being Factor 3, labeled Ego-Control (J. H. Block & J. Block 1980), reflects impulse under control and impetuousness versus impulse over-control and conformity.
For each of the three factors, factor scales were constructed by averaging the relevant CAQ items, after reversing the coding of items that were negative indicators of factors. The alpha reliabilities were .96, .94, and .89 for the Quality of Interpersonal Relations, Subjective Distress, and Ego-Control factors, respectively.7
Monotonic and Nonmonotonic Relations
Examination of the Q-sort findings presented in Table 1 reveals a mixture of monotonic and nonmonotonic relations between personality characteristics and level of drug use. Specifically, scores for certain CAQ items appear to increase (or decrease) monotonically as a function of drug use, whereas scores for other CAQ items appear to manifest somewhat "U"- (or inverted "U") shaped relations with drug use. The item "Undercontrols needs and impulses; unable to delay gratification" (CAQ item 53) illustrates a monotonic relation, with frequent users receiving higher scores than experimenters, who in turn receive higher scores than abstainers. In contrast, the item "Keeps people at a distance; avoids close interpersonal relationships" (CAQ item 48) illustrates a U-shaped relationship, with abstainers and frequent users both receiving higher scores than experimenters.
The pattern of monotonic and nonmonotonic relations appears to be orderly. Specifically, it appears that (a) items reflecting the Quality of Inter-personal Relations factor show somewhat U- (or inverted U) shaped relations with level of drug use, such that experimenters are judged to have healthier interpersonal relationships than either abstainers or frequent users; (b) items reflecting the Subjective Distress factor manifest somewhat U- (or inverted U) shaped relationship with level of drug use, such that experimenters are judged to have a greater sense of emotional well-being than either abstainers or frequent users; and (c) items reflecting the Ego-Control factor are related monotonically to level of drug use, such that abstainers
are judged to be relatively overcontrolled with respect to impulse expression and frequent users are judged to be relatively undercontrolled. Finally, the U-shaped relations described here do not appear to be symmetric: Although both abstainers and frequent users are evaluated relatively unfavorably in terms of the Quality of Interpersonal Relations and Subjective Distress factors, it also appears that frequent users are evaluated less favorably by far.
To test these observations formally, three separate hierarchical multiple regressions were performed, in which level of marijuana use served as a predictor of the Quality of Interpersonal Relations, Subjective Distress, and Ego-Control scales, respectively.8 Each regression equation included both a linear and a quadratic term. The regressions were "hierarchical" in the sense that the linear term entered the regression equation first, followed by the quadratic term, which provided a test of curvilinearity. The regression analyses are based on the full sample of 101 subjects for whom drug use and personality data were available; the previous distinctions between abstainers, experimenters, and frequent users were ignored.
Table 6 presents the results of the three regression analyses. Both linear and quadratic terms contributed significantly in the regression equation to predict Quality of Interpersonal Relations, and in the regression equation to predict Subjective Distress (i.e., the quadratic terms explained significant incremental variance, over and above the variance explained by the linear terms). The significant contributions of the quadratic terms indicate that these personality factors manifest, at least to some extent, U- or inverted U-shaped relations with level of drug use. The quadratic term did not significantly contribute to the regression equation to predict Ego-Control, indicating that the relation between drug use and Ego-Control is essentially linear. Ego-Control is quite strongly related to drug use (R = .52, p <.0001), a finding similar to that obtained when the subjects were 14 years of age (see Block et al., 1988). Quality of Interpersonal Relations and Subjective Distress are also significantly related, but less strongly (R = .33, p <.005, and R = .29, p <.05, respectively).
Figure 1 graphically illustrates the best-fit regression lines to predict Quality of Interpersonal Relations, Subjective Distress, and Ego-Control. It can be seen that both Quality of Interpersonal Relations and Subjective Distress show somewhat U- (or inverted U) shaped relations with level of drug use (i.e., moderate experimentation with marijuana is associated with more positive interpersonal relationships, and greater subjective well-being, than either no marijuana use or heavy use). These U-shaped relations are clearly asymmetric, with heavy marijuana
Summary of Major Findings
On the basis of the drug use information collected at age 18, subjects were divided into nonoverlapping groups made up of frequent drug users, experimenters, and abstainers. At age 18, frequent users were observed to be alienated, deficient in impulse control, and manifestly distressed, compared with experimenters. At age 18, abstainers were observed to be anxious, emotionally constricted, and lacking in social skills, compared with experimenters.
Differences between the groups were evident during childhood as well, at the age 7 and age 11 assessments. Consistent with the age 18 findings, frequent users were judged to be relatively insecure, unable to form healthy relationships, and emotionally distressed as children, compared with experimenters. Also consistent with the age 18 findings, abstainers were judged to be relatively anxious, inhibited, and morose as children, compared with experimenters.
Additionally, both frequent users and abstainers were judged to have received poorer maternal parenting than experimenters, as assessed by direct observations of mother-child interactions when the subjects were five years old. Compared with the mothers of experimenters, both the mothers of frequent users and the mothers of abstainers were perceived to be cold, critical, pressuring, and unresponsive to their children's' needs.
There were no noteworthy findings involving the fa-
At age 18, the underlying personality factors relevant to an understanding of drug use appear to be Quality of Interpersonal Relations, Subjective Distress, and Ego-Control. When level of marijuana use is treated as a continuous variable, both the Quality of Interpersonal Relations and the Subjective Distress factors show somewhat U-shaped relationships with level of marijuana use. Ego-Control appears to be linearly related to level of marijuana use, with heavy users showing the poorest impulse control.
On the Relation Between Drug Use and Psychological Health
When the psychological findings are considered as a set, it is difficult to escape the inference that experimenters are the psychologically healthiest subjects, healthier than either abstainers or frequent users. Psychological health is meant here in a global and nonspecific sense, consistent with ordinary conversational usage, and consistent also with empirical recognitions by mental health researchers that a general psychological health/psychological distress factor underlies diverse clinical syndromes (e.g., Dohrenwend, Shrout, Egri, & Mendelsohn, 1980; Tanaka & Huba, 1984; Watson & Clark, 1984). The inference that there is an inverted U-shaped relation between level of drug use and psychological adjustment is supported by the patterns of Q-sort items characterizing abstainers, experimenters, and frequent users, and also by the U- (and inverted U) shaped relations observed between level of marijuana use and the two personality factors, Quality of Interpersonal Relations and Subjective Distress.
The finding that frequent users are relatively maladjusted has been obtained by many other investigators. The finding that abstainers also show some signs of relative maladjustment (albeit of a very different kind) is, perhaps, unusual. In order to understand this latter finding, we suggest it is important to consider both the meaning of drug use within adolescent peer culture, as well as the psychology of adolescent development.
First, it is necessary to recognize that in contemporary American culture, there is wide prevalence and apparent acceptability of marijuana use in late adolescence. The majority of the 18-year-olds in our sample approximately two thirds had tried marijuana at one time or another. Such a high usage rate is consistent with the findings from national probability samples (Johnston et al., 1986; Johnston, Bachman, & O'Malley, 1981a, 1981b; Miller et al., 1983; NIDA, 1986). Thus, some experimentation with marijuana cannot be considered deviant behavior for high school seniors in this culture at this time. In a statistical sense it is not trying marijuana that has become deviant.
Second, the extended period of adolescence is a time of transition, a time when young people face the developmental task of differentiating themselves from parents and family and forging independent identities. Experimenting with values and beliefs, exploring new roles and identities, and testing limits and personal boundaries are normative behaviors during adolescence, and they serve important developmental ends (cf. Erikson, 1968; Havinghurst, 1972).
Given these factors---the ubiquity and apparent acceptability of marijuana in the peer culture and the developmental appropriateness of experimentation and limit-testing during adolescence---it is not surprising that by age 18, psychologically healthy, sociable, and reasonably inquisitive individuals would have been tempted to try marijuana. We would not expect these essentially normal and certainly normative adolescents to abuse the drug (and it is crucial to distinguish between experimentation and abuse) because they would have little need for drugs as an outlet for emotional distress or as a means of compensating for lack of meaningful human relationships---but we should not be surprised if they try it. Indeed, not to do so may reflect a degree of inhibition and social isolation in an 18-year-old.9
Although no prior study has focused explicitly on the psychology of adolescent abstainers, there is some empirical precedent for the present finding that abstainers are not the most well-adjusted of adolescents. Hogan, Mankin, Conway, and Fox (1970), using a self-report personality inventory, compared marijuana users with nonusers in a college population and found that users "are more socially skilled, have a broader range of interests, are more adventuresome, and more concerned with the feelings of others" (p. 63). Nonusers were characterized as "too deferential to external authority, narrow in their interests, and overcontrolled" (p. 61). These findings, based on entirely different methodology, are strikingly similar to our own. In a similar vein, Bentler (1987) reported a small but reliable association between marijuana use and the development of a positive self-concept.
We do not suggest that the inverted U-shaped relation between level of drug use and psychological health expresses a fundamental psychological "principle" or "law." Rather, we view this finding as a function of historical and social circumstances-specifically, of the current prevalence of drug use in this culture, conjoined with the developmentally appropriate propensity of adolescents to explore and experiment. Gergen's (1973) arguments regarding "social psychology as history" may be applicable here.
The U-shaped relations between psychological health and drug use are reminiscent of U-shaped relations between psychological health and alcohol use noted in an earlier generation of subjects. Thus, Jones (1968, 1971) found that moderate drinkers were psychologically healthier than either problem drinkers or abstainers.
Toward an Understanding of Frequent Drug Users
If drug use among experimenters reflects normative adolescent exploration and inquisitiveness, it reflects something quite different in the group we have labeled frequent users. Frequent users differ profoundly from the comparison group of experimenters, and indications of their social and psychological maladjustment are pervasive.
At age 18, the frequent users appear unable to invest in, or derive pleasure from, meaningful personal relationships. Indeed, they seem fortified against the possibility of such relationships through their hostility, distrust, and emotional withdrawal. Neither do they appear to be capable of investing in school and work, or of channeling their energies toward meaningful future goals. They are, then, alienated from the "love and work" that lend a sense of satisfaction and meaning to life. Consistent with this, they appear to feel troubled and inadequate. It is easy to see how these characteristics could create a vicious cycle: Feeling troubled and inadequate, these adolescents withdraw from work and relationships, and alienated from work and relationships, they feel all the more troubled and inadequate.
Such a pattern of alienation can be expected to go hand in hand with an impaired ability to control and regulate impulses. When there is little investment in either work or relationships, that is, when there is little connection with those things that give life a sense of stability and purpose, then the impulses of the moment become paramount. The impulses are not adequately transformed or mediated by a broader system of values and goals because such a system is lacking.
Shapiro (1965) has written eloquently on this point:
The normal person "tolerates" frustration or postpones the satisfaction of his whim at least in part because he is also interested in other things; his heart is set on goals and interests that are independent of the immediate frustration or extend beyond the whim and supercede it in subjective significance. This is not simply a matter of intellectual choice. Rather, the existence of these general goals and interests automatically provides a perspective, a set of dimensions in which a passing whim or an immediate frustration is experienced. In the absence of such goals and interests, the immediately present frustration or the promised satisfaction must, accordingly, gain in subjective significance, and under these conditions forbearance or tolerance is unthinkable. (pp. 145-146)
Drugs would have a special appeal to the alienated and impulsive individuals we are discussing. The temporary effects of various drugs "numb out" feelings of
isolation and inadequacy; they offer transient gratification to individuals who lack deeper and more meaningful gratifications (i.e., through relationships and work); and given the poor ability of these individuals to regulate impulse, the urge toward drug use would meet with little inner resistance and would be little modified by a broader value system.
The traits that characterize the frequent users can be seen, then, to form a theoretically coherent syndrome, characterized by the psychological triad of alienation, impulsivity, and subjective distress. The data indicate that the roots of this syndrome predate adolescence and predate initiation of drug use.
As early as age 7, the frequent users show signs of the alienation, undercontrol, and emotional distress that will characterize them at age 18. Relative to experimenters, they are described as not getting along well with other children, as not developing genuine and close relationships, as not showing concern for moral issues, as not trustworthy or dependable, as having bodily symptoms from stress, as afraid of being deprived, as appearing to feel unworthy, as inappropriate in emotive behavior, and so on (see Table 2). The data clearly indicate, then, that the relative maladjustment of the frequent users precedes the initiation of drug use.
This relative maladjustment perhaps may be traced, at least in part, to the maternal parenting that the frequent users received, as assessed by direct observations of mother-child interactions when the subjects were five years of age. Relative to the mothers of experimenters, the mothers of frequent users were perceived as hostile, critical and rejecting, and not sensitive or responsive to their children's needs. Moreover, they seemed to place their children in a double-bind: Although they gave their children little support and encouragement, they were simultaneously pressuring and overly concerned with their children's "performance." 10
Adolescents who have never experimented with marijuana or any other drug have not been the subject of research attention, if only because their behavior does not pose an obvious, confronting societal problem and because it has been presumed categorically that not using drugs goes hand in hand with psychological health. However, our data suggest that, relative to experimenters, abstainers in late adolescence are somewhat maladjusted.
Unlike the patent, blatant maladjustment of frequent drug users, however, the psychological inadequacies of abstainers are largely a private matter, limiting of life as it is led, and do not attract societal attention. The constriction, uneasiness with affect, and interpersonal deficiencies of abstainers are recognizable more by way of omission than commission.
By omission, we refer to personal potentialities that seem to remain unfulfilled, specifically, potentialities for emotional gratifications, friendship, and human warmth and closeness. It is the relative capacity (or rather, incapacity ) to experience these positive qualities of life that distinguishes abstainers from experimenters. Relative to experimenters, the abstainers are described at age 18 as overcontrolled and prone to delay gratification unnecessarily, not able to enjoy sensuous experiences, prone to avoid close interpersonal relationships, not gregarious, not liked and accepted by people, and so on. Thus, their avoidance of drugs seems less the result of "moral fiber" or successful drug education than the result of relative alienation from their peers and a characterological overcontrol of needs and impulses.
It seems likely that the relative overcontrol and emotional constriction of the abstainers serves the psychological purpose of containing or masking feelings of vulnerability. There is some evidence for this hypothesis in the age 18 personality descriptions of our subjects, for the abstainers are described at this age as relatively anxious. However, the strongest support for this hypothesis comes from the childhood data, when the relative maladjustment of these subjects is most manifest. At age 11, for example, prior to initiation of drug use, the abstainers are described (relative to experimenters) as fearful and anxious, inhibited and constricted, immobilized under stress, anxious in unpredictable environments, not curious and open to new experience, not vital or energetic or lively, not confident, not responsive to humor, and not cheerful. These traits would appear to reflect a susceptibility to anxiety and, perhaps, a consequent avoidance of circumstances or behaviors perceived as risky.
The hypothesis that emotional constriction serves the purpose of "containing" feelings of vulnerability is reinforced further by observations of mother-child interactions when the children were age five: The mothers of the abstainers were perceived (relative to the mothers of the experimenters) as unresponsive, cold, critical, and rejecting. Such parenting clearly has negative implications for psychological resiliency and well-being. It is interesting that the descriptions of the mothers of abstainers are strikingly similar to the descriptions of the mothers of the frequent users. These results at least raise the possibility that the behavior of the abstainers and the behavior of the frequent users, manifestly so different, represent alternative reactions to an underlying psychological vulnerability.
The question of why abstainers and frequent users traveled such different developmental pathways remains an open one. Character formation is a matter of complex temperamental and developmental vicissitudes, and such questions have no straightforward answer. Our findings suggest the speculation that the fathers of the abstainers; played a telling role in their character development. They acted toward their children in ways that would seem to increase anxiety, but their stern, authoritarian, and autocratic manner may have also provided a model for dealing with that anxiety. It is conceivable that the children internalized their fathers' attitudes, adopting an attitude toward their own impulses that paralleled their father's attitudes toward them.
Implications for Theory and Social Policy
Taken as a whole, the present data indicate that drug use and drug abstinence have theoretically coherent antecedents and must be understood within the context of an individual's total psychology. Because experimenters and frequent users are, psychologically, very different kinds of people, the meaning of drug use in these two groups is very different. In the case of experimenters, drug use appears to reflect age appropriate and developmentally understandable experimentation. In the case of frequent users, drug use appears to be a manifestation of a more general pattern of maladjustment, a pattern that appears to predate adolescence and predate initiation of drug use. Undoubtedly, drug use exacerbates this earlier established pattern but, of course, the logic of a longitudinal research design precludes invocation of drug use as causing this personality syndrome.
Current theories (e.g., Akers, Krohn, Lanze-Kaduce, & Radosevich, 1979; Jessor & Jessor, 1977, 1978; Kaplan, Martin, & Robins, 1982) tend to emphasize the role of peers in influencing drug use. The importance of peers in providing an encouraging surround for experimentation cannot be denied, but "peer-centered" or "environmental" explanations of problem drug use seem inadequate, given the present longitudinal findings (cf. Margulies, Kessler, & Kandel, 1977).
The discovery of psychological antecedents predating drug use (see also Block et al., 1988; Kellam et al., 1975; Kellam et al., 1983; Kellam, Ensminger, & Simon, 1980) can to some extent be integrated with explanations of drug use that emphasize environmental factors, once it is recognized that individuals, from early childhood, actively construct and seek out environments that, given their essential personality, motivational, and intellectual characteristics, they find particularly harmonious and vivifying (see, e.g., Scarr & McCartney, 1983). Rather than being passive recipients of "environmental" influences, by the time of adolescence, individuals are already appreciably formed psychologically and are actively evoking, actively seeking, and actively forging the circumstances that will suit them and that will then, in an adventitious way, "impinge" on them.
The recognition that problem drug use (and, for that matter, abstinence) has developmental antecedents, that it is a part of a broad and theoretically coherent psychological syndrome, and that it is not adequately explained in terms of peer influence has important implications for social policy.
Current social policy seems to follow from the assumption that peer influence leads to experimentation, which in turn leads to abuse. Thus, efforts at drug education are aimed at discouraging experimentation by emphasizing the need to "just say no" to peer influence. But adolescent experimentation in and of itself does not appear to be personally or societally destructive (see also Kandel, Davies, Karus, & Yamaguchi, 1986, and Newcomb & Bentler, 1988), and peer influence does not appear to be an adequate explanation for problem drug use. Moreover, given the developmental tasks of the prolonged adolescent period, efforts aimed at eliminating adolescent experimentation are likely to be costly and to meet with limited success.
Current efforts at drug "education" seem flawed on two counts. First, they are alarmist, pathologizing normative adolescent experimentation and limit-testing, and perhaps frightening parents and educators unnecessarily. Second, and of far greater concern, they trivialize the factors underlying drug abuse, implicitly denying their depth and pervasiveness. For so long as problem drug use is construed primarily in terms of "lack of education," so long is attention diverted from its disturbing psychological underpinnings: the psychological triad of alienation, impulsivity, and distress. Paradoxically, then, the "just say no" approach may be concerned with a "problem" that, from a developmental viewpoint, need not be seen as alarming (adolescent experimentation), and it may be dismayingly oblivious to a serious problem that is extremely alarming (the ubiquity of the psychological syndrome that appears to underlie problem drug use).
The concept of drug "education" may have its current popular appeal in part because the link between the problem (drugs) and the attempted solution (drug education) is self-evident and thus reassures concerned parents, educators, and policymakers that "something is being done." But educational approaches to drug prevention have had limited success (Tobler, 1986), and society's limited resources might better be invested in interventions focusing on the personality syndrome underlying problem drug use.
Given current understandings of personality development, it would seem that the psychological triad of alienation, impulsivity, and distress would be better addressed through efforts aimed at encouraging sensitive and empathic parenting, at building childhood self-esteem, at fostering sound interpersonal relationships, and at promoting involvement and commitment to meaningful goals. Such interventions may not have the popular appeal of programs that appear to tackle the drug problem "directly," but may have greater individual and societal payoff in the end.
The finding that experimenters are the psychologically healthiest adolescents, and the observation that some drug experimentation, in and of itself, does not seem to be psychologically destructive, may sit badly with some. In particular, it may sit badly with drug counselors who "know" from clinical experience that there is no level of drug use that is safe, that it is dangerous to suggest otherwise, and that the most effective intervention is one aiming at total abstinence. To avoid any misunderstanding, we wish to make clear that there is no contradiction between this therapeutic perspective and the findings we have reported. On the contrary, we are in agreement with the therapeutic perspective.
The present data indicate that in a nonselected late adolescent sample, occasional experimentation with marijuana is not personally or societally destructive. This view is supported by longitudinal studies of the consequences of drug use (as well as by the present study of the antecedents and concomitants of drug use; see, e.g., Kandel et al., 1986; Newcomb & Bentler, 1988), and by the fact that the majority of adolescents in the United States have experimented with marijuana but have not subsequently become drug abusers. The apparent contradiction between clinical wisdom, on the one hand, and the present findings, on the other, is resolved when it is recognized that individuals who present themselves for drug treatment are not representative of the general population of adolescents, but instead constitute a special, highly selected subpopulation. The psychological meaning of drug use is very different for this fractional group existing within the larger population of adolescents. For them, experimentation with drugs is highly destructive because drugs easily become part of a broader pathological syndrome. For adolescents more generally, some drug experimentation apparently does not have psychologically catastrophic implications.
In closing, one final clarification is in order. In presenting research on a topic as emotionally charged as drug use, there is always the danger that findings may be misinterpreted or misrepresented. Specifically, we are concerned that some segments of the popular media may misrepresent our findings as indicating that drug use might somehow improve an adolescent's psychological health. Although the incorrectness of such an interpretation should be obvious to anyone who has actually read this article, our concern about media misrepresentation requires us to state categorically that our findings do not support such a view, nor should anything we have said remotely encourage such an interpretation.
Akers, R. L., Krohn, M. D., Lanze-Kaduce, L., & Radosevich, M. (1979). Social learning and deviant behavior: A specific test of a general theory. American Sociological Review, 44. 636-655.
Bern, D. J., & Funder, D. C. (1978). Predicting more of the people more of the time: Assessing the personality of situations. Psychological Review, 85, 485-501.
Bentler, P. M. (1987). Drug use and personality in adolescence and young adulthood: Structural models with nonnormal variables. Child Development, 58, 65-79.
Block, J. (1960). On the number of significant findings to be expected by chance. Psychometrika, 25, 369-380.
Block, J. (1971). Lives through time . Berkeley, CA: Bancroft.
Block, J. (1978). The Q-sort method in personality assessment and psychiatric research. Palo Alto, CA: Consulting Psychologists Press. (Original work published 1961)
Block, J., Block, J. H., & Keyes, S. (1988). Longitudinally foretelling drug usage in adolescence: Early childhood personality and environmental precursors. Child Development, 59, 336-355.
Block, J. H., & Block, J. (1971). Manual for the Parent-Child Interaction Procedure. Berkeley: University of California, Department of Psychology. CA.
Block, J. H., & Block, J. (1980). The role of ego-control and ego-resiliency in the organization of behavior. In W. A. Collins (Ed.), Minnesota symposia on child psychology (Vol. 13, pp. 39-101). Hillsdale, NJ: Erlbaum.
Brook, J. S., Gordon, A. S., & Whiteman, M. (1985). Stability of personality during adolescence and its relationship to stage of drug use. Genetic, Social, and General Psychology Monographs, 111, 317-330.
Brook, J. S., Whiteman, M., Gordon, A. S., & Cohen, P. (1986). Dynamics of childhood and adolescent personality traits and adolescent drug use. Developmental Psychology, 22, 403-414.
Bush, P. J., & Iannotti, R. (1965). The development of children's health orientations and behaviors: Lessons for substance use prevention. In C. L. Jones & R. J. Battjes (Eds.), Etiology of drug abuse: Implications for prevention (Research Monograph No. 56, pp. 45-74). Rockville, MD: National Institute on Drug Abuse.
Cox, W. M. (1985). Personality correlates of substance abuse. In M. Galizio & S. A. Maisto (Eds.), Determinants of substance abuse: Biological, psychological, and environmental factors (pp. 209-246). New York: Plenum.
Dohrenwend, B. S., Shrout, P. E., Egri, G., & Mendelsohn, F. S. (1980). Nonspecific psychological distress and other dimensions of psychopathology. Archives of General Psychiatry 37, 1229-1236.
Duncan, 0. (1961). A socioeconomic index for all occupations. In A. J. Reiss, Jr. (Ed.), Occupation and social status (pp. 109-138). New York: Free Press.
Erikson, E. H. (1968). Identity: Youth and crisis. New York: Norton.
Gergen, K. J. (1973). Social psychology as history. Journal of Personality and Social Psychology, 36, 309-320.
Gjerde, P. F. (1988). Parental concordance on child-rearing and the interactive emphases of parents: Gender differentiated relationships during the preschool years. Developmental Psychology, 24, 700-706.
Gjerde, P. F., Block, J., & Block, J. H. (1988). Depressive symptomatology and personality during late adolescence: Gender differences in the externalization-internalization of symptom expression. Journal of Abnormal Psychology, 97, 475-486.
Haberman, P. W., Josephson, E., Zanes, A., & Elinson, J. (1972). High school drug behavior: A methodological report on pilot studies. In S. Einstein & S. Allen (Eds.), Proceedings of the First International Conference on Student Drug Surveys (pp. 103-121). Farmingdale, NY: Baywood.
Haley, J. (1976). Problem-solving therapy. New York: Harper & Row.
Havinghurst, R. J. (1972). Developmental tasks and education (3rd ed.). New York: McKay.
Hawkins, J. D., Lishner, D. M., & Catalano, R. F. (1985). Childhood predictors and-the prevention of adolescent substance abuse. In C. L. Jones& R. J. Battjes (Eds.), Etiology of drug abuse: Implications for prevention (Research Monograph No. 56, pp. 75-125). Rockville, MD: National Institute on Drug Abuse.
Hoffman, L. (1981). Foundations of family therapy: A conceptual framework for systems change. New York: Basic Books.
Hogan, R., Mankin, D., Conway, J., & Fox, S. (1970). Personality correlates of undergraduate marijuana use. Journal of Consulting and Clinical Psychology, 35, 58-63.
Jessor, R. (1979). Marijuana: A review of recent psychosocial research. In R. L. Dupont, A. Goldstein, & J. O'Donnell (Eds.), Handbook on drug abuse (pp. 337-355). Washington, DC: Government Printing Office.
Jessor, R., Chase, J. A., & Donovan, J. E. (1980). Psychosocial correlates of marijuana use and problem drinking in a national sample of adolescents. American Journal of Public Health, 70, 604-613.
Jessor, R., & Jessor, S. L. (1977). Problem behavior and psychological development: A longitudinal study of youth. New York: Academic Press.
Jessor, R., & Jessor, S. L. (1978). Theory testing in longitudinal research on marijuana use. In D. B. Kandel (Ed.), Longitudinal research on drug use. Empirical findings and methodological issues (pp. 41-71). Washington, DC: Hemisphere.
Johnston, L. D., Bachman, J. G., & O'Malley, P. M. (1981a). Highlights from student drug use in America 1975-1981. Rockville, MD: National Institute on Drug Abuse.
Johnston, L. D., Bachman, J. G., & O'Malley, P. M. (1981b). Student drug use in America 1975-1981, Rockville, MD: National Institute on Drug Abuse.
Johnston, L. D., O'Malley, P. M., & Bachman, J. G. (1984). Drugs and American high school students, 1975-1983 (National Institute of Drug Abuse, DHHS Publication No. ADM 84-1317). Washington, DC: Government Printing Office.
Johnston, L. D., O'Malley, P. M., & Bachman, J. G. (1986). Drug use among American high school students, college students, and other young adults: National trends through 1985. Rockville, MD: National Institute on Drug Abuse.
Jones, C. L., & Battjes, R. J. (Eds.). (1985). Etiology of drug abuse. Implications for prevention. (Research Monograph No. 56; DHHS Publication No. ADM 85-1335). Rockville, MD: National Institute of Drug Abuse.
Jones, M. C. (1968). Personality correlates and antecedents of drinking patterns in adult males. Journal of Consulting and Clinical Psychology 31, 1-12.
Jones, M. C. (1971). Personality antecedents and correlates of drinking patterns in women. Journal of Consulting and Clinical Psychology, 36, 61 69.
Kandel, D. B. (1978). Convergences in prospective longitudinal surveys of drug use in normal populations. In D. Kandel (Ed.), Longitudinal research on drug use: Empirical findings and methodological issues (pp. 3-38). Washington, DC: Hemisphere.
Kandel, D. B. (1980). Drug and drinking behavior among youth. Annual Review of Sociology, 6, 235-285.
Kandel, D. B., Davies, M., Karus, D., & Yamaguchi, K. (1986). The consequences in young adulthood of adolescent drug involvement. Archives of General Psychiatry 43, 746-754.
Kaplan, H. B., Martin, S. S., & Robbins, C. (1982). Application of a general theory of deviant behavior: Self-derogation and adolescent drug use. Journal of Health and Social Behavior, 23, 274-294.
Kellam, S. G., Branch, J. D., Agrawal, K. C., & Ensminger, M. E. (1975). Mental health and going to school, The Woodlawn program of assessment, early intervention, and evaluation. Chicago: University of Chicago Press.
Kellam, S. G., Brown, C. H., Rubin, B. R., & Ensminger, M. E. (1983). Paths leading to teenage psychiatric symptoms and substance use: Developmental epidemiological studies in Woodlawn. In S. B. Guze, F. J. Earls, & J. E. Barrett (Eds.), Childhood psychopathology and development (pp. 17-47). New York: Raven.
Kellam, S. G., Ensminger, M. E., & Simon, M. B. (1980). Mental health in first grade and teenage drug, alcohol, and cigarette use. Drug and Alcohol Dependence, 5, 273-304.
Keyes, S., & Block, J. (1984). Prevalence and patterns of substance abuse among early adolescents. Journal of Youth and Adolescence, 13, 1-14.
Margulies, R. Z., Kessler, R. C., & Kandel, D. B. (1977). A longitudinal study of onset of drinking among high school students. Journal of Studies of Alcohol, 38, 897-912.
Miller, J. D., Cisin, 1. H., Gardner-Keaton, H., Harrel, A. V., Wirtz, P. W., Abelson, H. I., & Fishburne, P. M. (1983). National survey on drug abuse: Main findings 1982. Rockville, MD: National Institute on Drug Abuse.
Minuchin, S. (1974). Families and family therapy. Cambridge, MA: Harvard University Press.
Mischel, W., Shoda, Y., & Peake, P. K. (1988). The nature of adolescent competencies predicted by preschool delay of gratification. Journal of Personality and Social Psychology, 54, 687-696.
National Institute on Drug Abuse. (1986). Capsules: Overview of the 1985 household survey on drug abuse. Rockville, MD: Author.
Newcomb, M., & Bentler, P. (1988). Consequences of adolescent drug use: Impact on the lives of young adults. Newbury Park, CA: Sage.
Perry, C. L., Killen, J., & Slinkard, L. A. (1980). Peer teaching and smoking prevention among junior high students. Adolescence, 15, 277-281.
Robins, L. N. (1984). The natural history of adolescent drug use. American Journal of Public Health. 74, 656-657.
Scarr, S., & McCartney, K. (1983). How people make their own environments: A theory of genotype (arrow) environmental effects. Child Development, 54, 424-435.
Shapiro, D. (1965). Neurotic styles. New York: Basic Books.
Single, E., Kandel, D., & Johnson, B. D. (1975). The reliability and validity of drug use responses in a large scale longitudinal survey. Journal of Drug Issues, 5, 426-443.
Smith, G. M., & Fogg, C. P. (1978). Psychological predictors of early use, late use, and nonuse of marijuana among teenage students. In D. B. Kandel (Ed.), Longitudinal research on drug use (pp. 101 -113). New York: Wiley.
Stein, J. A., Newcomb, M. D., & Bentler, P. M. (1986). The relationship of gender, social conformity, and substance use: A longitudinal study. Bulletin of the Society of Psychologists in Addictive Behaviors, 5, 125- 138.
Tanaka, J. S., & Huba, G. J. (1984). Confirmatory hierarchical factor analysis of psychological distress measures. Journal of Personality and Social Psychology 46, 621-635.
Tobler, N. S. (1986). Meta-analysis of 143 adolescent drug prevention programs: Quantitative outcome results of program participants compared to a control or comparison group. Journal of Drug Issues, 16, 537-568.
Warner, W. L., Meeker, M., & Eells, K. (1949). Social class in America. Chicago: Science Research Associates.
Watson, D. W., & Clark, L. A. (1984). Negative affectivity: The disposition to experience aversive emotional states. Psychological Bulletin, 96, 465- 490.