Journal of Adolescence 1991, 14, 255-273

Adolescent drinking behaviour and family socialization factors: a meta-analysis DAVID

R. FOXCROFT

AND

GEOFF

LOWE”

Family socialization processes are important influences on behaviour in childhood and adolescence. Two major dimensions of family socialization are Support and Control, and these two dimensions were assessed for their influence on adolescent drinking behaviour. Thirty recently published research studies, which reported the influence of (clearly identifiable) family socialization variables on (selfreported) adolescent drinking behaviour were selected for analysis. The results of these studies were subjected to meta-analysis using a sorting technique. Variables were sorted along the dimensions of Support and Control, and along a Family Structure dimension, which measured parental intactness. Results of the meta-analysis indicated a clear negative linear relationship between Support and adolescent drinking. There was also a negative linear relationship between Control and drinking behaviour. Thus low support and lax control were associated with increased drinking. However, there was some evidence of a possible curvilinear relationship between control and adolescent drinking. A negative relationship between Family Structure and adolescent drinking was also found, i.e. adolescents from non-intact families tend to drink more. The results were incorporated into a family systems perspective. It is suggested that extremes of Support and Control, when measured adequately, may be dysfunctional for adequate socialization into normal drinking behaviour, as defined by social and cultural norms, during adolescence.

ADOLESCENT

DRINKING

The literature on young people drinking of the socialization process from child 1989; Barnes, 1977; Stacey and apparent in the adolescent years, *Reprint Hull

HU6

requests should be addressed

7RX,

describes such behaviour as part to adult (e.g. Sharp and Lowe,

Davies, 1970). when physical

to G. Lowe, Department

This behaviour and psychological

of Psychology,

University

is most devel-

of Hull,

U.K.

0140-1971/91/030255+19

$03.00/O

0 1991 The

Association

for the Psychiatric

Study

of Adolescents

256

D. R. FOXCROFT

AND G. LOWE

opment, and age-related status, mean that adolescents try to behave more like adults. Also, parents and other adults treat adolescents transitionally more like adults. Given that drinking alcohol is a widespread and normal part of adult life, then adolescent drinking will increase from abstention to “adult-levels” throughout the adolescent years. During this period the acquisition of appropriate drinking behaviour does not seem to be a problem for the majority of teenagers (Sharp and Lowe, 1989; Barnes, 1977; Stacey and Davies, 1970). In Great Britain most adolescents (c.,94 per cent) have had an alcoholic drink by the time they are 16 years old (Marsh et al., 1986; Fogelman, 1978). This is the legal age for consuming certain alcoholic beverages with a meal in a restaurant. (The legal age for the purchase of an alcoholic beverage from any legitimate source, i.e. licensed premises, in Great Britain is 18 years.) Many adolescents begin their drinking much earlier than 16 years of age. Drinking at home under parental supervision begins during childhood and the early teenage years, peaking at the age of 13 or 14. After this, adolescent drinking tends to switch to settings away from the home (see Sharp and Lowe, 1989, for a review). Under-age drinking away from home, by obtaining alcohol from licensed premises, may well be considered deviant with regard to the law in Great Britain. Hawker (1978) found that 80 per cent of boys and 75 per cent of girls had tried to purchase alcohol illegally from licensed premises. Nevertheless, within an individual’s own social and cultural environment, under-age drinking may be perceived as just a normal step in the development from adolescent to adult status, and therefore non-deviant. However, when this drinking becomes excessive then there is cause for concern. There are no directly comparable studies to examine whether adolescent drinking has increased in recent years, but there is some evidence to suggest that drinking problems are becoming increasingly prevalent in much younger age groups (Special Committee of the Royal College of Psychiatrists, 1986; Wallace et al., 1987). Marsh et al. (1986) report that half the 13-year-old girls in their probability sample had been slightly drunk at least once, and 17 per cent had been very drunk. For the 17-yearold boys these figures are 80 per cent and 50 per cent, respectively. As Sharp and Lowe (1989) conclude: “Drinking per se is not as worrying as the amount of drunkenness and consequent problems. Children may get drunk as part of learning how to drink sensibly. However, when their sole reason for drinking is to get drunk, then young people may be heading towards both social and physical problems” (p.305).

ADOLESCENT

DRINKING

257

BEHAVIOUR

Parental characteristics, attitudes to drinking, and to under-age drinking in their own offspring, are important factors. An individual’s parents and family can be considered to be the major source of socialization and influence during childhood and early adolescence (e.g. Rollins and Thomas, 1979), and, as stated above, are the predominant introducers of alcohol to offspring. How an individual perceives his or her family, in terms of these socialization processes, will affect the amount of influence parental attitudes and characteristics have on that individual. Sharp and Lowe (1989) reported that the majority of parents were perceived by their teenage offspring as not really concerned about them drinking either at home or out with friends: “..less than 10 per cent said their parents actually disapproved per cent said their parent approved” (p.301).

and only 20

Hawker (1978) found that only 14 per cent of parents actually disapproved of their offspring drinking alcohol while under-age. Parker (1974) also reported that adults were generally ambivalent to under-age drinking. This apparent ambivalence to under-age drinking by parents and adults is possibly interpreted by the adolescent as a condoning attitude. The family socialization process also influences susceptibility to advertizing pressures. Strickland and Pittman (1984) report that teenagers in families with a communication structure which fosters interpersonal harmony, agreement, and mutual esteem (Socio-communication), and fosters independence of thought, rational problem-solving, and responsibility for decisions (Concept-communication), were less likely to be influenced by advertisements for alcoholic drinks. We now look at socialization processes within the family environment, and how they relate to adolescent drinking behaviour.

CHILDHOOD

AND THE

ADOLESCENT SOCIALIZATION

FAMILY

BEHAVIOUR

-

PROCESS

As stated above, parents and the family are the primary agent of socialization for children and early adolescents. As an individual moves towards mid-adolescence, late adolescence and adulthood then peers and social contacts outside the family become increasingly important as frames of reference for behaviour. However, the influence of the family is still predominant. Rosenberg (1979) found that parents ranked higher than peers in interpersonal significance throughout adolescence. Also, satisfaction with support from parents (especially mothers) was a better indicator of

258

D. R. FOXCROFT

AND G. LOWE

adolescent well-being than satisfaction with help from peers, in a study by Burke and Weir (1978). Greenberg et al. (1983) found that age did not appear to be a significant factor in relative parent/peer relationships. Older adolescents were no different than younger adolescents in their report of “quality” or “utilization” of relationships with parents or with peers. Noller and Callan (1991) remark:“Although peers become more important for adolescents, and they spend a lot of time talking with peers, there is little evidence that the peer group actually becomes more important than the family during adolescence” (P.51).

It seems that throughout the school years parents are highly valued, usually more so than peers, for their support, love and advice. The dimensions of family environment that appear most salient to the socialization process are two dimensions that family researchers, reviewers, and theorists have consistently identified. These two dimensions have been given various names. Argyle and Henderson (1985) state that “the most important dimension along which parents vary is that of acceptancerejection” (p.209). Accepting parents are warm, loving and rewarding. These authors also identify another major dimension, permissivenessstrictness, which indicates how parents control their children. Permissive parents do not restrict their children’s behaviour, and the children are often free to do as they please, with minimal guidance. The more usual labels from the literature, and the ones preferred here, are Family Support for one dimension (acceptance-rejection), and Family Control for the other (permissiveness-strictness). Family Support comprises behaviours which foster in an individual feelings of comfort and belonging, and that he or she is basically accepted and approved of as a person by the parents and family. Family Control comprises family interactions which direct the behaviour of an individual in a manner desirable to the parents, or to the power base in the family. Rollins and Thomas (1979) conducted a major review of the child socialization literature and identified 18 labels from 106 published studies and 129 unpublished dissertations, which could be subsumed under the dimension Control Attempts. They also identified nine labels, which could be subsumed under a Support dimension (see Table 1.) Rollins and Thomas’ (1979) categorization of variables into the dimensions of Control Attempts and Support (Table 1) is necessarily (i) subjective, and (ii) according to the labels and definitions used in the original studies. There may be some overlap for certain variables within the two dimensions. However, this categorization provides a workable model for

ADOLESCENT

Table 1.

DRINKING

BEHAVIOCR

259

Labels frequently used for two dimensions of parentat behaviour”

Control

attempts

Authoritative Authoritarian Autonomy Coercion ControlDemanding Democratic Discipline Dominance Induction Permissive Power Power Assertion Pressure Protective Punishment Restrictive Strictness

Support Acceptance Affection Hostile Love Neglect Nurturance Rejection Support Warmth

(5)T (35) (6) ,. ,::; (14) (15) (23) (25) (8) (8) (8)

(20) (15) (29) (15) (5) (37) (36) (11) (31)

(7) (5) (16) (47) (24) (7)

“Adapted from Rollins and Thomas (1979). tAl1 labels used in five or more studies are listed by frequency (in parentheses).

reducing family and parental behaviour variables into these two major socialization dimensions. There are cultural differences in parent-child relations. Devereux (1970) found that English families, as opposed to German and American families, used more physical punishment, were more nagging and scolding, and more indulgent. English families were also less nurturant, less protective, made fewer achievement demands, used less discipline by explanation and less punishment by love withdrawal, they were lower on close surveillance and on giving jobs around the house. It therefore seems that on the whole, for the dimensions of Support and Control, the cultural norms for English families are lower support and looser control than the norms for American and German families.

FAMILY

ENVIRONMENT

AND

ADOLESCENT

DRINKING

Kandel (1980) outlines the important contribution socialization research and theory have to make to drug (including alcohol) research. Barnes et al. (1986) firmly advocate the distinction of the dimensions of Support and

260

D. R. FOXCROFT

AND G. LOWE

Control when looking at family environment and adolescent drinking behaviour. Following this lead, we attempted to identify all family behaviour variables investigated in previous adolescent drinking research, and subsumed these variable into either a Family Support or a Family Control dimension. Although this is a subjective categorization variables were sorted along the lines of Rollins and Thomas’ work (see Table 1).

Method For the purpose of this paper articles were obtained from several sources: (1) The adolescent drinking behaviour library, Alcohol Research Laboratory, Hull University. (2) Keyword search of Psychological Abstracts CD-ROM. (3) Online keyword search of two U.S. Databases PSYCINFO and SOCIOLOGICAL ABSTRACTS. (4) Retrospective search of recent editions (past two years) of Current Contents - a weekly publication listing titles of articles in current journals in the social and behavioural sciences. (5) Following up all relevant references from available sources. A total of 30 published articles were selected for inclusion in the present review. The criteria for inclusion were:- (1) The article was published in a reputable academic journal, in the past 20 years. (2) The family socialization variables could be clearly classified along the appropriate dimensions. (3) The drinking behaviour variable was a self-report measure and was easily identifiable, either on its own, or less frequently, as part of a composite substance use measure. (4) The subjects’ age range could be classified as adolescent or teenage, (5) Relationships between family socialization variables were as reported. No re-analysis of data was carried out. (6) Only direct relationships were classified as significant. Those studies which showed an indirect relationship between drinking behaviour and support or control were classified as non-significant. It soon became apparent that three factors recurred throughout the literature. As expected, variables which could be subsumed under the dimensions of Family Support and Family Control were frequently reported as an important correlate of adolescent drinking behaviour. Family Structure, i.e. the extent of parental intactness, was also quite often reported as an important correlate of adolescent drinking behaviour. The family environment variables from each study, and appropriate subsumption are shown in Table 2. It can be seen from Table 2 that research into adolescent drinking and family environment consistently provides results which identify Family Structure, Family Support and Family Control as salient and important dimensions. Whether these dimensions are the most salient is open to

ADOLESCENT

question,

as research

expense

of others.

Growth

dimension

DRINKING

to date may have concentrated

Moos

and Moos

is also

an

261

BEHAVIOCTR

(1986)

on these

have suggested

important

factor

areas at the

that a Personal

in family

environment.

Personal growth comprises variables such as independence, achievement motivation, active-recreational orientation, intellectual-cultural orientaArgyle and Henderson (198.5) also tion, and moral-religious emphasis. mention

other

dimensions,

and Family

Support

but

accord

them

less important

than

Family

Control:-

“Parental behaviour varies along other dimensions.... for example in stimulating different degrees of creativity or achievement motivation. However no other dimensions affect the relationships with children so profoundly...” (p.211).

META-ANALYSIS

The

three

dimensions

Family

Support

sorting

method

appropriate tively

was used,

dimension

related,

are shown

in Table effects

from

Control,

the were

literature,

respect

related,

2. Table

results

to drinking

for Family

being

as either

These

A

on the posi-

classifications

of the meta-analysis.

the frequencies

from df=2,

(~~~10.75, and Family

Table

3 revealed

p12

Year 1: 12-15 Year 4: 16-19

Ninth graders

13-18

Ever had an alcoholic drink

12-14

structure

support

(-)

Q/F Index

Transition to drinker status

Parental Control (ns)

Parental (-) Willing, Involvement Strictness (-)

Adaptability

(-)

control

Parent-Youth Power (-)

Family

control*-continued

(-)

(-) (-) (-)

Support

Family Attachment Involvement Commitment

Parental

Trust and Concern (-)

Cohesion

Parent-Youth Sentiment (-)

Closeness to Family (-)

Family

support and family

Frequency of use in past six months

Parental (-) Intactness

Family (-) Structure

Family

family

Parental Support and Affection (n.s.)

-

structure,

Frequency of use in past year

Drug abusers (including alcohol)

Alcohol and other beverages (frequency only)

9-17

behaviour

Frequency ever used

Drinking

11-17

Age

The results qf studies classified into the dimensions of family

Coombs and Landsverk (1988)

Byram and Fly (1984)

Study

Table 2.

Prendergast and Schaefer (1974)

Plant ef al. (1985)

Pandina and Scheule (1983)

Needle et al. (1990)

Mercer et al. (1978)

Mercer and Kohn (1980)

(57)

H.S.students

(929)

Secondary students

H.S.students (1960)

(467)

Adolescents

(286)

H.S.students

(500)

H.S.students

H.S.students (1936)

Margulies (1977)

et al.

H.S.students (2626)

(499)

Marcos et al. (1986)

H.S.students

(N]

(1987)

Subjects

Frequency of use in past six months

Grade 9

Scale of SUI-substance use involvement Previous weeks alcohol consumption Frequency of drinking

12-18

15-16

16-19

Multiple substance use scale

Frequency of alcohol use

Grades ll->13

Onset of “hard liquor” use

Lifetime alcohol use

1+19

behaviour

Q/F index and P.D. status

Drinking

Grades 10 ->12

Age

family

Who raised child (-)

Divorced or intact parents (-)

structure

structure,

Family

The results of studies classaj5ed into the dimensions of family

Kline et a/.

Study

Table 2.

(-)

Love (n.s.)

Parental (-) AcceptanceRejection

Overall Parental Love (-)

Warmth, Support and Interest (n.s.)

Parental

Closeness with Family (ns.)

Parental Attachment

control

Parental Control (-)

(-)

(n.s.)

Overall Parental Autonomy

Organization

Parental Control (n.s.)

Parental Control (n.s.)

Family

control *-continued

and (-)

support

Family Conflict Disengagement

Family

support and family

et al.

2.

(N)

H.S.students (1923)

Grades 6->12

Grades 7->11

13-19

12-18

18-22

Age

behaviour

Frequency of heavy and light drinking

Initial use, Q/F index P.D. status

Frequency of drinking

Frequency of alcohol use

Control of drinking

Drinking structure

Parental (-) Intactness

(-)

family

Father Absence

Family

structure, support

Closeness to Family (-)

Parent-Adolescent Conflict (-)

Parental (-) Support

Family

support and family

control

Parental Control (-)

Family

control *-continued

* In order to establish directional continuity when comparing studies, each finding was classified according to its relationship with the main (ns.) related to the drinking behaviour variable. For dimension, as either significantly positively (+), significantly negatively (-), or non-significantly example, the study by Budd et al. (1985) found family conflict (a Support variable) to co-vary positively with adolescent drinking, but as family conflict is negatively related to Family Support, then the Budd et al. finding provides evidence that Family Support is negatively related to adolescent drinking behaviour, i.e. (-),

Wechsler and Thum (1973)

H.S.students

Thompson and Wilsnack (1987)

(839)

Native U.S. teenagers (74)

(813)

Adolescents

(494)

H.S.students

Subjects

The results qf studies classified into the dimensions qf family

Takei et al. (1988)

Stern et al. (1984)

(1987)

Schlegel

Study

Table

266

D. R. FOXCROFT AND G. LOWE

Table 3.

Total

Dimension Family Family Family

significant

and non-significant

results

for each dimension

Significantly positive (+)

Significantly negative (-)

0 0 1

7 23 9

Structure Support Control

CURVILINEAR

Non-significant n.s. 1 5 6

RELATIONSHIPS

Barnes et al. (1986) and Glynn (1981) report a clear curvilinear relationship between Family Control and adolescent drinking. High and low control are both associated with increased drinking in youth. This pattern is (Coombs and also apparent in other results, though not reported Landsverk, 1988). In Table 3 there were six non-significant results and one significantly positive report of the relationship between Family Control and adolescent drinking, compared to nine significantly negative results. Although this produced a significant effect in the Chi-squared analysis, this effect is not as clear cut as in the Family Structure and Family Support dimensions. It may be that the relationship between adolescent drinking behaviour and Family Control is not a linear one, thus confounding the results from previous studies. Studies which report a linear relationship between Family Control and adolescent drinking, although in the majority, may suffer from a problem with “thin” variable ranges. That is to say, the range of behaviour which a variable assesses is not sufficiently wide to allow a true picture to be obtained. For example, if Family Control does indeed have a curvilinear quadratic association with adolescent drinking, but a particular variable taps only the downward slope, or only the upward slope, then a linear trend could emerge. Many of the present studies may ask questions which consider only part of the range, and it is possible that a false picture may build up of a linear relationship between Family Control and adolescent drinking. To clarify this possibility, and to confirm the results of the present meta-analysis, future studies need to clearly identify and assess equally all parts of the Control dimension.

ADOLESCENT A FAMILY

DRINKING SYSTEMS

267

BEHAVIOUR

PERSPECTIVE

Minuchin (1974) proposed Family Cohesiveness to be the major dimension of family functioning. This theory has formed the basis of much subsequent family systems work. There is a considerable overlap between the concepts of Cohesion and Support (Bloom, 1985). These concepts have sometimes been used interchangeably, and it is apparent why when one considers the definition of Cohesion given by a leading systems theorist: “the emotional (Olson

bonding

that family

members

have toward

one another”

et al., 1983, p.70).

Minuchin’s theory specifies that extremes of Family Cohesionenmeshed and disengaged families-can be dysfunctional, with moderate cohesiveness optimal for family functioning. This is not the picture obtained from the present review of adolescent drinking behaviour, where Support is linearly related to drinking behaviour. However, other family systems theorists have found that Cohesion is essentially a linear function. Higher cohesion is indicative of better family functioning, and lower cohesion of poorer family functioning (Anderson and Gavazzi, 1990; Fristad, 1989; Lee, 1988). Cohesion and Support are similar concepts, but as outlined for Family Control in the previous section, there may be a measurement deficiency which explains the preponderance of findings of a linear relationship between family functioning and Family Support. Also, family systems theorists usually work with dysfunctional families, and it may be only in problem families that high support is considered dysfunctional (Olson et al., 1983). In studies of families with a schizophrenic member, mothers and fathers have sometimes been found to be over-protective, over-involved, and over-nurturant, i.e. enmeshed (e.g. Laing and Esterson, 1964). However, Friedman et al. (1987) f ound that drug abusing adolescents were almost all from disengaged families. Therefore it seems that the type of dysfunction may be important for the consequential behaviour (Beaver and Voeller, 1983). A linear relationship may well exist between Family Cohesion and adolescent behaviour for those individuals who use and abuse drugs and alcohol, but this linear relationship may not hold for other types of consequent behaviour. The method of assessment of family behaviour may also be important. Friedman et al. (1987) report that therapists’ perceptions of families differ substantially from a family’s own selfperceptions. Systems therapists are more likely to classify extremes of cohesion as dysfunctional, whereas families are more likely to perceive only low cohesion as dysfunctional.

268

D. R. FOXCROFT

AND G. LOWE

Minuchin (1974) and other systems theorists (Olson et al., 1979) outlined Adaptability as a second major factor influencing family functioning. Olson and co-workers view Adaptability as a curvilinear dimension, with the extremes dysfunctional. It is possible that Control Attempts and Adaptability can be viewed as similar concepts, with substantial overlap (Bloom, 1985), ,but this depends on the definition given to each. Olson et al. (1983) define Adaptability as: “the ability of a marital or family role relationships, and relationship developmental stress” (p.70).

system rules

to change its power structure, in response to situational and

Olson and his colleagues go on to state that concepts mainly from family sociology make up this dimension. Such concepts are family power (assertiveness, control, discipline), negotiation styles, role relationships and relationship rules. These concepts are very similar to those outlined in the present review as contributing to a Family Control dimension. Strict and lax control attempts might respectively equate with Olson’s rigid and chaotic adaptability. A curvilinear property of the Control dimension finds only limited support in Rollins and Thomas’ (1979) review, and in the present paper. A problem with the definition of Adaptability in Olson’s Circumplex Model is that it has been criticized as inferring a linear relationship between adaptability and family functioning, i.e. families more able to change are optimal. This has led to consequent confusion in the items of the test battery (FACES I to FACES III) designed by Olson and colleagues to measure this dimension (Lee, 1988; Anderson and Gavazzi, 1990).

DISCUSSION

Although Family Structure was extracted from the literature as a separate dimension, the absence of a parent may have profound effects on the amount of Support and Control provided within such a family. Most studies have found that children and adolescents from divorced families exhibit emotional distress and behaviour disorders, although this can depend on the recency of the parental separation. When quality of the parent-child relationship is controlled, then the effect of family structure is greatly reduced, but may still be significant (Flewelling and Bauman, 1990; Needle et al., 1990). Individuals from families in which the socialization process has provided a good psychological adjustment in the individual are generally more

ADOLESCENT

DRINKING

BEHAVIOUR

269

confident and autonomous, have better social skills, and are more likely to pass on these qualities to their own offspring, than individuals from families who have provided poorer socialization. Family Support and Family Control are two major dimensions of the socialization process, and are two of the most important factors in familial influence on adolescent drinking behaviour, as identified by this review. Therefore individuals from families deficient on these two dimensions are more likely to have less confidence, autonomy, and poorer social skills. Ford (1982) reported that social cognition variables (including social support networks) accounted for a large proportion of the variance in social competence. This accords with recent work by Bagnall (1990), who evaluated an alcohol education initiative, and reports that the way forward in alcohol education lies in an approach which emphasizes social influences and social skills. Adolescent drinking should be regarded as a normal developmental process, providing the adolescents social and cultural environment condones such behaviour. Thus, effective socialization should lead to controlled and sensible drinking behaviour, where appropriate. If a family is deficient in providing support and control, deviant or excessive drinking behaviour may result. In the preceding sections we have found that low support and lax control are associated with heavier drinking in adolescents. But, if an individual remains abstinent in an environment which condones and encourages drinking, then this too is deviant drinking behaviour, and it could be that high support and strict control are associated with abstention. Moderate amounts of Family Support and Family Control would therefore be the most functional for the socialization and development of sensible drinking in an individual. This hypothesis is also amenable to the family systems viewpoint, where extremes of Support and Control, or Cohesion and Adaptability, are viewed as dysfunctional. This is an important step, because it clarifies the family systems viewpoint on non-problem families. Research has shown that generally a linear relationship exists between these two dimensions and a target behaviour for non-problem families, and although these findings may be legitimate, it is the range of normality of the target variable which is important, For example, if the target variable is antisocial behaviour, then a linear relationship may indeed exist between the amount of antisocial behaviour and the two socialization dimensions, with low support and lax control associated with higher levels of antisocial behaviour. But, closer examination would reveal high support and high control to be associated with poorly autonomous, very socially conforming behaviour. This image is one with which readers of Orwell’s “Nineteen Eighty Four” (1954) will be familiar. In the present climate autonomous individuals with good social skills and independence of thought are viewed as more able and attractive.

270

D. R. FOXCROFT

AND G. LOWE

Moderate amounts of Family Support and Family Control would therefore socialize an individual towards this ideal. In this review we have also identified heavy drinkers who come from strictly controlling family environments. How does this fit into the above picture? As outlined above, strict control may be dysfunctional, and individuals from this type of family environment are more likely to abstain, contrary to social and cultural norms. However, there may be an interaction between Family Control and Family Support at the extremes of these dimensions. If there is optimal support for an individual, then they are more likely to be abstainers if they are from a strictly controlled environment. On the other hand, if there is strict control, but dysfunctional support, then individuals may become heavier drinkers. In the U.S.A. the level of under-age drinking has generally been found to be much less than in Great Britain. Only about 80 per cent of American adolescent aged 16 or over were reported to have consumed an alcoholic beverage (Rachel et al., 1980, cited by Plant et al., 1985). In Great Britain the level is nearer 9.5 per cent. Bearing this in mind, it is interesting to note the cultural variation in parent-child relations between England and the U.S.A. mentioned earlier (Devereaux, 1970) when lower support and looser control were found in English families. Intuitively, this accords with the results of the meta-analysis, in which the majority of studies came from the U.S.A. That is, higher support and firmer control were found to be associated with lower drinking levels. Given the established cultural variation in both adolescent drinking behaviour (Rachel et al., 1980; Bank et al., 1985) and parent-child relations (Devereaux, 1970), then it would be folly to directly compare results from different countries. As the majority (all but two) of the articles in the present review came from the U.S.A., research is needed in other countries to discover the pattern and impact of family factors on adolescent alcohol use in a particular country. Even within countries there may be regional variations (Marsh et al., 1986; Fogelman, 1978), and it would be wise to take measurements from different regions within different countries.

ACKNOWLEDGEMENT

We are grateful to the Alcohol viding support for D.R.F.

Education

and Research

Council

for pro-

ADOLESCENT

DRINKING

BEHAVIOUR

271

REFERENCES Adler, P. and Lotecka, L. (1973). Drug use among high school students: patterns and correlates. International Journal of the Addictions, $537-548. Anderson, S. and Gavazzi, S. (1990). A test of the Olson Circumplex Model: examining its curvilinear assumption and the presence of extreme types. Family Process, 29, 309-324. Argyle, M. and Henderson, M. (1990). The Anatomy of Relationships. London: Penguin. Baer, P., Garmezy, L., Mclaughlin, R., Pokorny, A. and Wernick, M. (1987). Stress, family conflict and adolescent alcohol use. Journal of Behavioural coping, Medicine, 10,449-466. Bagnall, G. (1990). Alcohol education for 13 year olds - does it work? Results from a controlled evaluation. British Journal of Addiction, 85,89-96. Bank, B., Biddle, B., Anderson, D., Hauge, R., Keats, D., Marlin, S. and Valantin, S. (198.5). Comparative research on social determinants of adolescent drinking. Social Psychology Quarterly, 48, 164-l 77. Barnes, G. (1977). The development of adolescent drinking behaviour: an evaluative review of the impact of the socialization process within the family. Adolescence, 12,571-591. Barnes, G., Farrel, M. and Cairns, A. (1986) Parental socialization factors and adolescent drinking behaviours. Journal of Marriage and the Family, 48,27-36. Barnes, G. and Windle, M. (1987). Family Factors in Adolescent Alcohol and Drug Abuse. Pediatrican, 14, 13-l 8. Beavers, W. and Voeller, M. (1983). Family Models: Comparing and contrasting the Olson Circumplex Model with the Beavers System Models. Family Process, 22, 85-98. Bloom, B. (1985). A factor analysis of self-report measures of family functioning. Family Process, 24, 225-239. Brook, J., Nomura, C. and Cohen, P. (1989). A network of influences on adolescent drug involvement: neighbourhood, school, peer, and family. Genetic, Social, and General Psychology Monographs, 115,123-145. Brook, J., Whiteman, M., Gordon, A., Nomura, C. and Brook, D. (1986). Onset of adolescent drinking: A longitudinal study of intrapersonal and interpersonal antecedents. Advances in Alcohol and Substance Abuse, 5,91-l 10. Budd, R., Eiser, J., Morgan, M. and Gammage, P. (1985). The personal characterisics and life style of the young drinker: the results of a survey of British adolescents. Drug and Alcohol Dependence, 16, 145-157. Burke, R. and Weir, T. (1978). Benefits to adolescents of informal helping relationships with parents and peers. Psychological Reports, 42, 1175-l 184. Burnside, M., Baer, P., McLauglin, R. and Pokorny, A. (1986). Alcohol use by adolescents in disrupted families. Alcoholism: Clinical and Experimental Research, 10, 274-278. Byram, 0. and Fly, J. (1984). Family structure, race, and adolescents’ alcohol use: a research note. American Journal of Drug and Alcohol Abuse, 10,467-478. Coombs, R. and Landsverk, J. (1988). Parenting styles and substance use during childhood and adolescence. Journal of Marriage and the Family, 50,473-482. Devereuz, E. (1970). Socialization in cross-cultural perspective: comparative study of England, Germany and the United States. In Families in East and West, Hill, R. and Konig, H. (Eds). Paris and the Hague: Mouton. F amily structure Flewelling, R. and Bauman, K. (1990). as a predictor of initial

272

D. R. FOXCROFT

AND G. LOWE

substance use and sexual intercourse in early adolescence. Journal of Marriage and the Family, 52, 171-181. Fogelman, K. (1978). Drinking among sixteen-years-olds. Journal of National Childrens Bureau London, 29, 19-25. Ford, M. (1982). Social cognition and social competence in adolescence. Developmental Psyci2ology, 18, 323-340. Friedman, A., Utada, A. and Morrissey, M. (1987). Families of adolescent drug abusers are “rigid”: are these families either “disengaged” or “enmeshed”, or both? Family Process, 26, 131-148. Fristad, M. (1989). A comparison of the McMaster and Circumplex family assessment instruments. Journal of Marital and Family Therapy, 15,259-269. Glass, G., McGraw, B., Smith, B. and Lee, M. (1981). Meta-analysis in Social Research. Beverley Hills (Calif.): Sage. Glynn, T. (1981). From family to peers: a review of transitions of influence among drug using youth. Journal of Youth and Adolescence, 10,363-383. Greenberg, M., Siegal, J. and Leitch, C. (1983). The nature and importance of attachment relationships to parents and to peers during adolescence. Journal of Youth and Adolescence, 12, 373-383. Hawker, A. (1978). Adolescents and Alcohol. London: Edsall. Hays, R., Stacy, A., Widaman, K., Dimatteo, M. and Douney, R. (1985). Multistage path models of adolescent alcohol and drug use: a re-analysis. Journal of Drug Issues, 16, 357-369. Hundleby, J. and Mercer, G. (1987). Family and friends as social environments and their relationship to young adolescents’ use of alcohol, tobacco, and marijuana. Journal of Marriage and the Family, 49, 1.51-164. Jessor, R. and Jessor, S. (1975). Adolescent development and the onset of drinking: a longitudinal study. Journal of Studies on Alcohol, 36, 27-51. Johnson, K. (1986). Informal control networks and adolescent orientation toward alcohol use. Adolescence, 21, 755-784. Kandel, D. (1980). Drug and drinking behavior among youth. Annual Review Sociology, 6,235-285. Kline, R., Canter, W. and Robin, A. (1987). Parameters of teenage alcohol use: a path analytic conceptual model. Journal of Consulting and Clinical Psychology, 55, 521-528. Laing, R. and Esterson, A. (1964). Sanity, Madness and the Family. London: Tavistock. Lee, C. (1988). Theories of family adaptability: toward a synthesis of Olson’s Circumplex and the Beavers Systems Models. Family Process, 27,73-85. Marcos, A., Bahr, S. and Johnson, R. (1986). Test of a bonding/association theory of adolescent drug use. Social Forces, 65, 13 5-l 61. Margulies, R., Kessler, R. and Kandel, D. (1977). A longitudinal study of onset of drinking among high school students. Journal of Studies on Alcohol, 38,897-912. Marsh, A., Dobbs, J, and White, A. (1986). Adolescent Drinking. London: HMSO. Mercer, G., Hundleby, J. and Carpenter, R. (1978). Adolescent drug use and attitudes toward the family. Canadian Journal of Behavioural Science, 10,79-90. Mercer, G. and Kohn, P. (1980). Child-rearing factors, authoritarianism, drug use attitudes, and adolescent drug use: a model. Journal of Genetic Psychology, 136, 159-171. Minuchin, S. (1974). Families and Family Therapy. New York: Tavistock. Moos, R. and Moos, B. (1986). Family Environment Scale Test Manual. Consulting Palo Alto (Calif.): Psychologists Press.

ADOLESCENT

DRINKING

BEHAVIOUR

273

Needle, R., Su, S. and Doherty, W. (1990). Divorce, remarriage, and adolescent substance use: a prospective longitudinal study. Journal of Marriage and the Family, 52,157-l 69. Noller, P. and Callan, V. (1991). The Adolescent in the Family. London: Routledge. Olson, D., Russell, C. and Sprenkle, D. (1983). Circumplex model of marital and family systems: VI. Theoretical update. Family Process, 22, 69-83. Olson, D., Sprenkel, D. and Russell, C. (1979). Circumplex Model of marital and family systems: I. Cohesion and adaptability dimensions, family types, and clinical applications. Family Process, 18, 3-28. Orwell, G. (1954). “-Nineteen Eighty Four”. London: Penguin. Pandina. R. and Scheule, J. (1983). Psychosocial correlates of alcohol and drug use and adolescent students and adolescents in treatment. Journal of Studies on Alcohol, 44,950-973. Parker, H. (1974). Viezu from the Boys. London: Tavistock. Plant, M., Peck, D. and Samuel, E. (1985). Alcohol, Drugs and School Leavers. London: Tavistock. Prendergast, T. and Schaefer, E. (1974). Correlates of drinking and drunkenness among high school students. Quarterly Journal of Studies on Alcohol, 35, 232-242. Rachel, J., Maisto, S., Guess, L,. and Hubbard, R. (1980). Use and misuse of alcohol by United States adolescents-recent information from two national surveys. Paper presented at the 26th International Institute on the Prevention and Treatment of Alcoholism, ICAA, Cardiff. Rollins, B. and Thomas, D. (1979). Parental support, power, and control techniques in the socialization of children. In Contemporary Theories about the Family (Vol. 11, Burr, W., Hill, R., Nye, I. and Reiss, I (Eds). New York: Free Press, pp. 317-364. Rosenberg, M. (1979). Conceiving the Self. New York: Basic Books. Schlegel, R., d’Avernes, J., Zanna, M., DiTecco, D. and Manusue, S. (1987). Predicting alcohol use in young adult males: a comparison of the Fishbein-Ajzen Model and Jessor’s Problem Behaviour Theory. Drugs and Society, 1,7-24. Sharp, D. and Lowe, G. (1989). Adolescents and alcohol-A review of the recent British research. Journal of Adolescence, 12,295-307. Special Committee of the Royal College of Psychiatrists (1986). Alcohol: Our Favourite Drug. London: Tavistock. Stacey, B. and Davies, J. (1970). Drinking behaviour in childhood and adolescence: an evaluative review. British Journal of Addiction, 65, 203-212. Stern, M., Northman, J. and Van Slyck, M. (1984). Father absence and adolescent “problem behaviours”: alcohol consumption, drug use and sexual activity. Adolescence, 19,301-312. Stickland, D. and Pittman, D. (1984). Social learning and teenage alcohol use: interpersonal and observational influences within the socio-cultural environment. Journal of Drug Issues, 14, 137-l 50. Takei, Y., Lynch, P. and Charleston, G. (1988). To drink or not to drink: the Indian adolescents choice between friends and family. Journal of American Indian education, 27, l-9. Thompson, K. and Wilsnack, R. (1987). Parental influence on adolescent drinking: modelling, attitudes, or conflict. Youth and Society, 19, 22-43. Wallace, P., Brennan, P. and Haines, A. (1987). Drinking patterns in general practice patients. Journal of the Royal College of General Practitioners, 37, 354-357. Wechsler, H. and Thum, D. (1973). Teenage drinking, drug use, and social correlates. Quarterly Journal of Studies on Alcohol, 34, 1220-l 227.