Вы находитесь на странице: 1из 13

Addiction Research & Theory

ISSN: 1606-6359 (Print) 1476-7392 (Online) Journal homepage: http://www.tandfonline.com/loi/iart20

Adolescents’ anticipated social outcomes for


drinking alcohol and being drunk

Catherine Angela Quinn & Kay Bussey

To cite this article: Catherine Angela Quinn & Kay Bussey (2015) Adolescents’ anticipated social
outcomes for drinking alcohol and being drunk, Addiction Research & Theory, 23:3, 253-264,
DOI: 10.3109/16066359.2015.1014346

To link to this article: http://dx.doi.org/10.3109/16066359.2015.1014346

Published online: 25 Feb 2015.

Submit your article to this journal

Article views: 89

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at


http://www.tandfonline.com/action/journalInformation?journalCode=iart20

Download by: [Universitat de les Illes Balears] Date: 10 February 2016, At: 09:37
http://informahealthcare.com/art
ISSN: 1606-6359 (print), 1476-7392 (electronic)

Addict Res Theory, 2015; 23(3): 253–264


! 2015 Informa UK Ltd. DOI: 10.3109/16066359.2015.1014346

Adolescents’ anticipated social outcomes for drinking alcohol and


being drunk
Catherine Angela Quinn and Kay Bussey

Department of Psychology, Macquarie University, North Ryde, Australia

Abstract Keywords
The aim of this study was to investigate the social outcomes adolescents anticipate from three Alcohol-related harm, anticipated social
Downloaded by [Universitat de les Illes Balears] at 09:37 10 February 2016

social figures (mother, father and peers) for two alcohol-related behaviours (drinking alcohol outcome, parent, peers, underage drinking
and being drunk) and how these anticipated social outcomes relate to adolescents’
engagement in underage drinking and experience of alcohol-related harm. The sample was History
comprised of 651 (329 female) adolescents (age range: 12–16 years; 81% White). Results
revealed that the anticipation of less social censure, from mother and peers, for drinking Received 30 January 2014
alcohol, related to greater engagement in underage drinking. Further, when underage drinkers Revised 23 December 2014
were examined separately, lower levels of censure anticipated from mothers and peers for Accepted 29 January 2015
being drunk were associated with higher levels of alcohol-related harm. These findings Published online 25 February 2015
highlight the importance of not considering social outcomes as a monolithic process. Instead,
the results underscore the complexity of adolescents’ social environment and the need for
research and interventions to examine this context in terms of the different social figures that
influence adolescents and the different alcohol-related behaviours adolescents may engage in.
Further, it highlights the notable influence of mothers and peers on adolescents’ drinking self-
regulation, and their potential roles in intervention programs aimed at reducing underage
drinking and its associated harms.

Introduction & Keen, 2003). Moreover, many, although not all, studies
find that alcohol use during adolescence predicts long-term
Alcohol misuse can result in severe health, social and
alcohol dependence and abuse (Guttmannova et al., 2011;
financial burdens to the individual, families, communities
Maimaris & McCambridge, 2013; Wells, Horwood, &
and wider society (Institute for Health Metrics and
Fergusso, 2004).
Evaluation, 2013; Laslett et al., 2010; Lim et al., 2012). For
To reduce adolescent drinking and its associated harms, it
adolescents, in particular, there are concerns that the neuro-
is therefore important for interventions to target key factors
toxic effects of alcohol may adversely affect the maturing
associated with adolescents’ engagement in underage drink-
brain (Bava & Tapert, 2010; Clark, Thatcher, & Tapert, 2008;
ing and experience of alcohol-related harm. As alcohol use
Crews, He, & Hodge, 2007; Crews, Mdzinarishvili, Kim, He,
predominately occurs in a social context, social influences on
& Nixon, 2006). In the short-term, many adolescents who
adolescent alcohol use have been identified as important
consume alcohol experience harms as a result of their
factors to address in interventions (Foxcroft & Tsertsvadze,
drinking including physical illness and injury, engagement
2011; Teesson, Newton, & Barrett, 2012). In particular, social
in high-risk sexual activity, academic underachievement,
evaluations, or the approval or censure adolescents anticipate
violence and interpersonal problems (Bradley & Greene,
for using alcohol, have been strongly associated with adoles-
2013; Chikritzhs, Pascal, & Jones, 2004; Fergusson &
cent drinking (Ford & Hill, 2012; Kristjansson, Sigfusdottir,
Lynskey, 1996; Mason et al., 2010; Stafström, 2007). These
James, Allegrante, & Helgason, 2010; Mrug & McCay, 2013;
consequences can result in significant long-term difficulties
Smith & Rosenthal, 1995). Yet, there is limited understanding
such as disability, teenage pregnancy, sexually transmitted
of how social evaluations differ for different types of alcohol-
disease and incarceration (Cauffman et al., 2007; Cavazos-
related behaviours and for different social figures. Further,
Rehg et al., 2011; Jernigan, 2001; Kelley, Borwaski, Flocke,
there is limited understanding of how these contextual factors
may differentially impact adolescents’ engagement in under-
age drinking or experience of alcohol-related harm.
Consequently, to address this gap, this study extended
previous research by examining the social evaluations mid-
Correspondence: Miss Catherine Angela Quinn, Centre for Youth
Substance Abuse Research, Queensland University of Technology, adolescents anticipate for two types of alcohol-related
Brisbane QLD 4000, Australia. E-mail: c6.quinn@qut.edu.au behaviours (drinking alcohol and being drunk) and from
254 C. A. Quinn & K. Bussey Addict Res Theory, 2015; 23(3): 253–264

each of three social figures (mothers, fathers and peers) experiencing the outcomes themselves (enactive experience),
from the perspective of Bandura’s (1986) social cognitive and by being informed of what the social standards are and
theory (SCT). what outcomes will ensue if a specific behaviour is performed
There is limited understanding of the social evaluations (direct tuition). By extracting, weighting and synthesising the
adolescents anticipate for different alcohol-related behav- information gained from these different modes of influence
iours, such as drinking alcohol and drinking to intoxication. adolescents then develop their own anticipatory outcomes for
When investigating social evaluations for underage drinking, different behaviour, different contexts, and different social
most studies have focused on the social sanctions adolescents figures (Bandura, 1986). SCT posits that it is through the
anticipate for consuming one or two drinks of alcohol (Ford & application of forethought, that these foreseeable social
Hill, 2012; Mrug & McCay, 2013; Yu, 1998). It is intoxicated outcomes become current motivators of behavioural perform-
drinking, however, which is commonly associated with ance, with individuals encouraged to engage in or abstain
problematic alcohol use and experience of alcohol-related from particular behaviours based on the social approval or
harm (Ministerial Counsel on Drug Strategy, 2006). To censure they anticipate for engaging in that behaviour
differentiate underage drinkers at an increased risk of (Bandura, 1986). Despite the comprehensive theoretical
experiencing alcohol-related harms it is therefore vital to framework offered by SCT to examine anticipated social
investigate the social approval or disapproval adolescents outcomes, the social cognitive conceptualisation of antici-
anticipate for being drunk, and not just the social evaluations pated social outcomes has not yet been applied to underage
they anticipate for drinking alcohol. Yet, despite evidence that drinking.
Downloaded by [Universitat de les Illes Balears] at 09:37 10 February 2016

a portion of adolescents do consume alcohol with the The central aim of the current study was therefore to
expressed intent of becoming intoxicated (White & Bariola, examine the relationship between anticipated social outcomes
2012), few, if any, studies have investigated the social and engagement in underage drinking and experience of
evaluations adolescents anticipate for being drunk. Hence, alcohol-related harms from a social cognitive framework.
there is a need to separately investigate the social outcomes Since intoxication is closely associated with alcohol-related
adolescents anticipate for drinking alcohol and for being harms (Ministerial Counsel on Drug Strategy, 2006), it was
drunk and how these relate to adolescents’ engagement in hypothesised that, for underage drinkers, anticipatory social
underage drinking and experience of alcohol-related harm. outcomes for being drunk would be more strongly associated
In addition to examining whether anticipated social evalu- with adolescents experiencing alcohol-related harm than
ations differ for different alcohol-related behaviours, it is also would anticipatory social outcomes for drinking. During
essential to consider whether anticipated social evaluations mid-adolescence, as peers begin to exert a stronger influence
differ for different social figures. Parents and peers were than parents on youths, peer attitudes to drinking have been
selected as social figures to examine because they have shown to more strongly predict adolescent drinking than
consistently been shown to strongly influence adolescents’ parental influences (Aas & Klepp, 1992; Andrews et al., 1993;
alcohol use (Bahr, Hoffman, & Yang, 2005). Although most Ford & Hill, 2012; Kristjansson et al., 2010; Yu, 1998).
research investigating the social evaluations adolescents Moreover, the limited research that has examined mothers and
anticipate receiving for their alcohol use has examined fathers separately, found that disapproval from mothers was a
anticipated social approval and censure from parents and stronger predictor of alcohol use than disapproval from fathers
peers separately (Aas & Klepp, 1992; Kristjansson et al., (Andrews et al., 1993). Consequently, it was expected that
2010), there has been limited differentiation between the social anticipatory social outcomes from peers would be most
influences of mothers and fathers. However, the limited strongly associated with experiencing harm as a result of
research examining paternal and maternal influences on drinking, followed by anticipatory social outcomes from
underage drinking demonstrates that mothers and fathers do mothers, then fathers. These hypothesised relationships,
not necessarily uniformly influence their children’s alcohol between anticipatory social outcomes and engagement in
consumption (Andrews, Hops, Ary, Tildesley, & Harris, 1993; underage drinking, and between anticipatory social outcomes
Pettersson, Linden-Boström, & Eriksson, 2009; Van der Vorst, and drinkers’ experience of alcohol-related harm, were
Engels, Meeus, & Dekovic, 2006). Rather, adolescents may expected to be consistent across gender and grade. SCT
anticipate different levels of approval or disapproval from their (Bandura, 1986) posits that even though the proclivity to
mothers than from their fathers, and these anticipated social anticipate social approval may vary depending on demo-
outcomes differentially predict adolescents’ alcohol use. To graphic factors, such as grade and gender, when approval is
target interventions more effectively, it is therefore essential to anticipated, it is expected to relate to the prospective behaviour
investigate the social evaluations adolescents anticipate from in the same way regardless of gender or age.
their mother, father and peers for drinking alcohol and being Despite no expected difference in the relationship between
drunk, and also how anticipated social evaluations relate to anticipated social outcomes and underage drinking across
adolescents’ engagement in underage drinking and experience grade and gender, the proclivity to anticipate social approval
of alcohol-related harm. may still differ by grade and gender. Therefore, another aim of
The comprehensive theoretical framework of SCT pro- this study was to examine mean differences in anticipated
vides a basis for examining the role of anticipated social social outcomes, for drinking alcohol and being drunk, for
evaluations. From the social cognitive perspective (Bandura, grade, gender and social figure. Mid-adolescents have been
1986), anticipated social outcomes for behaviour, are found to anticipate more approval for drinking alcohol from
formed by observing the outcomes experienced by others their peers than their parents (Aas & Klepp, 1992; Yu, 1998)
(modelling), by performing the behaviour and personally and fathers have been found to be more permissive towards
DOI: 10.3109/16066359.2015.1014346 Adolescents’ anticipated social outcomes 255

alcohol use than mothers (Pettersson et al., 2009; Van der Procedure
Vorst et al., 2006). The limited research that has examined
Questionnaires were administered in classrooms or halls in
grade and gender differences with mid-adolescents has found
groups of approximately 20 to 100 students. Students were
that both parents are equally more permissive towards older
assured verbally and in writing of the anonymity of their
than younger adolescents’ alcohol use (Van der Vorst et al.,
responses and were identified on their questionnaires through
2006) and that peer disapproval for alcohol use has been
a unique identification number. Students were supervised by
found to decline throughout adolescence (Mrug & McCay,
research assistants and their subject teacher. To ensure
2013). Additionally, girls have been found to anticipate
confidentiality, students were asked not to discuss their
greater disapproval from parents and anticipate less favour-
answers with peers and teachers remained at the front of the
able attitudes from peers for drinking than do boys (Griffin,
classroom throughout testing. All questions during the
Scheier, Botvin, & Diaz, 2000; Mrug & McCay, 2013).
administration of the survey were answered by trained
Consequently, it was hypothesised that regardless of adoles-
research assistants. Students took 45 to 50 min to complete
cents’ grade or gender, adolescents would anticipate the
the questionnaire.
least social censure, for drinking alcohol and being drunk,
from their peers, followed by the father and then their
mother. It was further hypothesised that older adolescents Measures
and boys would anticipate less social censure for drinking Lifetime drinking
alcohol and being drunk than would younger adolescents
Downloaded by [Universitat de les Illes Balears] at 09:37 10 February 2016

and girls. Students responded to the question ‘‘Have you ever tried
To address these aims, two scales were developed to alcohol?’’ using a 3-point scale (0 ¼ No, 1 ¼ Yes, a sip or a
separately assess adolescents’ anticipated social outcomes for taste, 2 ¼ Yes, I’ve had at least a full standard drink of
drinking and for being drunk. These scales were based on the alcohol) (McBride, Farringdon, Midford, Meuleners, &
current literature, qualitative data and Bandura’s (1986) Phillips, 2006; Newton, Vogel, Teesson, & Andrews, 2009).
SCT. Each scale was comprised of three sub-scales, repre- A full standard drink (10 g of alcohol) was visually depicted.
senting social outcomes anticipated from mothers, fathers
and peers. Alcohol-related harm
Experience of alcohol-related harm was measured using the
Methods Alcohol-Related Harm Index from the CLIMATE School
Participants survey (Newton et al., 2009). Only 10 of the 12 items from the
CLIMATE survey were used in this study. Two items, related
Six hundred and fifty-one (329 female) students in grades to sexual activity after consuming alcohol, were not included
8 (n ¼ 308, Mage ¼ 13.51 years, age range: 12–14 years) and because the ethics committee deemed that the risk of harm or
10 (n ¼ 343, Mage ¼ 15.43, age range: 14–16 years) from 10 discomfort to participants was not outweighed by the
regional and metropolitan non-government secondary potential benefits of these items. Respondents rated on a
schools participated in this study. The participants were 6-point scale (1 ¼ never and 6 ¼ 12 or more times) how often,
81% White, 8% Asian, 5% Middle Eastern and 6% other. in the past three months, they experienced harm as a
The majority of participants were raised by their biological consequence of drinking alcohol. Example items include
mother and father (94%), 3% were raised by a biological ‘‘How many times did you have a hangover after drinking?’’
parent and step parent, 1% were raised by adopted parents or and ‘‘On how many occasions did you damage something
grandparents and the remaining participants were raised only because you were affected by alcohol?’’ The Alcohol-Related
by their biological mother (2%). Of the participants raised Harm Index had high internal consistency (a ¼ 0.89), com-
only by their biological mother, seven participants identified parable to McBride et al.’s (2006) original scale (a ¼ 0.90).
a relative as a significant male figure who raised them (e.g. McBride et al.’s (2006) original Index had good test–retest
grandfather, uncle) and the remaining five participants did reliability (a ¼ 0.89).
not identify a significant paternal figure. The final sample
only included those adolescents who identified both a
Anticipated social outcomes for drinking alcohol
significant maternal and paternal figure who raised them
and for being drunk
(N ¼ 646).
Convenience sampling was used to select schools, with The Drink and the Drunk ASO Scales both consisted of three
Sydney metropolitan and regional schools approached to identical stems separately assessing social outcomes from
participate in the study. Information and consent forms were mothers, fathers, and friends. Friends were specifically
sent home to parents/guardians of all Year 8 and 10 students referenced instead of peers, as friends exert a stronger
from the 10 participating schools. All students who returned influence on adolescents’ drinking than do acquaintances (Dal
their consent forms entered a draw to win 1 of 40 cinema Cin et al., 2009; Song, Smiler, Wagoner, & Wolfson, 2012).
tickets. Of the potential participants, 30% participated in the These three stems had the same three response items, yielding
study. In total 35% of potential participants returned their a total of 9-items for each scale.
consent forms and of these 84% of parents agreed to their The response scales were derived from a separate quali-
child’s participation in the study. Students who received tative study (Quinn & Bussey, 2012; see Table 1). In the
parent/guardian consent, and who gave written consent qualitative study, adolescents were asked an open-ended
themselves (99%) participated in the study. question regarding the social outcomes they anticipated for
256 C. A. Quinn & K. Bussey Addict Res Theory, 2015; 23(3): 253–264

Table 1. Responses to six open-ended questions: ‘‘What would your mother/father/friend think of you for drinking alcohol/being
drunk?’’.

Anticipated social outcome Mother Father Friend


Social censure 77.5 (91.4) 69.3 (87.5) 40.3 (58.7)
Disapproving, sad, upset 9.0 (6.1) 5.8 (6.0) 4.3 (3.3)
Angry, mad, furious 14.3 (20.4) 14.9 (23.3) 3.0 (5.3)
Shock, surprise 1.3 (1.6) 1.3 (0.8) 5.3 (3.2)
Stupid, irresponsible 18.2 (21.5) 15.3 (20.2) 12.1 (19.6)
Teaching, bad decision, wrong choice, too young, against law 6.7 (4.9) 4.6 (4.5) 3.0 (5.0)
Hanging round bad group, bad friends 0.3 (0.7) 0.3 (0.7) 0.0 (0.0)
In trouble, lecture, grounding 1.8 (3.0) 2.2 (4.0) 0.0 (0.0)
Beating, ‘‘kick out of home’’ 2.3 (3.3) 2.5 (2.6) 0.0 (0.0)
Not be my friend anymore, avoid 0.0 (0.0) 0.0 (0.0) 0.3 (1.0)
Disappointed, loss of trust/respect, embarrassed, ashamed 17.9 (21.7) 16.7 (18.2) 5.5 (8.8)
Disgust, repulsion, screw up, unworthy, loser, idiot 2.6 (2.5) 2.7 (3.5) 4.0 (5.7)
Where did I go wrong? What didn’t I teach you? 0.3 (1.3) 0.5 (1.0) 0.0 (0.0)
Concern, worry, help, call parents, Why? 2.8 (4.4) 2.5 (2.7) 2.8 (6.8)
Neutral 4.7 (2.8) 6.3 (3.5) 19.2 (10.6)
Not care, not notice, nothing 4.7 (2.8) 6.3 (3.5) 17.4 (9.6)
My choice, my decision 0.0 (0.0) 0.0 (0.0) 1.8 (1.0)
Downloaded by [Universitat de les Illes Balears] at 09:37 10 February 2016

Social approval 15.2 (2.5) 21.2 (4.5) 29.9 (19.0)


Ok if small amount; as long as I don’t pass out 13.0 (1.0) 17.6 (2.3) 2.3 (0)
Happy, grateful 0.0 (0.2) 0.0 (0.0) 2.6 (1.2)
Smart, good, sensible, responsible 0.2 (0.0) 0.7 (0.7) 1.2 (0.7)
Ok, alright, that’s normal 1.5 (0.5) 1.7 (0.2) 9.6 (3.8)
Cool, legend, gangster, I’m awesome 0.0 (0.0) 0.3 (0.0) 6.8 (2.3)
Encourage, joining in, finally 0.5 (0.0) 0.7 (0.3) 6.9 (2.7)
Funny, amusing 0.0 (0.8) 0.2 (1.0) 0.5 (8.3)
Other (e.g. I don’t know, it depends) 2.6 (3.3) 3.2 (4.5) 10.6 (11.6)

In the table the percentage of responses for ‘‘drinking’’ items is presented first followed by percentage of responses for ‘‘being drunk’’
items in parentheses.
The analysis is based on a sample of 619 (362 female) students, predominately white Australian (80%) in grades 9 (n ¼ 309, Mage ¼ 14.33
years, age range: 13–16 years) and 11 (n ¼ 310, Mage ¼ 16.21 years, age range: 15–17) from four non-government secondary schools.
Informed written consent was obtained from participants and passive consent was obtained from their parents.
Bandura’s (1986) SCT, posits that while the majority of social outcomes reflect social approval or social censure there may also be
neutral responses that are neither approving nor disapproving. In accord with this theoretical approach open-ended questions were
divided into the following categories: social approval, neutral and social censure. The sub-categories within these responses were
derived using the modified analytic induction method (Gilgun, 1995). Twenty percent of all questions were double-coded. There was an
acceptable level of agreement with the kappa statistic for lower level categories ranging from 0.55 to 0.90 (Landis & Koch, 1977).

drinking alcohol and for being drunk. Three broad categories 4-point response scales which followed were the same as
of anticipated social outcomes, consistent with SCT those for the Drink ASO Scale, except the responses
(Bandura, 1986), were identified: social approval, indiffer- specifically referred to being drunk instead of drinking
ence/neutrality and social censure. Each of these categories alcohol.
were represented in the Drink and Drunk ASO Scales, with
the most frequent socially approving (i.e. think it is ok) and
Data management
neutral (i.e. not care) open-ended responses included in the
scales. The second most frequent socially censuring open- Drinker sub-sample. Students were classified as drinkers if
ended response (i.e. be disappointed) was also included, since they had consumed a full standard drink of alcohol
it reflected the theoretical construct of a social evaluation of (Agostinelli & Grube, 2005; Callas, Flynn, & Wordon,
the person more than did the most frequent social censuring 2004; Kelly et al., 2011). Therefore, for subsequent analyses
open-ended response, which reflected an attribution towards engagement in underage drinking is assessed through a
behaviour. dichotomised version of the lifetime drinking item (0 ¼ have
For the Drink ASO Scale students were presented with the not consumed a full standard drink in their lifetime; 1 ¼ have
stem ‘‘You are drinking an alcoholic drink. Your (mother) consumed a full standard drink in their lifetime). The drinker
would . . .’’. This stem was answered for each of three 4-point sub-sample included those students who had consumed at
response scales (1 ¼ be totally disappointed with you for least one standard drink in their lifetime. In this sample, there
drinking alcohol and 4 ¼ be totally pleased with you were 244 (145 male) students in grades 8 (n ¼ 55,
for drinking alcohol; 1 ¼ think it is totally not ok that you Mage ¼ 13.53 years, age range: 13–15 years) and 10
are drinking alcohol and 4 ¼ think it is totally ok that you are (n ¼ 189, Mage ¼ 15.46, age range: 15–16 years).
drinking alcohol; 1 ¼ totally care that you are drinking alcohol
and 4 ¼ totally not care that you are drinking alcohol). Missing data. There were small amounts of missing data at
For the Drunk ASO Scale, students were presented with the item level (0 to 2.8%). All missing data was imputed using
the stem ‘‘You are drinking an alcoholic drink and it is the expectation-maximisation (EM) algorithm in SPSS. This
obvious you are drunk. Your (mother) would . . .’’. The three procedure has been shown to be superior to means
DOI: 10.3109/16066359.2015.1014346 Adolescents’ anticipated social outcomes 257

substitution, pair-wise deletion or list-wise deletion (Enders, was conducted on the total sample, with the Drink ASO Scale
2001; Schafer & Graham, 2002). as the dependent variable. The second repeated measures
analysis of variance was conducted on the drinker sub-sample
School clustering. The data used in the present study are with the Drunk ASO Scale as the dependent variable.
nested (students within different schools). Therefore, analyses The repeated measures design for both analyses was
were conducted to examine the potential effect of clustering 2 (gender: boy, girl)  2 (grade: 8, 10)  3 (social figure:
of responses within schools on subsequent analyses. The mother, father, peers) with gender and grade as between
linear-mixed model procedure in SPSS was used to examine subject factors and social figure as a within subject
the possible lack of independence of responses which could factor. Where significant interactions were examined post
have resulted from belonging to a particular school. Each of hoc, the Bonferroni method was used with an overall alpha
the variables within the study was fitted to a random intercept of 0.01.
model, with school entered as a random factor. The intra-class Partial correlations between all measures in the study were
correlations ranged from 0.00 to 0.14 indicating a significant conducted for the total sample and for the drinker sub-sample.
clustering of school. To account for this non-independence, School was controlled in these analyses. Next, to examine the
the mixed model procedure and generalised estimating association between the Drink ASO sub-scales and engage-
equations were used when conducting analysis of variance ment in underage drinking a logistic regression was conducted
and regression analyses, as detailed below. In effect, these using generalised estimation equations, on the total sample,
procedures enable a multilevel model, with individual with underage drinking as the dependent variable and grade
Downloaded by [Universitat de les Illes Balears] at 09:37 10 February 2016

students at level 1 and school at level 2 (Peugh & Enders, and gender as control variables. Finally, to examine the
2005). association between the Drunk ASO sub-scales and alcohol-
related harm, for underage drinkers, a linear regression, using
mixed model procedure, was conducted on the drinker sub-
Statistical analysis
sample with alcohol-related harm as the dependent variable
First, the factor structure of the Drink and Drunk ASO Scales and grade and gender as control variables. For the latter two
were examined using exploratory and confirmatory factor analyses, all Drink ASO sub-scales and Drunk ASO scales
analyses. For the Drink ASO Scale, factor analyses were were initially interacted with grade and gender to test for
conducted on the total sample. For the Drunk ASO Scale, possible differences in the results between grades and gender.
factor analyses were conducted on the drinker sub-sample. For As the alcohol-related harm and the Drunk ASO sub-scales
both 9-item scales, which consisted of three response options were positively skewed, each of these variables were log10
for three social figures (mother, father and peers), exploratory transformed for all analyses.
factor analyses using principle axis extraction and Oblimin
rotation were conducted. Confirmatory factor analyses (CFA) Results
were then conducted using maximum likelihood estimation to
Factor structure of the Drink and Drunk ASO Scales
examine whether the factor structure obtained in the explora-
tory factor analysis could be replicated. To determine good- The exploratory factor analysis, for both 9-item scales,
ness-of-fit of the models the 2 statistic and several other fit revealed a scree plot with a four-factor solution. However,
indices were examined. Since the 2 statistic is sensitive to the fourth factor did not have any factor loadings above 0.25.
sample size and the number of variables included in a model, Therefore, a factor analysis was conducted for each ASO
the Comparative Fit Index (CFI), Tucker Lewis Index (TLI) Scale, with a three-factor solution specified. This resulted in a
and Root Mean Square Error of Approximation (RMSEA) meaningful three-factor solution: mother, father and peers,
were also examined (Hox & Bechger, 2001; Hu & Bentler, which accounted for 84% of the variance of the Drink ASO
1999; Kaplan, 2000; Vandenberg & Lance, 2000). Due to scale and 81% of the variance for the Drunk ASO scale. The
similarity in wording, the errors of the three sets of items with final correlation matrix revealed correlations between the
the same stem response were allowed to correlate since the three factors ranging from 0.50 to 0.72 for the Drink ASO
failure to allow these correlations could result in a misleading Scale and 0.32 to 0.53 for the Drunk ASO Scale. The mother,
interpretation of the model (Cole, Ciesla, & Steiger, 2007). father and peers sub-scales consisted of three items each, with
Measurement and structural invariance across grade and all factor loadings above 0.70. The Cronbach Alpha
gender were examined for both scales. Configurable invariance reliabilities indicated good internal consistency for the
was first determined through an examination of the uncon- Drink ASO sub-scales (mother: a ¼ 0.88; father: a ¼ 0.92;
strained model. Then, the factor loadings were constrained to peers: a ¼ 0.91) and for the Drunk ASO sub-scales (friends:
be the same across groups and finally the parameters and a ¼ 0.85; father: a ¼ 0.91; mother: a ¼ 0.87).
structural components of the model were constrained to be The model for the CFA, depicted in Figure 1, obtained
equivalent across grade and across gender. When the con- reasonable fit for the Drink ASO Scale, 2(15,
strained models were compared to the unconstrained model, a N ¼ 646) ¼ 41.14, p50.001, CFI ¼ 0.99, TLI ¼ 0.99,
DCFI50.01 was used to determine measurement and struc- RMSEA ¼ 0.05, and excellent fit for the Drunk ASO Scale,
tural invariance (Cheung & Rensvold, 2002). 2(15, N ¼ 244) ¼ 19.71, p ¼ 0.183, CFI ¼ 1.00, TLI ¼ 0.99,
Repeated measures’ analyses of variance using mixed RMSEA ¼ 0.04. For both scales, invariance across gender and
model procedure were then conducted to examine grade, grade was indicated for the factor loadings and the structural
gender, and social figure differences for the Drink and Drunk components of the model with the DCFI ranging from50.001
ASO Scales. The first repeated measures analysis of variance to 0.006.
258 C. A. Quinn & K. Bussey Addict Res Theory, 2015; 23(3): 253–264

Figure 1. Confirmatory factor analysis for the (a)


Drink and Drunk ASO Scales. F F F
M M M
ASO1 ASO2 ASO3 ASO1 ASO2 ASO3

.91 .79 .93 .84


.84 .91

Mother Father
Drink
.86 ASO .84

.60

Peers

.84
.85
.95
Downloaded by [Universitat de les Illes Balears] at 09:37 10 February 2016

P P P
ASO1 ASO2 ASO3

(b)

M M M F F F
ASO1 ASO2 ASO3 ASO1 ASO2 ASO3

.83 .82 .91 .86


.86
.85
Father
Mother Drunk
ASO .68
.77

.46

Peers
.85
.86 .74

P P P
ASO1 ASO2 ASO3

Due to the large sample size it was deemed important to variance yielded significant main effects for grade, F(1,
replicate the findings for the exploratory and confirmatory 642) ¼ 167.37, p50.001, and social figure, F(2,
factor analyses on a smaller sample. The total sample and the 641) ¼ 363.12, p50.001. These significant main effects
drinker sub-sample were randomly divided into two. were qualified by the significant two-way interactions
Exploratory factor analyses were conducted on the first between social figure and gender, F(2, 641) ¼ 10.36,
samples (N ¼ 327; N ¼ 124), obtaining the same three-factor p50.001, and between social figure and grade, F(2,
solution as described above. CFAs were then conducted 641) ¼ 6.99, p ¼ 0.001.
on the second samples. Adequate model fit was obtained For the interaction between social figure and gender
for the three-factor model for both the Drink ASO, 2 (15, (Figure 2), boys and girls anticipated more social approval
N ¼ 319) ¼ 17.19, p50.001, CFI ¼ 0.99, TLI ¼ 0.99, from their peers than from their mother, t(644) ¼ 19.62,
RMSEA ¼ 0.02) and Drunk ASO, 2 (15, N ¼ 120) ¼ 18.49, p50.001, t(644) ¼ 13.53, p50.001 and father, t(644) ¼
p ¼ 0.238, CFI ¼ 0.99, TLI ¼ 0.99, RMSEA ¼ 0.04. 17.71, p50.001, t(644) ¼ 15.18, p50.001. However, there
was no significant difference in the social outcomes that boys
Social figure effects by grade and gender and girls anticipated receiving from their father compared
to their mother, t(644) ¼ 1.91, p ¼ 0.171, t(644) ¼ 1.65,
Drink ASO
p ¼ 0.300. Further comparisons revealed that boys anticipated
The 2 (gender: boy, girl)  2(grade: 8, 10)  3(social figure: more social approval from their peers than did girls, however,
mother, father and peers) repeated measures analysis of there was no significant difference for father or mother,
DOI: 10.3109/16066359.2015.1014346 Adolescents’ anticipated social outcomes 259
8 p50.001, or fathers, t(644) ¼ 14.07, p50.001, but did not
7.5 differ in the social outcomes anticipated from mothers
7
compared to fathers, t(644) ¼ 1.20, p ¼ 0.700. There was also
a significant main effect for grade, F (1, 240) ¼ 6.12,
6.5
p ¼ 0.014, with Grade 10 students anticipating more positive
Mean Drink ASO Score

6 social outcomes for being drunk (M ¼ 0.69, SD ¼ 0.02) than


5.5 Mother
Grade 8 students (M ¼ 0.69, SD ¼ 0.02). The main effect for
Father gender was not significant, F (2, 639) ¼ 0.01, p ¼ 0.922.
5
Peers
4.5 Correlations
4 Means, standard deviations and partial correlations between
3.5 the observed variables are depicted in Table 2 for the total
sample and for the drinker sub-sample. The partial correl-
3
Boy Girl ations, which controlled for school effects, revealed a signifi-
Gender cant positive relationship between all ASO sub-scales. These
correlations were stronger for the total sample than for the
Figure 2. Mean drink ASO score as a function of social figure and drinker sub-sample. For the total sample, the Drink and Drunk
gender.
Downloaded by [Universitat de les Illes Balears] at 09:37 10 February 2016

ASO sub-scales were all positively related to engagement in


underage drinking and this positive association was stronger
8
for the Drink than the Drunk ASO sub-scales. For the drinker
sub-sample, the Drunk ASO sub-scales were all positively
7.5
associated with alcohol harm. However, of the Drink ASO sub-
7 scales, only the peers and father sub-scale were significantly
6.5
associated with alcohol-related harm, and this relationship was
Mean Drink ASO Score

weaker than for the Drunk ASO sub-scales.


6

5.5 Mother Engagement in underage drinking and anticipated social


5
Father outcomes
Peers
4.5 For the generalised estimation equation, examining the
association between the Drink ASO sub-scales and underage
4
drinking, none of the interactions involving grade and gender
3.5 interactions were significant, so they were removed from the
3 analysis. In the final model, as presented in Table 3 (Model
Grade 8 Grade 10 1), results of the generalised estimation equation revealed that
Gender
grade was significantly associated with underage drinking.
Figure 3. Mean drink ASO score as a function of social figure and Older students were more likely to drink underage than were
grade. younger students. The sub-scales for peers and mother also
significantly predicted underage drinking. More social
approval from mother or peers was related to an increased
t(644) ¼ 0.78, p ¼ 0.436, t(644) ¼ 2.40, p ¼ 0.016. For the probability of engaging in underage drinking.
interaction between social figure and grade (Figure 3),
students in Grade 8 and Grade 10 anticipated more social Alcohol-related harm and anticipated social outcomes
approval from their peers than from their mother,
For the linear-mixed model analysis, while examining the
t(644) ¼ 14.15, p50.001, t(644) ¼ 19.18, p50.001 and
association between the Drunk ASO sub-scales and alcohol-
father, t(644) ¼ 13.69, p50.001, t(644) ¼ 19.37, p50.001.
related harm, none of the interactions involving grade and
As expected, there was no significant difference in the social
gender interactions were significant, so they were removed
outcomes that Grade 8 and Grade 10 students anticipated
from the analysis. In the final model, as seen in Table 3
receiving from their father compared to their mother,
(Model 2), results revealed that gender was significantly
t(644) ¼ 0.46, p ¼ 1.000, t(644) ¼ 0.18, p ¼ 1.000. Further
related to alcohol-related harm. Girls were more likely to
comparisons revealed that Grade 10 students anticipated more
experience harm as a consequence of their drinking than were
positive social outcomes from their mother, father and peers
boys. The sub-scales for peers and mother were also
than did Grade 8 students, t(644) ¼ 6.42, p50.001,
significantly associated with alcohol-related harm. More
t(644) ¼ 5.79, p50.001, t(644) ¼ 10.61, p50.001.
social approval from mother or peers was related to higher
Drunk ASO levels of alcohol-related harm.

This repeated measures analysis of variance yielded significant


main effects for social figure, F(2, 239) ¼ 139.26, p50.001.
Discussion
Post hoc comparisons revealed that students anticipated more This is the first study to separately examine the social
social approval from peers than mothers, t(644) ¼ 15.27, outcomes adolescents anticipate from their mother, father and
260 C. A. Quinn & K. Bussey Addict Res Theory, 2015; 23(3): 253–264

Table 2. Summary of partial correlations, means, and standard deviations for analszed variables, for the total sample (N ¼ 646) and drinker sub-sample
(N ¼ 244).

Alcohol harmb 2 3 4 5 6 7 M SD
a
1. Initiated drinking _ 0.46*** 0.42*** 0.54** 0.37*** 0.34*** 0.48*** 402 62.2
2. Drink ASO – Mother 0.10 _ 0.67** 0.47*** 0.64*** 0.50*** 0.38*** 4.67 1.78
3. Drink ASO – Father 0.14* 0.56** _ 0.48*** 0.45*** 0.71*** 0.39*** 4.69 1.91
4. Drink ASO – Peers 0.17** 0.20*** 0.26*** _ 0.36*** 0.35*** 0.79*** 7.08 2.34
5. Drunk ASO – Motherb 0.31*** 0.61*** 0.33*** 0.23*** _ 0.55*** 0.41*** 3.67 1.25
6. Drunk ASO – Fatherb 0.24*** 0.39*** 0.69*** 0.15* 0.48*** _ 0.41*** 3.99 1.48
7. Drunk ASO – Peersb 0.24*** 0.12 0.18** 0.66*** 0.32*** 0.29*** _ 5.91 2.20
M 12.90 5.73 5.73 8.71 4.25 4.63 7.26
SD 4.37 1.82 1.98 1.73 1.55 1.69 1.91

All correlations were conducted controlling for school. Inter-correlations for the total sample are presented above the diagonal, and inter-correlations
for the drinker sub-sample are presented below the diagonal. Means and standard deviations for the total sample are presented in the vertical columns,
and means and standard deviations for the drinker sub-sample are presented in horizontal rows. For all ASO scales higher scores are indicative of less
social censure and higher alcohol harm and initiated drinking scores are indicative of greater alcohol engagement.
a
Number and frequency presented instead of mean and standard deviation (0 ¼ never consumed a full standard drink).
b
Log10 transformed due to initial skewness, however, means and standard deviations are for the untransformed variable.
*p50.05, **p50.01, ***p50.001.
Downloaded by [Universitat de les Illes Balears] at 09:37 10 February 2016

Table 3. Generalised estimated equation for the effects of drink ASO on initiated drinking and linear-mixed
model for the effects of drunk ASO on alcohol harm.

Model 1 Model 2
Estimate of
OR 95% CI fixed effects SE t
Grade 3.11*** 2.03–4.76 0.03 0.02 1.68
Gender 0.79 0.51–1.23 0.03 0.02 2.05*
Drink SOE – Mother 1.39*** 1.17–1.65
Drink SOE – Father 1.11 0.96–1.28
Drink SOE – Peers 1.64*** 1.44–1.88
Drunk SOE – Mother 0.19 0.06 3.16**
Drunk SOE – Father 0.07 0.05 1.29
Drunk SOE – Peers 0.14 0.06 2.30*

OR, odds ratio; CI, confidence interval. Model 1 was conducted on the total sample (N ¼ 646); has initiated
drinking (0 ¼ never consumed a full standard drink, 1 ¼ have consumed full standard drink) as the
dependent variable and depicts logistic odds ratio derived from generalised estimation equations. Model 2
was conducted on the drinker sub-sample (n ¼ 244) using mixed model procedure. Alcohol harm (log10
transformed) was the dependent variable. All analyses were conducted controlling for clustering within
school.
*p50.05, **p50.01, ***p50.001.

peers for drinking alcohol and being drunk, using a social anticipated social outcomes from mothers or fathers, which
cognitive framework (Bandura, 1986). As expected, the is in agreement with other literature and highlights the
exploratory and confirmatory factor analyses showed that importance of peer influence on underage drinking (Allen,
anticipated social outcomes from mother, father and peers Donohue, Griffin, Ryan, & Turner, 2003; Ford & Hill, 2012;
formed separate sub-scales, within the Drink and Drunk ASO Mrug & McCay, 2013; Nash, McQueen, & Bray, 2005; Yu,
Scales. These sub-scales had good internal consistency, and 1998). Although social approval from peers related to
multiple group analysis, for grade and gender groups, increased engagement in underage drinking, social censure
supported invariance of the factor models across males and also related to increased abstinence. Often research empha-
females, and across older and younger mid-adolescents. These sises peers as negative influences on adolescents’ alcohol
findings support the investigation of anticipated social use. However, the protection peers may offer against
outcomes from different social figures, as being related, but engagement in underage drinking should also be emphasised
conceptually independent. Furthermore, high levels of social (Allen, Chango, Szwedo, Schad, & Marston, 2012; Maxwell,
approval for drinking alcohol were correlated with greater 2002). As Kumar, O’Malley, Johnston, Schulenberg,
engagement in underage drinking in the overall sample. and Bachman (2002) suggest creating an environment of
However, when underage drinkers were examined separately, disapproval of substance use within schools is likely to
high levels of social approval, or low levels of social censure, reduce adolescent alcohol use. Creating such an environment
for being drunk were most strongly correlated with adoles- may enable peers to more actively express disapproval of
cents’ experience of alcohol-related harm. alcohol use, and consequently reduce the potential of
Anticipated social outcomes from peers were more adolescents perceiving social acceptance of underage
strongly associated with engagement in drinking than were drinking.
DOI: 10.3109/16066359.2015.1014346 Adolescents’ anticipated social outcomes 261

It is notable that anticipated social outcomes from mothers Meeus, Deković, & Van Leeuwe, 2005). Therefore, it may
related more strongly to underage drinkers experiencing be that the negative social outcomes anticipated from mothers
alcohol-related harm than did anticipated social outcomes are more salient for adolescents than those for fathers and
from peers. This finding emphasises the importance of are consequently more strongly related to their engagement
parents, even after adolescents start drinking, and the role in underage drinking and experience of alcohol-related
they play in curbing or facilitating the alcohol-related harm harm. Future research should explore maternal and paternal
underage drinkers’ experience. In examining the implication influences on underage drinking to a greater extent and
of these findings it is necessary to note that the majority of what factors account for the stronger influence of anticipated
adolescents anticipated social censure from their mother for social outcomes associated with mothers than with fathers.
drinking alcohol and being drunk. Therefore, it is not just As hypothesised, when the relationships between antici-
overt approval that relates to drinking but also the degree of pated social outcomes and engagement in underage drinking,
social censure adolescents anticipate receiving. As Nash et al. and experience of alcohol-related harm, were examined no
(2005) found, it is strong parental disapproval that deters grade or gender interactions were significant. The lack of
adolescent alcohol use. This has important implications for significant interactions suggests that the relationship between
interventions. anticipated social outcomes and underage drinking is the
Parents may introduce their children to alcohol in same regardless of the gender or age of the adolescent.
‘‘supervised settings’’ with the aim of minimising alcohol- However, mean differences in anticipated social outcomes
related harm and providing a safe social development to using were found for boys and girls, and for older and younger mid-
Downloaded by [Universitat de les Illes Balears] at 09:37 10 February 2016

alcohol (Gilligan, Kypri, & Lubman, 2012). Some research adolescents.


suggests that such practices can reduce adolescents’ excessive In line with previous research, all adolescents anticipated
alcohol consumption (Foley, Altman, Durant, & Wolfson, less social censure from their peers than from their mothers or
2004). However, other research suggests that parental provi- fathers (Mrug & McCay, 2013). Unexpectedly, however,
sion of alcohol may normalise underage drinking and adolescents did not differ in the levels of social censure they
implicitly condone more risky drinking behaviours (Komro, anticipated receiving from their mothers compared to their
Maldonado-Molina, Tobler, Bonds, & Muller, 2007; fathers. It is possible that despite fathers consuming more
Livingston, Testa, Hoffman, & Windle, 2010). An area for alcohol than mothers, they may still communicate equal levels
future research is to determine how adolescents interpret of disapproval for their adolescents’ drinking.
parental supply of alcohol. This study shows that adolescents As expected, older students anticipated less social censure
do anticipate different social outcomes for drinking alcohol for drinking alcohol, from their mothers, fathers and
and being drunk and that, particularly in relation to mothers, it peers than did younger students (Mrug & McCay, 2013;
is the social sanctions underage drinkers anticipate receiving Van Der Vorst et al., 2006) and this finding was consistent for
for being drunk, not for drinking, which relates to their social outcomes related to drinking alcohol, and for being
experience of alcohol-related harm. Particularly in Australia, drunk. The only difference in anticipated social outcomes
there is an increased focus on the need to educate and support for boys compared to girls, however, was from peers for
parents in relation to their adolescents’ alcohol use (Gilligan drinking alcohol; boys anticipated less social censure from
et al., 2012; Hayes, Smart, Toumbourou, & Sanson, 2004). To their friends for drinking alcohol than did girls. A possible
aid such efforts, future research could examine how parental reason for this is that in Australia, despite a recent increase
supply of alcohol in different quantities, different settings and in the prevalence of female drinking, males still typically
within different ethnic groups impacts adolescents perception drink more alcohol than females (Australian Institute of
of the social outcomes they will receive if they drink alcohol Health and Welfare, 2011). In addition, historically, drinking
or become drunk. has been discouraged among women and encouraged among
This study found that anticipated approval or censure from men to promote mateship, solidarity and virility (Sargent,
fathers did not relate to adolescent engagement in underage 1973). It is possible that this view is still communicated
drinking or experience of alcohol-related harm, in the among adolescent peers, particularly for drinking but not
presence of anticipated social outcomes from mothers and necessarily for the more severe behaviour of drinking to
peers. It may be that fathers exert their influence on intoxication.
adolescents indirectly through mothers (Cabrera, Fitzgerald, The current study had several limitations. First, this study
Bradley, & Roggman, 2007; Mares, Van Der Vorst, Engels, & was cross-sectional in design, therefore, although associations
Lichtwarck-Aschoff, 2011). Other studies have shown that between anticipated social outcomes and underage drinking
fathers are generally less present at home and spend less time were explored, no causal statements can be made. It cannot be
with adolescents than do mothers (Lamb, 2000; William & ascertained whether less anticipated social sanctions leads to
Kelly, 2005). Adolescents also feel more emotionally close to more adolescents drinking, whether adolescents’ drinking
and cared for by their mothers than their fathers (Ackard, leads to decreased social sanctions or whether anticipated
Neumark-Sztainer, Story, & Perry, 2006; William & Kelly, social outcomes and underage drinking co-influence each
2005) and have greater ease communicating with their other. Future longitudinal studies are needed to map the
mothers than their fathers (Luk, Farhat, Iannotti, & Simons- relationship between anticipated social outcomes and under-
Morton, 2010). Furthermore, adolescents are more likely to age drinking. Further, the study did rely on self-report
ask mothers about alcohol than fathers (Kelly, Comello, & measurements; however, substantial evidence indicates that
Hunn, 2002) and mothers are more likely than fathers to adolescents provide reliable and valid reports of their
initiate conversations about alcohol (Van Der Vorst, Engels, substance use when they are assured of the confidentiality
262 C. A. Quinn & K. Bussey Addict Res Theory, 2015; 23(3): 253–264

and anonymity of their responses (Dolcini, Adler, & Research). We thank Alan Taylor for his consultation on the
Ginsberg, 1996), as was the case in this study. statistical analysis and Lorna Allen, Lina Asquini, Madeline
Another limitation of this study is that over 80% of the Jarrett, Joseph Lennon, Loretta Moore and Kate Stainlay for
current sample identified as White Australian, making it help with data collection.
impossible to investigate differences in anticipated social
outcomes for different racial and ethnic groups. Moreover, Declaration of interest
this study did not assess, or examine the contribution of
different family structures, such as separated families, on The authors declare no conflicts of interests. The authors
adolescents’ anticipated social outcomes for alcohol use. The alone are responsible for the content and writing of this
low participation rate obtained in this study also potentially article.
biased the current sample and limits the generalisability of the
current findings. However, this participation rate was not a References
reflection of children and their parents actively deciding not Aas, H., & Klepp, K. (1992). Adolescents’ alcohol use and perceived
to participate in the study. Rather, it was simply a result of norms. Scandinavian Journal of Psychology, 33, 315–325.
forms not being returned which is unlikely to have biased the Ackard, D.M., Neumark-Sztainer, D., Story, M., & Perry, C. (2006).
Parent–child connectedness and behavioral and emotional health
sample of the participants. Despite these limitations, the study among adolescents. American Journal of Preventive Medicine, 30,
had many strengths. These include, the use of an Australian 5966.
sample which contributes to the sparse research investigating Agostinelli, G., & Grube, J. (2005). Effects of presenting heavy drinking
Downloaded by [Universitat de les Illes Balears] at 09:37 10 February 2016

norms on adolescents’ prevalence estimates, evaluative judgments,


parental influences on alcohol use among Australian adoles-
and perceived standards. Prevention Science, 6, 89–99.
cents (Hayes et al., 2004). Additionally, this is one of a few Allen, J.P., Chango, J., Szwedo, D., Schad, M., & Marston, E. (2012).
studies to separately assess the social outcomes adolescents Predictors of susceptibility to peer influence regarding substance use
anticipate from their mothers and fathers, rather than parents in adolescence. Child Development, 83, 337–350.
Allen, M., Donohue, W.A., Griffin, A., Ryan, D., & Turner, M.M.
in general, for drinking alcohol and for being drunk. It is
(2003). Comparing the influence of parents and peers on the first
important for future research to extend the generalisability of choice to use drugs. Criminal Justice and Behavior, 30, 163–186.
these findings by replicating this study with a more diverse Andrews, J.A., Hops, H., Ary, D.V., Tildesley, E., & Harris, J. (1993).
sample obtained with a high participant response rate. It Parental influence on early adolescent substance use: Specific and
nonspecific effects. The Journal of Early Adolescence, 13, 285–310.
would then be possible to examine whether ethnicity and Australian Institute of Health and Welfare. (2011). 2010 National Drug
family structure moderate the relationships found between Strategy Household Survey (Drug Statistics Series No. 25. Cat. No.
anticipated social outcomes and adolescent alcohol use. PHE 145).
Moreover, although the ASO Scales used in this study were Bahr, S.J., Hoffmann, J.P., & Yang, X. (2005). Parental and peer
influences on the risk of adolescent drug use. The Journal of Primary
theoretically driven, consisted of items derived from qualita- Prevention, 26, 529–551.
tive research and demonstrated a sound factor structure across Bandura, A. (1986). Social foundations of thought and action.
grade and gender, to further enhance the applicability of these Englewood Cliffs, NJ: Prentice Hall.
scales for intervention, further validation and replication of Bava, S., & Tapert, S. F. (2010). Adolescent brain development and the
risk for alcohol and other drug problems. Neuropsychology Review,
the current results is required in future studies. 20, 398–413.
A key strength of this study is that it is the first to Bradley, B.J., & Greene, A.C. (2013). Do health and education agencies
separately investigate the social outcomes adolescents in the United States share responsibility for academic achievement and
anticipate from their mother, father and peers for drinking health? A review of 25 years of evidence about the relationship of
adolescents’ academic achievement and health behaviors. Journal of
alcohol and being drunk, underscoring the importance of Adolescent Health, 52, 523–532.
considering how social outcomes for different social Cabrera, N., Fitzgerald, H.E., Bradley, R.H., & Roggman, L. (2007).
figures and different alcohol-related behaviours may Modeling the dynamics of paternal influences on children over the life
differentially impact adolescents’ engagement in underage course. Applied Developmental Science, 11, 185–189.
Callas, P.W., Flynn, B.S., & Worden, J.K. (2004). Potentially modifiable
drinking and experience of alcohol-related harm. This psychosocial factors associated with alcohol use during early adoles-
study also highlights the notable influence of parents and cence. Addictive Behaviors, 29, 1503–1515.
peers on adolescents’ drinking self-regulation, and their Cauffman, E., Piquero, A.R., Kimonis, E., Steinberg, L., Chassin, L., &
potential role in intervention programs aimed at reducing Fagan, J. (2007). Legal, individual, and environmental predictors of
court disposition in a sample of serious adolescent offenders. Law and
underage drinking and its associated harms. Additionally, Human Behavior, 31, 519–535.
in accord with social norms theory (Perkins, 2003) and Cavazos-Rehg, P., Krauss, M.J., Spitznagel, E.L., Schootman, M.,
SCT (Bandura, 1986), it emphasises the importance for Cottler, L.B., & Bierut, L.J. (2011). Substance use and the risk for
intervention programs and research to not only consider sexual intercourse with and without a history of teenage pregnancy
among adolescent females. Journal of Studies on Alcohol and Drugs,
external social influences on underage drinking but also 72, 194–198.
how adolescents perceive these influences and how these Cheung, G.W., & Rensvold, R.B. (2002). Evaluating goodness-of-fit
conceptions subsequently motivate adolescents’ underage indexes for testing measurement invariance. Structural Equation
Modelling, 9, 233–255.
drinking and experience alcohol-related harm.
Chikritzhs, T., Pascal, R., & Jones, P. (2004). Under-aged drinking
among 14–17 year olds and related harms in Australia. (National
Alcohol Indicators, Bulletin No. 7). Retrieved from National Drug
Acknowledgements Research Institute website: http://www.ndri.curtin.edu.au/local/docs/
If possible can the following acknowledgement please be pdf/naip/naip007.pdf.
Clark, D.B., Thatcher, D.L., & Tapert, S.F. (2008). Alcohol, psycho-
included: Approval for this study was provided by the logical dysregulation, and adolescent brain development. Alcoholism:
Macquarie University Ethics Review Committee (Human Clinical and Experimental Research, 32, 375–385.
DOI: 10.3109/16066359.2015.1014346 Adolescents’ anticipated social outcomes 263
Cole, D.A., Ciesla, J.A., & Steiger, H. (2007). The insidious effects of Komro, K.A., Maldonado-Molina, M.M., Tobler, A.L., Bonds, J.R., &
failing to include design-driven correlated residuals in latent-variable Muller, K.E. (2007). Effects of home access and availability
covariance structure analysis. Psychological Methods, 12, 381–398. of alcohol on young adolescents’ alcohol use. Addiction, 102,
Crews, F., He, J., & Hodge, C. (2007). Adolescent cortical development: 1597–1608.
A critical period of vulnerability for addiction. Pharmacology, Kristjansson, A.L., Sigfusdottir, I.D., James, J.E., Allegrante, J.P., &
Biochemistry and Behavior, 86, 189–199. Helgason, A.R. (2010). Perceived parental reactions and peer respect
Crews, F.T., Mdzinarishvili, A., Kim, D., He, J., & Nixon, K. (2006). as predictors of adolescent cigarette smoking and alcohol use.
Neurogenesis in adolescent brain is potently inhibited by ethanol. Addictive Behaviors, 35, 256–259.
Neuroscience, 137, 437–445. Kumar, R., O’Malley, P.M., Johnston, L.D., Schulenberg, J.E., &
Dal Cin, S., Worth, K.A., Gerrard, M., Gibbons, F.X., Stoolmiller, M., Bachman, J.G. (2002). Effects of school-level norms on student
Wills, T.A., & Sargent, J.D. (2009). Watching and drinking: substance use. Prevention Science, 3, 105–124.
Expectancies, prototypes, and friends’ alcohol use mediate the effect Lamb, M.E. (2000). The history of research on father involvement: An
of exposure to alcohol use in movies on adolescent drinking. Health overview. Marriage and Family Review, 29, 23–42.
Psychology, 28, 473–483. Landis, J.R., & Koch, G.G. (1977). The measurement of observer
Dolcini, M.M., Adler, N.E., & Ginsberg, D. (1996). Factors influencing agreement for categorical data. Biometrics, 33, 159–174.
agreement between self-reports and biological measures of smoking Laslett, A., Catalano, P., Chikritzhs, Y., Dale, C., Doran, C., Ferris, J. . . .
among adolescents. Journal of Research on Adolescence, 6, 515–542. Wilkinson, C. (2010). The range and magnitude of alcohol’s harm to
Enders, C. (2001). A primer on maximum likelihood algorithms others. Fitzroy, Victoria: AER Centre for Alcohol Policy Research,
available for use with missing data. Structural Equation Modeling, Turning Point Alcohol and Drug Centre, Eastern Health.
8, 128–141. Lim, S.S., Vos, T., Flaxman, A.D., Danaei, G., Shibuya, K., Adair-
Fergusson, D.M., & Lynskey, M.T. (1996). Alcohol misuse and Rohani, H. . . . Ezzati, M. (2012). A comparative risk assessment
adolescent sexual behaviors. Pediatrics, 1, 91–96. of burden of disease and injury attributable to 67 risk factors and
Downloaded by [Universitat de les Illes Balears] at 09:37 10 February 2016

Foley, K.L., Altman, D., Durant, R.H., & Wolfson, M. (2004). Adults’ risk factor clusters in 21 regions, 1990–2010: A systematic analysis
approval and adolescents’ alcohol use. Journal of Adolescent Health, for the Global Burden of Disease Study 2010. Lancet, 380,
35, 17–26. 2224–2260.
Ford, J.A., & Hill, T.D. (2012). Religiosity and adolescent substance use: Livingston, J.A., Testa, M., Hoffman, J.H., & Windle, M. (2010). Can
Evidence from the national survey on drug use and health. Substance parents prevent heavy episodic drinking by allowing teens to drink at
Use & Misuse, 47, 787–798. home? Addictive Behaviors, 35, 1105–1112.
Foxcroft, D.R., & Tsertsvadze, A. (2011). Universal school-based Luk, J.W., Farhat, T., Iannotti, R.J., & Simons-Morton, B.G. (2010).
prevention programs for alcohol misuse in young people. Cochrane Parent-child communication and substance use among adolescents:
Database of Systematic Reviews, 2011, CD009113. Do father and mother communication play a different role for sons and
Gilgun, J.F. (1995). We shared something special: The moral discourse daughters? Addictive Behaviors, 35, 426–431.
of incest perpetrators. Journal of Marriage and the Family, 57, Maimaris, W., & McCambridge, J. (2013). Age of first drinking
265–281. and adult alcohol problems: Systematic review of prospective
Gilligan, C., Kypri, K., & Lubman, D. (2012). Changing paternal cohort studies. Journal of Epidemiology & Community Health, 68,
behavior to reduce risky drinking among adolescents: 268–274.
Current evidence and future directions. Alcohol and Alcoholism, 47, Mares, S.H., Van Der Vorst, H., Engels, R.C., & Lichtwarck-Aschoff, A.
349–354. (2011). Parental alcohol use, alcohol-related problems, and alcohol-
Griffin, K.W., Scheier, L.M., Botvin, G.J., & Diaz, T. (2000). Ethnic and specific attitudes, alcohol-specific communication, and adolescent
gender differences in psychosocial risk, protection, and adolescent excessive alcohol use and alcohol-related problems: An indirect path
alcohol use. Prevention Science, 1, 199–212. model. Addictive Behaviors, 36, 209–216.
Guttmannova, K., Bailey, J.A., Hill, K.G., Lee, J.O., Hawkins, J.D., Mason, W.A., Hitch, J.E., Kosterman, R., McCarty, C.A.,
Woods, M.L., & Catalano, R.F. (2011). Sensitive periods for Herrenkohl, T.I., & Hawkins, J.D. (2010). Growth in adolescent
adolescent alcohol use initiation: Predicting the lifetime occurrence delinquency and alcohol use in relation to young adult crime, alcohol
and chronicity of alcohol problems in adulthood. Journal of Studies on use disorders, and risky sex: A comparison of youth from low- versus
Alcohol, 72, 221–231. middle-income backgrounds. Journal of Child Psychology and
Hayes, L., Smart, D., Toumbourou, J.W., & Sanson, A. (2004). Parenting Psychiatry, 51, 1377–1385.
Influences on Adolescent Alcohol Use (Research Report No. 10). Maxwell, K.A. (2002). Friends: The role of peer influence across
Retrieved from Australian Institute of Family Studies website: http:// adolescent risk behaviors. Journal of Youth and Adolescence, 31,
www.aifs.gov.au/institute/pubs/ resreport10/parentinginfluences.html. 267–277.
Hox, J.J., & Bechger, T.M. (2001). An introduction to structural equation McBride, N., Farringdon, F., Meuleners, L., & Midford, R. (2006).
modeling. Family Science Review, 11, 354–373. School health and alcohol harm reduction project: Details of
Hu, L., & Bentler, P.M. (1999). Cutoff criteria for fit indexes in intervention development and research procedures. Perth, WA:
covariance structure analysis: Conventional criteria versus new National Drug Research Institute, Curtin University of Technology.
alternatives. Structural Equation Modeling, 6, 1–55. Ministerial Council on Drug Strategy. (2006). National alcohol strategy
Institute for Health Metrics and Evaluation. (2013). The Global Burden 2006–2011. Retrieved from http://www.alcohol.gov.au/internet/
of Disease: Generating Evidence, Guiding Policy. Retrieved from alcohol/publishing.nsf/Content/B83AD1F91AA632ADCA25718E0081
http://www.healthdata.org/policy-report/global-burden-disease-gener- F1C3/$File/nas-06-09.pdf.
ating-evidence-guiding-policy. Mrug, S., & McCay, R. (2013). Parental and peer disapproval of alcohol
Jernigan, D. (2001). Global status report: Alcohol and young people. use and its relationship to adolescent drinking: Age, gender, and racial
Retrieved from World Health Organization website: http:// differences. Psychology of Addictive Behaviors, 27, 604–614.
www.who.int/substance_abuse/publications/alcohol/en/. Nash, S.G., McQueen, A., & Bray, J.H. (2005). Pathways to adolescent
Kaplan, D. (2000). Structural equation modeling: Foundations and alcohol use: Family environment, peer influence and parental
extensions. Thousand Oaks, CA: Sage. expectations. Journal of Adolescent Health, 37, 19–28.
Kelley, S.S., Borawski, E.A., Flocke, S.A., & Keen, K.J. (2003). The role Newton, C., Vogel, L.E., Teesson, M., & Andrews, G. (2009). CLIMATE
of sequential and concurrent sexual relationships in the risk of Schools: Alcohol module: Cross-validation of a school-based preven-
sexually transmitted diseases among adolescents. Journal of tion programme for alcohol misuse. Australian and New Zealand
Adolescent Health, 32, 296–305. Journal of Psychiatry, 43, 201–207.
Kelly, A.B., O’Flaherty, M., Toumbourou, J.W., Connon, J.P., Hemphill, Perkins, H.W. (2003). The emergence and evolution of the social norms
S.A., & Catalano, R. F. (2011). Gender differences in the impact of approach to substance abuse prevention. In Perkins, H.W. (Ed.), The
families on alcohol use: A lagged longitudinal study of early social norms approach to preventing school and college age substance
adolescents. Addiction, 106, 1427–1436. abuse: A handbook for educators, counselors, clinicians (pp. 3–18).
Kelly, K.J., Comello, M.L.G., & Hunn, L.C.P. (2002). Parent-child San Francisco: Jossey-Bass.
communication, perceived sanctions against drug use, and youth drug Pettersson, C., Linden-Boström, M., & Eriksson, C. (2009). Parental
involvement. Adolescence, 37, 775–787. attitudes and behavior concerning adolescent alcohol consumption:
264 C. A. Quinn & K. Bussey Addict Res Theory, 2015; 23(3): 253–264

Do sociodemographic factors matter? Scandinavian Journal of Public Van Der Vorst, H., Engels, R.C., Meeus, W., & Deković, M. (2006). The
Health, 37, 509–517. impact of alcohol specific rules, parental norms about early drinking
Peugh, J.L., & Enders, C.K. (2005). Using the SPSS mixed procedure to and parental alcohol use on adolescents’ drinking behavior. Journal of
fit cross-sectionaland longitudinal multilevel models. Educational and Child Psychology and Psychiatry, 47, 1299–1306.
Psychological Measurement, 65, 717–741. Van Der Vorst, H., Engels, R.C., Meeus, W., Deković, M., & Van
Quinn, C., & Bussey, K. (2012). [Qualitative anticipated social outcome Leeuwe, J. (2005). The role of alcohol-specific socialization in
responses in an underage drinking context]. Unpublished raw data. adolescents’ drinking behavior. Addiction, 100, 1464–1476.
Sargent, M. (1973). Alcoholism as a social problem. Queensland: Vandenberg, R.J., & Lance, C.E. (2000). A review and synthesis of the
University of Queensland Press. measurement invariance literature: Suggestions, practices and recom-
Schafer, J., & Graham, J. (2002). Missing data: Our view of the state of mendations for organizational research. Organizational Research
the art. Psychological Methods, 7, 147–177. Methods, 3, 4–70.
Smith, A.M.A., & Rosenthal, D.A. (1995). Adolescents’ perceptions of Wells, J.E., Horwood, L.J., & Fergusso, D.M. (2004). Drinking patterns
their risk environment. Journal of Adolescence, 18, 229–245. in mid-adolescence and psychosocial outcomes in late adolescence
Song, E-Y., Smiler, A.P., Wagoner, K.G., & Wolfson, M. (2012). and early adulthood. Addiction, 99, 1529–1541.
Everyone says it’s ok: Adolescents’ perceptions of peer, parent, and White, V., & Bariola, E. (2012). Australian secondary school students’
community alcohol norms, alcohol consumption, and alcohol-related use of tobacco, alcohol, and over-the-counter and illicit substances in
consequences. Substance Use & Misuse, 47, 86–98. 2011. Canberra, Australia: Drug Strategy Branch, Australian
Stafström, M. (2007). Kick back and destroy the ride: Alcohol-related Government Department of Health and Ageing.
violence and associations with drinking patterns and delinquency in William, S.K., & Kelly, F.D. (2005). Relationships among involvement,
adolescence. Substance Abuse Treatment, Prevention, and Policy, 2, attachment, and behavioral problems in adolescence: Examining
18. doi:10.1186/1747-597X-2-18. father’s influence. The Journal of Early Adolescence, 25, 168–196.
Teesson, M., Newton, N.C., & Barrett, E.L. (2012). Australian school- Yu, J. (1998). Perceived parental/peer attitudes and alcohol-related
Downloaded by [Universitat de les Illes Balears] at 09:37 10 February 2016

based prevention programs for alcohol and other drugs: A systematic behaviors: An analysis of the impact of the drinking age law.
review. Drug and Alcohol Review, 31, 731–736. Substance Use & Misuse, 33, 2687–2702.