FOR DEBATE doi:10.1111/j.1360-0443.2010.03160.

Alcohol marketing research: the need for a

new agenda add_3160 466..471

Petra S. Meier
School of Health and Related Research (ScHARR), University of Sheffield, Sheffield, UK

ABSTRACT

Aims This paper aims to contribute to a rethink of marketing research priorities to address policy makers’ evidence
needs in relation to alcohol marketing. Method Discussion paper reviewing evidence gaps identified during an
appraisal of policy options to restrict alcohol marketing. Findings Evidence requirements can be categorized as
follows: (i) the size of marketing effects for the whole population and for policy-relevant population subgroups, (ii) the
balance between immediate and long-term effects and the time lag, duration and cumulative build-up of effects and
(iii) comparative effects of partial versus comprehensive marketing restrictions on consumption and harm. These
knowledge gaps impede the appraisal and evaluation of existing and new interventions, because without understand-
ing the size and timing of expected effects, researchers may choose inadequate time-frames, samples or sample sizes. To
date, research has tended to rely on simplified models of marketing and has focused disproportionately on youth
populations. The effects of cumulative exposure across multiple marketing channels, targeting of messages at certain
population groups and indirect effects of advertising on consumption remain unclear. Conclusion It is essential that
studies into marketing effect sizes are geared towards informing policy decision-makers, anchored strongly in theory,
use measures of effect that are well-justified and recognize fully the complexities of alcohol marketing efforts.

Keywords alcohol advertising, alcohol marketing, policy, research needs.

Correspondence to: Petra S. Meier, School of Health and Related Research (ScHARR), University of Sheffield, 30 Regent Street, Sheffield S1 4DA, UK.
E-mail: p.meier@sheffield.ac.uk
Submitted 9 April 2010; initial review completed 21 April 2010; final version accepted 11 August 2010

IDENTIFYING THE EVIDENCE GAPS
Classically, marketing companies use four tools to bring
products and customers’ needs into alignment: product
design, pricing, distribution and promotion. This paper is
primarily concerned with promotion, which covers a
variety of communications with the consumer, including
mass media advertising; product placements; sponsor-
ship; public relations and point of sale display [1]. It also
includes price signals (e.g. discounting); however, price
policies are a slightly different issue not discussed in this
paper. Unlike some other elements of alcohol marketing,
advertising has been researched heavily, with well over
1000 studies available. Longitudinal studies and recent
systematic reviews now contribute considerable support
for causal links [2–5], and have been instrumental in
generating a growing scientific consensus that advertis-
ing in the traditional media influences drinking initia-
tion, levels of consumption and drinking patterns in
young people. This is an important step in refuting
industry arguments that marketing influences brand
share but not total consumption.
With regard to the question of policy action there is, as
often in public health, a tension between economic and
libertarian interests versus the need to protect citizens
from harm. Hence, there is a need to demonstrate the
degree of effectiveness of different policy options with
varying levels of restrictiveness. While scientists are not
always in a position to influence policy decision-making,
in 2008 the author’s team was tasked with appraising
advertising and promotion policies for the UK govern-
ment. The following evidence needs were identified,
specifically:
• Size of marketing effects (i.e. a baseline quantification
of the effect sizes of marketing on the whole popula-
tion, and on policy-relevant population subgroups
including young binge drinkers, heavy and dependent
drinkers trying to reduce consumption) and quantifica-
tion of effects of the range of marketing efforts
employed by industry.

© 2010 The Author, Addiction © 2010 Society for the Study of Addiction Addiction, 106, 466–471

• Timing of effects (i.e. the balance between immediate
and long-term effects, information on time lags, dura-
tion and cumulative build-up of effects).
• Comparative effectiveness of policy options in reducing
alcohol consumption, and associated health and social
harms (i.e. effect sizes for a range of comprehensive and
partial marketing restrictions on content, times of day
or media channels, and including estimates for effects
on population subgroups).
Surprisingly, given the large body of literature on
alcohol marketing, none of this evidence was available.
This paper aims to discuss evidence needs specific to
policy appraisal and evaluation in more detail, to present
first ideas for future directions, and to call on scientists
to develop the necessary methods and research pro-
grammes to take the field forward.
A FOCUS ON EXISTING DRINKERS
Many researchers and politicians have concentrated on
potential harm to young people. There is an implicit
assumption that alcohol marketing only, or at least pre-
dominantly, carries risks for children, presumably
because they lack the cognitive capacities to distinguish
the reality portrayed in adverts from real-life experiences.
Researchers may also prefer youth populations because
drinking onset effects are easier to detect than changes in
established drinkers. This focus needs to be rethought in
light of (i) most marketing being designed to influence
existing consumers and (ii) emerging evidence on the
heuristic processing of marketing messages.
Concerning the first point, only pre-experience market-
ing is targeted at those who have never tried alcohol, or
those who have not tried the type or brand of alcohol in
question. The most common reason is age—young
cohorts have not started drinking at all, and older cohorts
mature into different market segments, e.g. life-style prod-
ucts such as single malt whisky. The aim of pre-experience
marketing is to increase the odds of a first trial.
However, marketing targeted at existing drinkers,
either during or in between drinking experiences, is much
more common. In this case, marketing aims to introduce
new product versions, show new settings or contexts in
which a product can be used, refamiliarize former users,
frame the next purchasing or consumption experience
cognitively or frame outcome expectations and interpre-
tations [6]. Drinkers are targeted frequently during a
drinking situation in order to increase consumption.
Examples are wide-ranging and include ‘happy hours’,
beer festivals, sponsorship of events that involve heavy
drinking, such as the first week at university, sports or
music events and well-timed TV advertising, e.g. wine
advertisements during situation comedies aimed at female
and older viewers. Marketing that prompts consumption
© 2010 The Author, Addiction © 2010 Society for the Study of Addiction
Alcohol marketing research priorities 467
in connection with an activity that the drinker enjoys
seeks to enhance the overall experience of the product.
Cognitive research shows that during-experience market-
ing both reshapes the consumption experience itself and
the memory of it by framing it positively [7,8].
The aim of postexperience marketing is to organize
consumers’ memory using aural and visual cues. Cogni-
tive psychology research shows that memories of adver-
tisements and consumption experiences are not separate
‘replays’ but are encoded with relevant context informa-
tion and reconstructed in a personally relevant way.
Advertising can trigger memories of previous exposure
and, similarly, the consumption experience can cue the
retrieval of advertising memories. Experimental evidence
from taste tests shows that ‘objective’ experiences can be
altered by subsequent advertising exposure, such that the
product is remembered as better than experienced origi-
nally [6]. These types of marketing are all targeted at
existing drinkers, but effects on their consumption have
not been quantified.
Recent neuroscience and cognitive psychology
research also raises concern about the effect of marketing
on heavy drinkers, showing that subconsciously these
take more notice of alcohol-related cues, especially
alcohol imagery, compared to light drinkers and abstain-
ers. This alcohol attention bias has been confirmed for
both dependent and heavy social drinkers (e.g. see Town-
shend & Duka [9]). While, to the author’s knowledge,
research has not yet been extended to marketing commu-
nications, it is plausible that heavier drinkers would also
be more sensitive to alcohol cues in marketing. This could
explain findings that memories of positive drinking out-
comes are more accessible for heavy compared to light
drinkers [10]. Experimental evidence also shows that
alcohol portrayals in films and adverts lead to distinct
patterns of brain activation, causing craving responses
and affecting consumption decisions in heavy drinkers
[11].
With regard to the second point raised, it may not only
be children who lack the logical capacities to fully evalu-
ate the marketing to which they are exposed. Cognitive
theories, and in particular information processing and
decision-making theories, provide a useful explanatory
framework for both short-term and cumulative influ-
ences of marketing on purchasing and consumption
decisions. For example, message interpretation process
and dual processing models hypothesize how marketing
communications are processed, interpreted and internal-
ized by the recipient, while decision-making theories
hypothesize how internalized representations can influ-
ence both conscious (analytical) and subconscious (heu-
ristic) decisions about consumption.
Dual processing theories suggest that all messages—
including marketing messages—are interpreted in two
Addiction, 106, 466–471
468 Petra S. Meier
ways, through affect-based automatic processes and
through logical processes, and that these separate pro-
cesses involve different neural structures [12,13]. The
balance between the two processes depends on whether
or not situational factors (e.g. motivation as well as cog-
nitive capacity) promote full evaluation. Logic-based pro-
cesses require cognitive effort—they are resource-
intensive, slower and involve scepticism and testing of the
realism and credibility of the message. In contrast, affect-
based cognitive processes tend to be largely automatic,
they are unconscious (implicit), fast and impulsive, are
not controlled by conscious desires or intentions and do
not involve extensive cognitive processing of the merits of
the arguments presented. Affect-based processes use heu-
ristic cues such as message length, source attractiveness
and credibility (e.g. celebrity or expert endorsement). We
cannot avoid noticing advertising, and once aware, we
cannot choose to become ‘unaware’. At the same time,
we are unable to invest cognitive resources to evaluate
consciously every advertisement we encounter, and heu-
ristic mechanisms may therefore dominate.
Message interpretation process models hypothesize
that people interpret marketing communications actively
(rather than receive passively). Depth of message inter-
pretation is related to whether the message is perceived as
personally relevant; for example, through identification
(similarity to own life circumstances), emotional engage-
ment with presented storylines or desirability of por-
trayed situations. Personal relevance is a key concept in
the population segmentation approach used in market-
ing, where the focus is often not on product features but
on life-style advertising. Advertising aimed at younger
markets shows drinking in realistic situations involving
fun and socializing outside the home. Marketing targeted
at older men often focuses on mature pleasures of a high-
quality drink, with signals of fraternity, vitality and pros-
perity such as playing golf or fishing. Similarly, beauty,
family relations and romance are used to address female
audiences. Message interpretation process models predict
that with repeat exposure, alcohol-related messages per-
ceived as relevant will become progressively internalized,
so that they are cognitively more available during the
actual decision-making process. Others have referred to
this process as normalization, with alcohol appearing
ubiquitous and a part of normal life.
Decision-making research shows consistently that
people tend to choose only between decision options that
are easy to retrieve (heuristic decision-making). They do
not go through a full analytical process involving calcu-
lations of probabilities and values of the benefits and
costs of each decision option in terms of an expected
utility analysis, unless there are particular situational
aspects promoting analytical decision-making. Heuristic
decision-making is more likely in situations involving
© 2010 The Author, Addiction © 2010 Society for the Study of Addiction
repeat or routine behaviours (e.g. Friday-after-work
drinks), in the presence of peers behaving in a uniform
way (e.g. round-buying), or where decision-makers are
intoxicated. Importantly, ease of retrieval of decision
options is influenced not only by previous encounters
with real objects or events, but also by priming effects,
source credibility, repetitions and recency of exposure.
Studies that have applied cognitive models explicitly in
the context of alcohol marketing report that depth of
message interpretation and the degree of audience con-
nectedness, identification with protagonists and recall of
advertisements and product placements predict alcohol-
related attitudes and consumption decisions in young
people [14–17].
Research also shows that prior decisions influence
future decisions strongly, and this is again understood as
a mental short-cut. If a decision led to an acceptable
outcome, people are much more likely to decide in the
same way the next time they face a similar situation.
Experimental research shows that marketing can reshape
memories of consumption experiences [6–8], and so it
alters the mental representations of previous decision
outcomes. By making previous decision outcomes appear
more favourable, marketing could promote repeat deci-
sions to consume alcohol.
To sum up:
1 Marketing messages may be processed using uncon-
scious affect-based processes rather than slower logic-
based processes.
2 Personally relevant messages may be more likely to be
internalized and cognitively available during decision-
making.
3 Easy cognitive availability of a decision option predicts
decision outcomes in situations where the default posi-
tion is heuristic rather than analytical decision-
making.
4 Because marketing can affect memories of drinking
occasions, and because memories of positive outcomes
promote repeat behaviour, marketing may play a
special role in driving repeat consumption.
Thus, while research on young people should con-
tinue, it is important that research also establishes to
what degree marketing reinforces consumption among
existing drinkers, and whether it hampers attempts to
drink in moderation.
RESEARCH DESIGNS OUGHT TO
REFLECT MARKETING COMPLEXITY
Industries spend significant resources on market research
to inform coherent, multi-component campaigns that are
targeted expertly at relevant population segments using
data on demographic/life-style characteristics and drink-
ing preferences. In contrast, academic research has
Addiction, 106, 466–471

struggled to design quantitative measures taking full
account of the breadth and diversity of marketing behav-
iour, and much of the complexity is so far missing from
the research evidence base. For example, the most
common measures of advertising effects are broad indi-
cators such as estimates of changes of the volume of
overall media exposure (e.g. hours of television viewing),
the volume of advertising exposure (e.g. number of TV
advertisements) or relatively small variations observed in
marketing expenditure over time. These are then related
to (often equally aggregated) per capita consumption or
alcohol sales indicators. Such models of advertising work
from the central tenet that if advertising creates demand,
then more advertising should create more demand.
Aggregate-level studies of this type tend to report very
small effect sizes, due partly to the fact that at least in
developed countries there is only limited variability over
time in the levels of marketing relative to the total mar-
keting effort in developed countries. However, if we
assume that targeting of messages is important and that
exposure to messages that are not pertinent to us does not
affect us (or not as much), then small effect sizes could
also be caused by inadequate study designs. As an
example for illustration: a new type of product is mar-
keted to affluent young women and this explains a signifi-
cant share of the growth in marketing expenditure in a
given period. The campaign is successful and leads to sub-
stantial increases in this group’s alcohol consumption.
Any effect would hardly be discernible in aggregate-level
analyses, as affluent young women may make up only
around 2% of the adult population, and studies that do
not differentiate between targeted and non-targeted
populations would be expected to produce patterns of
inconsistent and non-significant effects.
RESEARCH NEEDS TO CLARIFY THE
TIMING OF EFFECTS
An important uncertainty concerns the timing of mar-
keting effects, assumptions about which are required
when estimating policy impact: how soon after imple-
mentation of a policy should researchers expect con-
sumption effects—immediately, or possibly only once an
unexposed generation has grown up? So far, data are
limited.
There is empirical support for direct, immediate effects
of marketing on consumption. What is meant is the
increased likelihood of opening a beer from the fridge in
response to watching a beer advertisement, or the degree
to which a promotion leads to impulse purchases that
would not otherwise have been made. A suggested
mechanism for immediate effects is that alcohol portray-
als act as cues for imitative behaviour and/or prompt
craving [18,19]. For example, a Dutch team showed in an
© 2010 The Author, Addiction © 2010 Society for the Study of Addiction
Alcohol marketing research priorities 469
experiment that young men watching movies in which
protagonists drank frequently, or commercial breaks with
alcohol advertising, drank substantially more alcohol
during and immediately after a TV-watching episode than
controls watching movies with infrequent drinking or
non-alcohol advertisements [18]. Thus, portrayals of
actual drinking behaviours, whether via product place-
ments or advertisements, appear to influence drinking
levels directly. This study should be replicated in other
population groups and using other marketing channels.
Future studies could vary (or at least control for) the level
of identification with the drinking protagonists, and
attempt quantification of effect sizes through variation of
exposure levels.
There are, however, also likely to be longer-term influ-
ences via individuals’ drinking-related affective and/or
cognitive responses (e.g. decision-making propensities,
approach–avoidance behaviours, social norms). Longitu-
dinal studies have begun to provide first estimates of
cumulative advertising effect sizes for young people, fol-
lowing children and young adults for up to 8 years [2–5].
Young people are more likely to continue to increase their
drinking into their 20s in markets with greater overall
exposure to alcohol advertising than in markets with less
exposure [2]. It is important that such cohorts are fol-
lowed into adulthood, with careful recording of market-
ing and policy changes. The feasibility of using existing
cohort data to learn more about individual subgroups
such as advertising ‘non-responders’ needs to be explored
(i.e. those who, despite high exposure to marketing, did
not start drinking heavily), as this could generate hypoth-
eses about the relevance of marketing messages to differ-
ent groups.
POSSIBLE DIRECTIONS FOR
FUTURE RESEARCH
A key argument of this paper is that it is unreasonable to
expect observable effects in people who are neither tar-
geted nor otherwise perceive marketing messages as
relevant, and so traditional marketing measures are
unlikely to provide much-needed answers. Studies need
to be designed such that, while ideally covering the whole
population and the breadth of marketing communica-
tions, they allow levels of disaggregation of exposure and
measured effects similar to those used by marketing
industries.
New indicators should combine measures of market-
ing volume and receptivity, i.e. individual-level measures
of exposure to relevant marketing. These are necessary for
the important first step of quantifying baseline levels of
exposure to the full range of alcohol marketing, against
which postintervention data can then be compared.
Survey methods are unlikely to result in convincing
Addiction, 106, 466–471
470 Petra S. Meier
estimates, as it is difficult for most people to remember
how many internet alcohol advertisements they saw, or
how often they noticed a promotional banner during a
televised sport event. Field study approaches using direct
observation may be worth exploration. For example,
researchers could follow volunteers for a day and record
systematically (e.g. using visual and audio footage) expo-
sure to alcohol-related marketing communications,
whether seen while commuting, browsing the internet,
watching television or in shops. Such data could provide
information about marketing mix, possibly providing a
quantitative breakdown of intensity of exposure to differ-
ent channels, and allow detailed analysis of the types of
messages targeted at different population segments. If it
were possible to undertake such observations over time
and/or with large samples and to record concurrent
purchasing behaviours, such research might provide a
powerful tool for understanding how marketing mix
influences purchasing in different populations. It is
acknowledged that there are significant financial, and in
the case of field observations also ethical, obstacles to
overcome and that it may be overly optimistic to hope that
scientific marketing research will be able to compete with
the extremely well-funded marketing research carried
out by industries. One solution could be a new tax on all
alcohol marketing expenditure, including industry mar-
keting research, with some of this money earmarked for
targeted research in this field. There may also be ways to
compel industry players to share their routinely collected
sales and promotion data with policy makers and
researchers.
To understand further the effects of message target-
ing, it may be possible to examine historical changes in
advertising portrayals and target markets; for example,
whether increased marketing to affluent young women is
matched by consumption changes in this group that are
not observed in other groups. Where good-quality, fine-
grained consumption and advertising data are available
or can be constructed from media archives, time–series
analyses to derive effect size estimates may be possible.
Moreover, a crucial reverse demonstration could be
attempted—that marketing not perceived as personally
relevant is not recalled as well and does not affect con-
sumption. Comparative research with samples of people
who are exposed similarly to marketing, but for whom
marketing messages are likely to feel less relevant, could
be informative. Alternatively, retrospective cohort designs
could be used to investigate personal characteristics that
predict differential responses to marketing, comparing
those with similar levels of marketing exposure but dif-
ferent drinking outcomes.
If heavy drinkers show cognitive changes that
promote attention to alcohol-related stimuli, then mar-
keting exposure could have a disproportionate effect on
© 2010 The Author, Addiction © 2010 Society for the Study of Addiction
their consumption and interfere with attempts to cut
down. So far, studies on alcohol attention bias have taken
a clinical experimental perspective to understand how
alcohol stimuli influence clinical outcomes in dependent
drinkers. Testing whether alcohol attention bias can be
confirmed using actual marketing communications is an
important next step.
Finally, radical policy changes are rare, thus it is
crucial that the international research community is
ready to conduct evaluations when they do occur. For
example, when the French Loi Evin (a law restricting
alcohol advertising) was introduced in 1991, no appro-
priate baseline measurement for pre-policy advertising
appears to have been carried out and it has been difficult
to control for underlying consumption trends. Baseline
data might have allowed researchers to evaluate the
effects on advertising and consumption, to investigate
possible counterveiling industry responses and to under-
stand more clearly the timing of long-term effects.
CONCLUSION
To date, studies have tended to rely on simplified models of
marketing and have focused disproportionately on youth
populations. The cumulative effects of exposure across
multiple marketing channels, targeting of messages at
certain population groups and indirect effects of advertis-
ing on consumption have rarely been considered. The
identified knowledge gaps impede the appraisal and
evaluation of policy interventions, because without
understanding the size and timing of expected effects,
researchers may choose inadequate time-frames or
sample sizes. The lack of understanding of how different
population groups respond to marketing and to market-
ing restrictions prevents us from devising targeted inter-
ventions and hampers scientists’ ability to inform policy
decision-making. It is essential that future research into
marketing effects is anchored strongly in theory, that
measures of effect are well justified and that the complexi-
ties of alcohol marketing efforts are recognized fully.
Declaration of interests
None.
Acknowledgements
Thanks go to my collaborators on the Independent
Review of Alcohol Pricing, especially Andrew Booth,
Alan Brennan and Robin Purshouse. Gerard Hastings,
Lucy Gell and Sabrina Schulte provided valuable com-
ments on an early draft of the manuscript. The author’s
salary is paid by the University of Sheffield and no sepa-
rate funding was obtained for the preparation of this
manuscript.
Addiction, 106, 466–471

References
1. Hastings G., Brooks O., Stead M., Angus K., Anker T., Farrell
T. Failure of self regulation of UK alcohol advertising. BMJ
2010; 340: b5650.
2. Anderson P., de Bruijn A., Angus K., Gordon R., Hastings G.
Impact of alcohol advertising and media exposure on ado-
lescent alcohol use: a systematic review of longitudinal
studies. Alcohol Alcohol 2009; 44: 229–43.
3. Booth A., Meier T., Stockwell T., Sutton A., Wilkinson A.,
Wong R. et al. 2008. Independent review of the effects of
alcohol pricing and promotion. Part A: Systematic reviews.
Project report for the Department of Health. Available at:
www.dh.gov.uk/prod_consum_dh/groups/
dh_digitalassets/documents/digitalasset/dh_091366.pdf
(accessed 17 September 2010). Archived by WebCite® at
http://www.webcitation.org/5qmVZSYcc.
4. Science Group of the European Alcohol and Health Forum.
2009. Does marketing communication impact on the volume
and patterns of consumption of alcoholic beverages, especially by
young people? A review of longitudinal studies. Available at:
http://ec.europa.eu/health/ph_determinants/life_style/
alcohol/Forum/docs/science_o01_en.pdf (accessed 17
September 2010). Archived by WebCite® at http://
www.webcitation.org/5qmVkRnkg.
5. Smith L. A., Foxcroft D. R. The effect of alcohol advertising,
marketing and portrayal on drinking behaviour in young
people: systematic review of prospective cohort studies.
BMC Public Health 2009; 9: 51.
6. Braun-LaTour K. A., LaTour M. S., Pickrell J. E., Loftus E. F.
How and when advertising can influence memory for
consumer experience. J Advertising 2004; 33: 7–25. Avail-
able from http://www.jstor.org/pss/4189273 (accessed 1
October 2010).
7. Allison R., Uhl K. Influence of beer brand identification on
taste perception. J Market Res 1964; 1: 36–9.
8. Vakratsas D., Ambler T. How advertising works: what do we
really know? J Market 1999; 63: 26–43.
© 2010 The Author, Addiction © 2010 Society for the Study of Addiction
Alcohol marketing research priorities 471
9. Townshend J. M., Duka T. Attentional bias associated with
alcohol cues: differences between heavy and occasional
social drinkers. Psychopharmacology 2001; 157: 67–74.
10. McCusker C. G. Cognitive biases and addiction: an evolution
in theory and method. Addiction 2001; 96: 47–56.
11. Tapert S. F., Cheung E. H., Brown G. G., Frank L. R., Paulus
M. P., Schweinsburg A. D. et al. Neural response to alcohol
stimuli in adolescents with alcohol use disorder. Arch Gen
Psychiatry 2003; 60: 727–35.
12. Bechara A. Decision making, impulse control and loss of
willpower to resist drugs: a neurocognitive perspective. Nat
Neurosci 2005; 8: 1458–63.
13. Yin H. H., Knowlton B. J. The role of the basal ganglia in
habit formation. Nat Rev Neurosci 2006; 7: 464–76.
14. Austin E., Chen M., Grube J. How does alcohol advertising
influence underage drinking? The role of desirability, iden-
tification and scepticism. J Adolesc Health 2006; 38: 376–
84.
15. Stacy A. W., Zogg J. B., Unger J. B., Deut C. W. Exposure to
advertised alcohol ads and subsequent adolescent alcohol
use. Am J Health Behav 2004; 28: 498–509.
16. Ellickson P. L., Collins R. L., Hambarsoomians K., McCaffrey
D. F. Does alcohol advertising promote adolescent drinking:
results from a longitudinal assessment. Addiction 2005;
100: 235–46.
17. Russell C. A., Russell D. Nature and impact of alcohol mes-
sages in a youth-oriented television series. J Advertising
2009; 38: 97–111.
18. Engels R. C. M. E., Hermans R., van Baaren R. B.,
Hollenstein T., Bot S. M. Alcohol portrayal on television
affects actual drinking behaviour. Alcohol Alcohol 2009;
44: 244–9.
19. Larson H., Engels R. C., Granic I., Overbeek G. An experi-
mental study on imitation of alcohol consumption in same-
sex dyads. Alcohol Alcohol 2009; 44: 250–5.
Addiction, 106, 466–471
This document is a scanned copy of a printed document. No warranty is given about the accuracy of the copy.
Users should refer to the original published version of the material.