One Size (Never) Fits All: Segment Differences Observed
Following a School-Based Alcohol Social Marketing
Program
Author
Dietrich, Timo, Rundle-Thiele, Sharyn, Leo, Cheryl, Connor, Jason
Published
2015
Journal Title
Journal of School Health
Version
Accepted Manuscript (AM)
DOI
https://doi.org/10.1111/josh.12244
Copyright Statement
© 2015 American School Health Association. This is the peer reviewed version of the following
article: One Size (Never) Fits All: Segment Differences Observed Following a School-Based
Alcohol Social Marketing Program, Journal of School Health, Volume 85, Issue 4, April 2015,
Pages 251–259, which has been published in final form at 10.1111/josh.12244. This article may
be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Self-
Archiving (http://olabout.wiley.com/WileyCDA/Section/id-828039.html)
Downloaded from
http://hdl.handle.net/10072/141069
Griffith Research Online
https://research-repository.griffith.edu.au
Author postprint of Dietrich, T., Rundle-Thiele, S., Leo, C., & Connor, J. (2015). Please refer to
published version for tables and figures mentioned in the text. http://dx.doi.org/10.1111/josh.12244
1
One Size (Never) Fits All: Segment Differences Observed Following a School-Based Alcohol Social
Marketing Program
Timo Dietrich,Sharyn Rundle-Thiele, Cheryl Leo, Jason Connor
The authors would like to express their gratitude to all 3 Journal of School Health reviewers for their
feedback which has assisted to greatly improve this work. We also thank the support of the
students, parents, teachers and support staff employed at the schools. Finally, we thank the
Queensland Catholic Education Commission and Griffith University for their financial support
throughout this project. The funders played no role in study design, collection, analysis,
interpretation of data, or the decision to submit the paper for publication. They accept no
responsibility for contents.
ABSTRACT
BACKGROUND
According to commercial marketing theory, a market orientation leads to improved performance.
Drawing on the social marketing principles of segmentation and audience research, the current
study seeks to identify segments to examine responses to a school-based alcohol social marketing
program.
METHODS
A sample of 371 year 10 students (aged: 14–16 years; 51.4% boys) participated in a prospective (pre-
post) multisite alcohol social marketing program. Game On: Know Alcohol(GO:KA) program included
6, student-centered, and interactive lessons to teach adolescents about alcohol and strategies to
abstain or moderate drinking. A repeated measures design was used. Baseline demographics,
drinking attitudes, drinking intentions, and alcohol knowledge were cluster analyzed to identify
segments. Change on key program outcome measures and satisfaction with program components
were assessed by segment.
RESULTS
Three segments were identified; (1) Skeptics, (2) Risky Males, (3) Good Females. Segments 2 and 3
showed greatest change in drinking attitudes and intentions. Good Femalesreported highest
satisfaction with all program components and Skeptics lowest program satisfaction with all program
components.
CONCLUSION
Three segments, each differing on psychographic and demographic variables, exhibited different
change patterns following participation in GO:KA. Post hoc analysis identified that satisfaction with
program components differed by segment offering opportunities for further research.
Author postprint of Dietrich, T., Rundle-Thiele, S., Leo, C., & Connor, J. (2015). Please refer to
published version for tables and figures mentioned in the text. http://dx.doi.org/10.1111/josh.12244
2
Alcohol consumption is an intrinsic part of the youth culture in industrialized countries and is
considered integral to social interactions and celebratory events.[1, 2] Research indicates that youth
often engage in “drinking” with the primary intention of becoming so inebriated as to experience a
loss of control.[3] One in 5 Australians over the age of 14 years drinks at risky levels at least once a
month[4] – defined as 6 or more standard drinks (10 g of alcohol) per occasion for males and 5 or
more for females. One in 3 Australian youths directly experience alcohol-related violence by
adulthood.[4] In the 15 to 34 year age group, alcohol causes more deaths and hospitalizations than
tobacco and all illicit drugs combined.[5] Overall, alcohol-related social costs in Australia are
estimated at Australian dollar (AUD) 15 billion per annum.[6] Public health initiatives addressing
youth alcohol consumption have been based on the principle that prevention is more cost-effective
than treatment.[7] There is an urgent need to address youth drinking cultures and school-based
delivery offers an attractive and efficient means to reach a large number of adolescents.[8, 9]
School-based alcohol education programs have produced mixed results, with many showing modest
effect sizes in the short-term and poor long-term outcomes.[9, 10] Efforts to improve long-term
effectiveness have involved education programs being expanded to include several components
delivered across multiple settings (communities, schools, and parents), and the incorporation of
regular, follow-up “booster” sessions.[11] It is not yet clear if these resource intensive programs
offer improved outcomes[12] and further research is required to disentangle more precise change
mechanisms.[9]
A limitation of many alcohol and drug education programs is that they are designed for a general
population using a 1-size-fits-all approach.[13] A 1-size-fits-all approach may limit program
effectiveness as large numbers of the audience may be left dissatisfied, uninterested, or
unchallenged.[14] Studies dating back to the early 1990s have indicated differential program
effects[15] which challenged the effectiveness of 1-size-fits-all approaches.[16] Understanding
differential success rates in education programs requires further research into the key components
of target behaviors and their change mechanisms. A potential avenue for further investigation is
identifying variation in program responses by student segments.
Social marketing is the adaptation of commercial marketing principles for social good and social
marketing has been effective in addressing a broad range of issues in diverse areas including (but not
limited to) breastfeeding, road safety, saving, recycling, and safer sex.[17-21] In contrast to allied
health behavior change areas, social marketing is audience-centered and competitively minded as
noted in the National Social Marketing Centre's benchmark criteria for designing social marketing
programs.[22] Programs are developed based on audience research and competitor analysis
ensuring social marketing gains necessary insights to meet the needs and wants of the target
audience. Further, target audience feedback is used to inform program improvement.[23] In short,
social marketing starts and ends with the target audience. According to commercial marketing
theory this target audience or market orientation leads to improved performance (ie, greater levels
of behavior change).[24, 25] Despite success in allied health behavior change areas, social marketing
has received limited attention in alcohol education programs in school settings. The effectiveness of
social marketing's application in an alcohol program in a school setting has been documented by
Rundle-Thiele et al.[26] Despite success in allied health behavior change areas, social marketing has
received only limited attention in the context of alcohol interventions delivered in school settings.
Author postprint of Dietrich, T., Rundle-Thiele, S., Leo, C., & Connor, J. (2015). Please refer to
published version for tables and figures mentioned in the text. http://dx.doi.org/10.1111/josh.12244
3
An effective strategy and key principle[22] in social marketing is market “segmentation” which is a
core commercial marketing strategy.[27] A segmentation study aims to identify, often via cluster
analysis methods, groups of individuals who are likely to exhibit similar needs, preferences,
motivations, and/or behaviors. According to Kotler[28] a segmentation study can comprise up to 4
bases: demographic, psychographic, geographic, and behavioral, to identify meaningful segments.
Rundle-Thiele et al[29] indicate that whereas there has been mixed used of demographic,
behavioral, psychographic, and geographic segmentation bases, the majority of studies use 2 or
more segmentation bases. Once different segments are identified, a social marketer targets the
group or groups (target marketing) that promise most behavior change potential.[21] In achieving
different profiles of adolescents, it is possible that more effective programs can be tailored catering
to each segment's needs and wants to build on the modest effect sizes of existing, generic
approaches to school-based alcohol education programs.
It is proposed that school-based alcohol education programs tailored to the characteristics of
targeted adolescents may be more effective than generic approaches. Against this background, the
purpose of the current study was to identify if segments (groups) were evident at baseline on key
factors across 2 segmentation bases, and whether the segments differ in terms of response to the
program and satisfaction with measured program components. Interestingly, whereas some
programs have begun to focus on ethnic,[30] high risk,[31-33] and sex,[34] to date, studies have not
implemented a complete segmentation process to identify homogenous groups of adolescents
within their study population.
In contrast to earlier studies, this research adopts key social marketing benchmark criteria of
segmentation and audience research to ascertain whether unique groups exist and how these
groups differ in their response to and satisfaction with a social marketing alcohol education
program. This research assessed whether segments within the target population of 14- to 16-year-
old adolescents were evident at baseline using 2 segmentation bases (psychographic and
demographic). As part of the Game On: Know Alcohol (GO:KA) pilot study, demographics (age and
sex) and psychographics (knowledge about alcohol), attitudes toward binge drinking (defined as
more than 6 standard drinks in any one day), and intentions toward moderate drinking (defined as
more than 2 standard drinks in any 1 day) were selected to evaluate pilot study effectiveness.
Alcohol knowledge, attitudes toward binge drinking, and intentions toward no or moderate drinking
are key modifiable constructs included in effective alcohol education programs.[9, 35, 36]
Social marketing, with its audience-oriented focus, provides a bottom-up research philosophy where
insights gained from the target audience can be used to inform program improvement.[37] On the
basis that satisfaction leads to loyalty (repeated behavior) in commercial marketing,[38] this study
also undertook an examination of student satisfaction with program components to gain segment-
based audience insights. The main purpose of the current study is to identify if segments (groups)
are evident at baseline and whether the segments differ in terms of response to the program and
satisfaction with the program components that they experienced during GO:KA. The analysis
performed seeks to gain additional audience insight for subsequent program development.
METHODS
Author postprint of Dietrich, T., Rundle-Thiele, S., Leo, C., & Connor, J. (2015). Please refer to
published version for tables and figures mentioned in the text. http://dx.doi.org/10.1111/josh.12244
4
Participants
Three out of 4 schools approached participated in the GO:KA pilot study, following active informed
parental and student consent. These schools were selected on the basis of geographical
convenience, and were within a broader catchment of 92 Catholic schools in Queensland, Australia.
A total of 382 students (mean age: 14.77) from inner-city metropolitan Catholic high schools took
part in the pre-survey. A total of 343 students (183 males and 160 females) completed the post-
survey, representing a 70.7% retention rate.
The Game On: Know Alcohol Program
The GO:KA program was designed by social marketing researchers using 8 social marketing
benchmark criteria.[22] These criteria include behavior change, customer orientation, theory,
insight, competition, exchange, segmentation, and methods mix. A complete description outlining
how GO:KA was based on the 8 social marketing benchmark criteria is available in Rundle-Thiele et
al.[26] GO:KA seeks to increase knowledge relating to alcohol, change drinking attitudes, and
decrease intentions to drink alcohol.[26] Starting with formative and competitive research to
understand student and educator preferences for alcohol education, GO:KA was developed to
provide an interactive student-centered program involving both online and offline activities to
maximize learning. The GO:KA program featured a combination of online games and practical
activities in a 6-lesson program targeting 14- to 16-year-old adolescents. A total of 9 activities, 4
online activities (3 games and 1 quiz), and 5 experiential activities underpinned the pilot study
design. The specific aims of the program were to (1) illustrate the effects of alcohol; (2) promote an
understanding of units of alcohol measurement (standard drinks); and (3) provide strategies to drink
alcohol in moderation and improve drinking refusal self-efficacy skills.
Instruments
An online repeated measures survey was completed by all participants prior to participating in the
GO:KA program and post-delivery. Pre- and post-survey responses were on average 3 weeks apart
with the GO:KA program taking place immediately after the first survey. The survey included
demographic items of age and sex, and psychographic measures designed to capture knowledge
about alcohol, attitudes toward binge drinking (defined as more than 6 standard drinks in any 1 day),
and intentions toward moderate drinking (defined as 2 standard drinks in any 1 day). Attitude
measures toward binge drinking were adapted from prior research[35, 39] and consisted of 5 bipolar
items rated on a 7-point bipolar rating scale (α: 0.90) with −3 indicating a negative attitude toward
binge drinking and +3 indicating a positive attitude. Behavioral intention measures were adapted
from previous research.[35, 39] A total of 3 bipolar items were used to measure students' intentions
toward moderate drinking (2 standard drinks) on a 7-point rating scale (α = 0.83) with −3 indicating
negative intentions and +3 positive intentions. An alcohol knowledge index was created for this
study following procedures used in prior studies.[35, 40] Alcohol knowledge was tested asking
students 10 questions relating to low-risk alcohol consumption levels (6 items) and drink driving
questions (2 items) and number of standard drinks in different alcoholic beverages (2 items).
Answers were awarded with a score of 1 if the respondent gave an accurate answer. An overall
knowledge score (maximum 10) was then calculated. Students' satisfaction with the GO:KA program