You are on page 1of 9

Journal of Consumer Behaviour, J. Consumer Behav.

10: 152–160 (2011)


Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/cb.359

Can advergames boost children’s healthier eating habits? A comparison


between healthy and non‐healthy food
MÓNICA DIAS* and LUÍSA AGANTE
NOVA School of Business and Economics, Campus de Campolide, 1099 - 032 Lisbon, Portugal

ABSTRACT

The purpose of this paper is to provide a contribution towards increasing knowledge on how healthy food stimuli can improve eating behaviour,
by making a comparison between advergames’ food content effects on children’s food choices, product liking and nutritional knowledge.
A sample of 231 elementary‐school‐age children (7 and 8 years old) was randomly assigned to one of the following tasks: (i) playing the
healthy advergame and (ii) playing the less healthy advergame. A post‐treatment pictured questionnaire was used to assess children’s
behaviour in terms of immediate food choice, food liking and nutritional knowledge, following exposure to treatment. Results show that
children tend to choose according to what was being advertised in the game. In terms of food liking, children who played the less healthy
version of the game report a higher preference for some of the less healthy options. Regarding children’s nutritional knowledge, no
differences between groups were registered, indicating that children already have a solid understanding of what are ‘good’ and ‘bad’ foods
for their health. Our findings are aligned with previous research on dual‐process model: although children’s nutritional knowledge is good,
they might not employ it when selecting snacks. According to these results, legal, educational and management contributions should be
employed: regulation should protect children from less healthy contents on advergames; schools may use digital games to promote
healthier eating habits; enterprises can invest on this market opportunity (healthy food targeting children); and finally, social marketing can
use advergames in order to reduce child obesity.
Copyright © 2011 John Wiley & Sons, Ltd.

INTRODUCTION WHY TARGET CHILDREN?

Obesity is now considered the fifth leading global risk factor Psychology has been witnessing the birth of several
contributing to mortality and has reached an epidemic status. cognition stage perspectives based on social perspective
According to the World Health Organisation, ‘Childhood (Selman, 1980), information processing perspective (Roedder,
obesity is one of the most serious public health challenges of the 1981) and, more recently, mind theories (Moses and
21st century’, with 43 million children under 5 years old now Baldwin, 2005). This study targets elementary‐school‐age
considered overweight (WHO, 2010). This is a matter of great children in the concrete operational stage (7–8 years old)
concern, especially because these health‐related behaviours, of Piaget’s Cognitive Development Theory (Piaget and
developed during childhood, influence children’s quality of life Inhelder, 1972). Despite conceptual differences, it is generally
(Williams et al., 2005) and are likely to persist in adulthood accepted that children around 7–8 years old start to solidify
(McGinnis et al., 2006). The food industry has contributed to their analytical, logical and rational judgment (Piaget and
childhood obesity through the promotion and strong commu- Inhelder, 1972; Roedder, 1981), being able to retain and
nication campaigns of energy‐dense food directed to children process information from a more ‘self‐reflective’ perspective
(Story and French, 2004; Hastings et al., 2006; Lobstein, (Selman, 1980). Children’s nutritional knowledge also depends
20101). Moreover, a food content analysis of popular children on their cognitive development stage (Lytle et al., 1997; Hart
websites shows that the most marketed food products are et al., 2002). Although parents and schools educate young
candy, cereal, quick‐serve restaurant meals and snacks (Alvy consumers on their eating habits, advertising contributes
and Calvert, 2008). Consequently, children’s food decisions largely to their nutritional beliefs (Singleton and Rhoads,
often take place in high‐calorie or nutrient‐poor environments. 1984; Hastings et al., 2006; Contento, 2008). Edwards and
Yet, we have little empirical understanding about how children Hartwell (2002:373) concluded that 8‐ to 11‐year‐old children
make decisions in these environments or how healthy food have ‘an appreciation of the term healthy eating and could
stimuli might improve children’s eating behaviour. The relate this to what they should be consuming’. Dietary
purpose of this paper is to make a comparison between awareness of children is considered to be good because they
advergames’ food content effects on children’s food choices, seem to be able to distinguish healthy (good) from non‐healthy
product liking and nutritional knowledge. (bad) food (Wellman and Johnson, 1982; Lytle et al., 1997;
Noble et al., 2000).
*Correspondence to: Mónica Dias, NOVA School of Business and Computer games are one of the ‘megahits’ for this age
Economics, Campus de Campolide, 1099‐032 Lisbon, Portugal. segment. Because of their cognitive abilities, children look for
E‐mail: monica.srdias@gmail.com more engaging, stimulating and interactive games that can
provide different levels of challenge and a ‘great deal of
1
At the 11th International Congress on Obesity in Stockholm in 2010, visual and auditory stimulation’ (Acuff and Reiher,
Lobstein stated that ‘Heavy marketing of energy dense foods and the
promotion of fast food outlets is a likely risk factor for obesity and children 1997:88). Today’s kids are characterised as the socially
are a prime target’. empowered generation, with Wi‐Fi, MySpace and YouTube.

Copyright © 2011 John Wiley & Sons, Ltd.


Advergames: Boosting children’s eating habits 153

They are technologically literate and have been shaped to are generally the central feature of the game, engaging the
multitasking (Bardhi et al., 2010). In such technological child in a fun and playful environment. In fact, advergames
environment, it is essential to understand how media are able to deliver subtle associations and experiences
interacts with young people’s lives in order to develop (ratherthan a single factual message) and would therefore
efficient and responsible marketing campaigns (Tufte and persuade children to adapt a certain behaviour (Nairn and
Rasmussen, 2005, 2011). When addressing the fairness of Fine, 2008).
advertising to children, no consensus regarding the ‘magic Culp (2010) analysed the content of all websites
age’ in which a child is able to understand its selling advertised on two children’s TV networks, and games were
purposes was found. Nevertheless, there is a general found on 81 per cent of the websites. The logo was the most
agreement that children do not possess sufficient cognitive used brand identifier, and only one in every 45 exposures to
defences against implicit persuasion from contemporary brand identifiers contained a ‘healthy’ message.
marketing techniques (Nairn and Fine, 2008) even though The food industry is among the top three industries using
this is considered the time when they start developing their advergames as part of their promotional strategy (Lee and
‘moral sense’ perceiving things as right or wrong, good or Youn, 2008). Seventy‐three per cent of the major food
bad (Acuff and Reiher, 1997). advertisers’ websites analysed in 2005 included advergames
(Moore, 2006). Still, as one of the most common categories
targeting children through advergaming, the food promoted
DIGITAL MEDIA is usually energy dense and nutrient poor (Story and French,
2004; Hastings et al., 2006; Alvy and Calvert, 2008).
The emergence and dynamics of new types of entertainment, Research on advergames has focused on its efficacy in
especially on the Web, are changing children’s ‘playground’ delivering a promotional message, building brand equity or
and how they spend their time. Livingstone and Haddon formulating attitudes towards brands or towards the game
(2009) reported that 75 per cent of European children itself (Webber et al., 2006; Winkler and Buckner, 2006;
between 6 and 17 years old use the Internet for various Mallinckrodt and Mizerski, 2007; Hernandez and Chapa,
purposes. This figure is increasing every year, especially 2010). Despite the increasing popularity of advergames in
among younger children (6 –10 years old). In Portugal, the the food industry (Moore, 2006; Culp, 2010), we know
percentage of children using the Internet increased from 54 relatively little about the effects of food advergaming on
per cent in 2005 to 68 per cent in 2008, with more than half children’s healthier eating behaviour and learning after
of these children, between 6 and 10 years old (54%), having playing the game.
access to the Internet.
McCrindle (2006) has argued that experimental, interac-
tive and visual communication is a more attractive and BUILDING EATING BEHAVIOURS
efficient approach towards today’s multimedia literate AND PREFERENCES
generation. To stay competitive, companies have been
adapting their marketing strategies to this new environment. Repeated exposure to marketing campaigns has an impact on
It is estimated that 98 per cent of websites designed for children’s food choice and preferences, because it builds
children permit advertising and that more than two‐thirds rely awareness, reinforces food familiarity, reduces neophobia (fear
on advertising as their primary revenue stream (Neuborne, of trying new foods) and enhances consumption (Sullivan and
2001). This communication channel became very attractive Birch, 1994; Contento, 2008). Most literature addresses the
and has been increasing over the last few years, especially impact of TV messages advertising food and beverages on
Internet marketing techniques targeting children and adoles- children’s dietary and eating behaviours (McGinnis et al.,
cents (Tufte and Rasmussen, 2005; Webber et al., 2006). 2006). The findings suggest that children’s food choices and
Furthermore, in terms of format and content, advertisements preferences reflect their TV exposure experience (Goldberg
are evolving from a factual message format to more implicit et al., 1978; Galst, 1980; Gorn and Goldberg, 1982; Coon et al.,
techniques such as viral marketing, SMS texting, celebrity or 2001). However, the fun and entertainment environment of the
character endorsement, product placement in films, TV advergame plays an important role regarding food familiarity
shows, games and social networks (Tufte and Rasmussen, and product liking once a child is attracted by the fun dimension
2005; Nairn and Fine, 2008). These new formats are able to of a product (Mathiot, 2010). In the case of cereal websites for
capture children’s attention for several minutes compared instance, children are manipulated by brands through online
with 30‐second TV commercials (Moore, 2006; Lee et al., marketing campaigns that increase brand knowledge through
2009), thus increasing product and brand exposure. multiple product/brand interaction, rewards for consuming/
manipulating the desired food item or for mastering the game
(Thomson, 2010).
ADVERGAME Mallinckrodt and Mizerski (2007) studied the effects of
advergames on young consumers’ brand perceptions,
This new form of advertising, where the brand entertains its preferences and requests of a branded cereal breakfast. They
audience, consists of playing ‘online games designed for the were able to show that although children perceived fresh
specific purpose of marketing a single brand or product’ fruit as being healthier than the cereal advertised (Froot
(Winkler and Buckner, 2006:24). The product and the brand Loops), they still planned to request the cereal after playing

Copyright © 2011 John Wiley & Sons, Ltd. J. Consumer Behav. 10: 152– 160 (2011)
DOI: 10.1002/cb
154 M. Dias and L. Agante

the advergame. More recent studies confirmed that children a) The group who played the healthy version will like
choose significantly more snacks that were being advertised healthy food more than the other group.
on advergames compared with others that were not b) The group who played the non‐healthy version will
advertised (Pempek and Calvert, 2009; Hernandez and like non‐healthy food more than the other group.
Chapa, 2010). Wise et al. (2008) concluded that advergames
H3: Children who played the healthy version will exhibit
are more effective when they are associated with the brand or
greater nutritional knowledge, when compared with
product being advertised, engaging players in activities
children who played a less healthy version of the same
related to a behaviour they would display when using the
game.
product. This enhances implicit associations between game
content and behaviours. Recognizing this ‘implicit persua-
sion’, psychologists have also been discussing how adver-
tising techniques may be using specific formats to METHODOLOGY
manipulate behaviour through implicit attitude change,
where preferences are driven by non‐conscious and non‐ A total of 234 children (second and third graders) from five
rational means provided by subliminal messages (Nairn and different middle‐class schools2 returned the signed consent
Fine, 2008). In the dual‐process discussion, Gawronski and forms from their parents (response rate = 84%). However, on
Bodenhausen (2006) distinguish ‘explicit attitudes’ as the day of the experiment, three children were absent and did
deliberate evaluations, from ‘implicit attitudes’ triggered by not participate. The final sample was composed of 231
more intrinsic associations and not deliberated stimuli. young students. The group is considered to be homogeneous
Although children may have good nutritional understanding because children share similar cognitive and social devel-
and knowledge (explicit attitude), that does not mean they opment levels and because there was approximately the
would spontaneously choose one food over the other same number of girls (49.4%) and boys (50.6%) (Table 1).
(implicit attitude) because they are easily influenced or Initial qualitative research involved discussions with
manipulated by what they see or experience (Lytle et al., specialised professionals, psychologists and nutritionists3 in
1997; Contento, 2008). order to develop the most appropriate method. Moreover,
There is also empirical evidence on how promotion of several interviews with children in the age range were
healthy food influences children’s preferences and purchase conducted to define the best character to use in the
behaviour over this category, decreasing the consumption of advergame that was later developed.
less healthy available items (Rexha et al., 2010). The Pictorial cues were used in the questionnaires. These are
distinction between healthy/non‐healthy foods is still contro- easy for children to understand, organise and use in
versial but has already started to converge into a more defined communicating their choices in a way that is both enjoyable
classification. To avoid broad ‘healthy’ guidelines, the food and playful (Birch, 1980; Sullivan and Birch, 1990; Domel
‘nutrient profile’, which identifies food nutrition quality and et al., 1993). These non‐verbal measures minimise a child’s
composition, was created (Lobstein and Davies, 2009). This is dependence on the spoken language in understanding the
based on the presence or absence of nutrients that are linked to researcher’s questions and in providing his or her own
certain disease‐related or health‐related conditions (Parker, answers (Donohue et al., 1980; Macklin 1985). Question-
2003). For the purpose of this article, healthy food is defined naires, rather than semi‐structured interviews, were a more
in terms of higher nutrient and lower energy density following efficient way to reduce social desirability effect. To further
guidance from a qualified nutritionist. reduce the probability of this social desirability response, the
Advergames’ impact on children’s health has stimulated children were informed that there were no right or wrong
the interest for this project. Food marketing is considered to answers, so they could be as honest as possible, and that
influence children’s preference (Hastings et al., 2006) and to their anonymity was guaranteed (Podsakoff et al., 2003).
encourage childhood obesity growth rates, which brings us Two versions of the same advergame were designed: one
several concerns. Are children easily manipulated by the with healthy food4 and one with unhealthy food.5 The
nature of food advergames? To what extent will advergames’ selection of foods was discussed with a qualified nutritionist,
food content influence their eating behaviour? and the main character of the game was based on a well‐
This paper provides a critical view on this topic and known game character. Each time the character grabs a snack,
examines how children’s eating behaviour can be manipu- the child gains five points. To make it challenging enough for
lated through fun and entertainment. In particular, this this age group, there were bombs and bullets (enemies), which
research looks at whether the promotion of healthy food can penalised the player 10 points if he or she accidentally
lead to healthier choices and influence perceived preferences grabbed them. After getting 10 enemies, the game ends as
and knowledge by comparing two experimental groups
exposed to different stimuli (healthy and less healthy
2
advergames). The schools are situated in an urban district in the surrounding area of
Lisbon and were selected randomly from the list of schools. Prior consent
H1: Children’s spontaneous snack selection will reflect was obtained from the school board to do the research in the school.
3
Member of the educational department for health and social services—
the food content of the advergame they play. Portuguese Ministry of Education.
4
Banana, milk, strawberries, lettuce, tomatoes, fruit salad, carrots and bread.
H2: Children’s food liking will reflect the food content of 5
Ice cream, potato chips, cookies, soda, pizza, lollipop, hamburger and
the advergame they play. chocolate mousse.

Copyright © 2011 John Wiley & Sons, Ltd. J. Consumer Behav. 10: 152–160 (2011)
DOI: 10.1002/cb
Advergames: Boosting children’s eating habits 155

Table 1. Sample constitution per age and group condition


Age Food type Total

7 years 8 years Healthy Less healthy


Boys 62 55 56 61 117 (50.6%)
Girls 60 54 59 55 114 (49.4%)
Total 122 (52.8%) 109 (47.2%) 115 (49.8%) 116 (50.2%) 213 (100%)

Note: Valid n = 231.

‘game over’. There were two different game levels. In the first used were based on Portuguese culture and eating habits,
one, the character walks to the right and to the left to get food food nutritional value and children’s access to each one.
and run from enemies. Two hundred points are needed to Pictures of the selected snacks were divided into two
move on to the second level, which is faster and harder. This different 3 × 2 cards (Goldberg et al., 1978) (Figure 1). Each
time, the character is able to fly, using every direction (right, card included three healthy snacks, more wholesome and
left, up and down). Pre‐tests to evaluate the suitability and generally higher in nutrient value (here assumed healthy
general understanding of both the game and the questionnaire food), and three less wholesome snacks, higher in glucose
were conducted with five children. Based on these tests, and considered to be non‐healthy, especially if eaten in
adjustments were made to the game (comprising initial rule excess. From these, only two foods were not advertised in
clarification) and to the questionnaire (some food pictures the game.
were improved or even changed). Following the procedure used by Goldberg et al. (1978),
The experiment took place in the schools’ computer the researcher asked the children to imagine that they were in
laboratories or the in classrooms, with groups of one to four the following situation:
children at a time. A teacher or a school supervisor was
Situation 1: Now, let us pretend that your parents went to
present at all times during the study. Participants were
work and that they asked me to take care of you while
randomly assigned to either the healthy or the unhealthy
they are out. But I do not know the kind of foods you would
game and played for about 5 minutes. This is considered to
like to eat. So, suppose I said ‘here are six snacks; you can
be long enough to engage the child’s attention (Acar, 2007;
choose three to eat’. You can tell me which three you
Mallinckrodt and Mizerski, 2007; Wise et al., 2008). After
would want to eat now by putting a big X on your page.
playing the game, the participants completed the questionnaire,
which took on average 5 to 7 minutes. Each step of the process After the first set of choices was made on Card 1, they
was introduced and properly explained by the researcher who were then asked to imagine a second situation and make a
read the questions to the children. The children were allowed selection for the snacks on Card 2:
to ask questions at any time, which were answered by the
Situation 2: OK, now you can turn your page, and let us
researcher, addressed to the entire group.
pretend again that this was the second day I was taking
Explicit consent was obtained from parents, as well as the
care of you. And now, I brought six new snacks. Which
consideration of children’s participation and the adaptation
three would you want to eat?
of all instruments to the age range of the children (UNICEF,
2002; Greig et al., 2007). Besides that, we followed the Answers were coded 0 if the snack was not selected and 1
mandatory procedures in the Portuguese context, which if it was selected. A new variable was then created
involved getting permission from the education authorities corresponding to an additive index, the ‘Health Index’,
by submitting all the methodologies, which were analysed which was the sum of all selected healthy snacks in both
by an expert. All children were informed about the activity cards. Children were able to make a maximum of six healthy
dynamics, their role and their freedom to express whatever choices.
thoughts and feelings they had during the experiment
process. Their willingness to participate (whether they Food liking
wanted to participate or not, even if their parents have To evaluate food‐liking levels, children were asked to indicate
consented) was always taken into consideration as well as how much they liked each food presented (Edwards and
their personal views along the study. Hartwell, 2002; Wardle et al., 2003; Cooke and Wardle, 2005;
Nicklaus et al., 2005). In this study, a four‐point Likert scale
Measures was used to evaluate the children’s food likes and dislikes,
Snack selection which is composed of ‘I hate it’, ‘I don’t like it’, ‘I like it’ and
After the game exposure, the children were asked to choose ‘I love it’, and ‘never tried it’ options. To facilitate their
between several alternate food types, following the proce- understanding, ‘smiley faces’ illustrated each option.
dure used by Goldberg et al. (1978), Hernandez and Chapa
(2010), Mallinckrodt and Mizerski (2007) and Pempek and Nutritional knowledge
Calvert (2009). A total of 12 final snacks were chosen To analyse their conceptions of nutrition and healthy food,
(Wardle et al., 2001; Cooke and Wardle, 2005). The criteria five different pairs of food (one healthier than the other) were

Copyright © 2011 John Wiley & Sons, Ltd. J. Consumer Behav. 10: 152– 160 (2011)
DOI: 10.1002/cb
156 M. Dias and L. Agante

Figure 1. Pictures of snacks available for children to choose.

presented to the children. In each pair, they had to choose American brand, which is not present in Portugal (Table 2).
which one they thought was healthier, following the The question used pictures of each food, and the children
procedure used by Signorielli and Staples (1997) and were asked to put a cross below their selection. Selection of
Harrison (2005). Some changes from the original measure a less healthy food was coded as 0, and selection of the
of Signorielli and Staples (1997) were carried out: (i) the pair healthy option was coded as 1. The answers to this question
frosted flakes and corn flakes was removed, because children were combined into an additive index identified as the
in the pre‐test did not seem to understand the difference ‘Knowledge Index’, which is the sum of each respondent’s
between the two, and (ii) fruit roll‐up in the original question answers. The maximum attainable value is five if a child
was replaced with fruit cereal bar, because the first is an correctly chooses all the healthier options.

Copyright © 2011 John Wiley & Sons, Ltd. J. Consumer Behav. 10: 152–160 (2011)
DOI: 10.1002/cb
Advergames: Boosting children’s eating habits 157

Table 2. Food pairs hamburger in H2 (b), there is evidence not to reject a


1. Yogurt Ice cream possible effect of the game food content.
2. Big Mac Grilled chicken sandwich
3. Coca Cola Natural orange juice H3: Children’s nutritional knowledge
4. Popcorn Chocolate bar It was hypothesised that the nature of food to which children
5. Fresh fruit Fruit cereal bar
have been exposed would influence their nutritional
Note: Correct answers are in boldface. knowledge. Children were asked to choose, in each pair of
food products, the healthier option. The majority of the
children (71%) correctly identified the healthy option in all
FINDINGS five pairs. To analyse any possible association between the
game version and nutritional knowledge, a χ2‐test was
H1: Children’s snack selection conducted using the likelihood ratio.7 The results showed no
As explained previously, the children had to select six statistically significant association between both variables
snacks after playing the advergame. It was hypothesised, H1, (G = 8.082, p = 0.089), which suggests that children’s
that those who played the healthy version of the game would nutritional knowledge is not influenced by the version of
tend to select more healthy snacks. Indeed, children exposed the game they play. To further analyse this question, we
to the less healthy version have selected more frequently recoded the variable nutritional knowledge to a new variable
nutrient‐poor snacks (63% chose three or more of less healthy with two levels8, and again, H3 was rejected.
food or, in other words, only up to two healthy snacks),
whereas those who played the healthy version selected more
frequently healthier options (70% selected three or more DISCUSSION
nutritious snacks) (see Table 3). To test if there was any
association between these two variables (game type and snack The findings provide evidence to support concerns about
selection), a χ2‐test of association was used. Results show that marketing activities’ promoting unhealthy food in advergames.
these variables are associated (χ2 = 27.323, p < 0.000) but Our investigation revealed that advergames might also be
only slightly (Cramer V = 0.344). As a consequence, there is used to instigate healthier eating habits. Results show that
sufficient evidence to accept H1. children are significantly more likely to choose from those
foods that they have played with in one of the advergames’
H2: Children’s food liking versions. This is consistent with previous studies regarding
The 12 food products were grouped together according to their advergames (Mallinckrodt and Mizerski, 2007; Pempek
nutritional value (six healthy versus six less healthy), and the and Calvert, 2009; Hernandez and Chapa, 2010). The
mean of each group was computed. Analysis shows that the explanations for this behaviour might be related to the
mean preference for healthy products is higher for those active engagement provided by the interactive, fun and
children who played the healthy version of the game, whereas entertaining environment of the game (Mathiot, 2010),
the mean preference for less healthy products is higher for those memory (Hernandez and Chapa, 2010) and increasing food
who played the less healthy version (Table 4a). Those familiarity (Sullivan and Birch, 1994; Contento, 2008).
differences were found to be non‐significant6 in the case of Additionally, players are ‘disciplined through play’ into a
healthy food products (t = 1.643, p = 0.102) but significant for specific behaviour, which will depend on the food nature
less healthy ones (t = −3.718, p < 0.000). These results indicate present in the advergame (Thomson, 2010). As a
that children who played the less healthy game version tend to consequence, ‘fun food’ can be considered an opportunity
report a higher liking for the less healthy products. instead of a threat to healthy eating. Companies may find in
We also conducted the same mean comparisons on each healthier products a new profitable market, as long as they
product to see any differences in specific food items can reach young consumers’ attention and interest and, at
(Table 4b). Results confirmed that there were no significant the same time, enhance their corporate social responsibility.
differences on healthy food products. However, in half of the These results provide further support for social marketing
non‐healthy food items [potato chips (t = 3.658, p < 0.000), campaigns to fight childhood obesity through health‐
hamburger (t = 3.122, p = 0.002) and pizza (t = 3.063, promoting advergames.
p = 0.003)], we found statistically significant differences
between the children who played different game versions. As Food liking
expected, these three products registered a higher mean Children’s food liking also seemed to be affected by the
preference in the group of children who have played the less content of the game but only for the less healthy food items
healthy version of the game. (particularly potato chips, hamburger and pizza). The fact
Therefore, there is insufficient evidence to support that there are no significant differences between preferences
hypothesis H2 (a) or hypothesis H2 (b) for cookies, lollipop for most foods might be explained by some intrinsic factors
and chocolate mousse. However, for pizza, chips and
7
This is because 40 per cent of the cells had an expected count of less than
6
Because of the metric nature of this measure, independent t‐tests were used five. The Pearson χ2 should not be analysed when more than 20 per cent of
to establish if any significant difference existed between game groups with a the cells have an expected count of less than five.
8
statistical significance of 0.05 for all tests. Medium/low and high nutritional knowledge (Table 5).

Copyright © 2011 John Wiley & Sons, Ltd. J. Consumer Behav. 10: 152– 160 (2011)
DOI: 10.1002/cb
158 M. Dias and L. Agante

Table 3. Health Index (sum of healthy selected items) per game version
0 1 2 3 4 5 6
Less healthy game 6 (5.2%) 27 (23.3%) 40 (34.5%) 17 (14.7%) 14 (12.1%) 7 (6.0%) 5 (4.3%)
Healthy game 2 (1.7%) 12 (10.4%) 21 (18.3%) 27 (23.5%) 21 (18.3%) 13 (11.3%) 19 (15.5%)
Total 8 (3.5%) 39 (16.9%) 61 (26.4%) 44 (19.0%) 35 (15.2%) 20 (8.7%) 24 (10.4%)

Note: Valid n = 231.

Table 4a. Food liking averages by group: healthy food versus less healthy food
Healthy fooda Less healthy foodb

Mean SD Mean SD
Less healthy game 3.11 0.588 3.55 0.420
Healthy game 3.23 0.542 3.28 0.645

Note: SD = standard deviation.


Banana, tomatoes, lettuce, strawberries, carrot and bread.
Cookies, potato chips, hamburger, lollipop, pizza and chocolate mousse.

Table 4b. Food liking averages of each single food item by game version
Less healthy game Healthy game Mean
difference
Valid n Mean SD Valid n Mean SD
Banana 115 3.43 0.762 115 3.39 0.746 0.04
Strawberries 111 3.40 0.937 112 3.36 0.948 0.04
Tomatoes 111 2.49 1.242 109 2.77 1.191 −0.28
Lettuce 116 3.03 1.071 114 3.15 1.107 −0.12
Carrots 116 2.92 1.136 112 3.19 1.027 −0.27
Bread 116 3.32 0.787 115 3.46 0.704 −0.14
Lollipop 114 3.43 0.841 115 3.23 0.976 0.20
Pizza 115 3.80 0.463 115 3.55 0.752 0.25
Cookies 113 3.60 0.662 115 3.42 0.805 0.18
Potato chips 115 3.68 0.539 115 3.33 0.866 0.35
Hamburger 115 3.46 0.798 115 3.08 1.044 0.38
Chocolate mousse 112 3.30 1.003 114 3.07 1.095 0.23
Note: SD = standard deviation.

Table 5. Nutritional knowledge (recoded) Nevertheless, although food exposure in advergames did
Low/Medium High knowledgeb not enhance preferences for healthy products, it does appear
knowledgea to increase liking for some less healthy products. Although
more research is needed to analyse this effect, its verification
Less healthy 11 (9%) 105 (91%)
game version would recommend restrictions on the advertising of such
Healthy game version 5 (4%) 110 (96%) products. We would recommend a combined approach of
Total 16 (7%) 215 (93%) regulation for this type of products near children’s
programmes, with self‐regulation from marketers (like the
Note: Valid n = 231.
Up to three correct answers (zero, one, two or three correct). European Union pledge signed by some food and drink
Four or more correct answers (four or five correct). advertisers in Europe), and finally, the introduction of media
literacy programmes in the school curriculum as recom-
mended by the European Union.
that are harder to change or influence in a one‐shot However, as mentioned earlier, the food industry has
interaction, for instance, biological predispositions to a mostly been promoting energy‐dense food and strong
specific food or taste (children’s higher tolerance to sweet communication campaigns directed towards children (Story
rather than to bitter) and experience with food, which and French, 2004; Hastings et al., 2006). On the contrary,
consists of physiological learning arising from previous food not many campaigns using the same appealing strategies that
exposure (Contento, 2008). According to Contento (2008), are used by the ‘unhealthy’ products have been devoted to
these are the factors that mostly affect preferences and food promoting healthy eating among children. This raises the
dislikes. following question: was the effect on the food liking only

Copyright © 2011 John Wiley & Sons, Ltd. J. Consumer Behav. 10: 152–160 (2011)
DOI: 10.1002/cb
Advergames: Boosting children’s eating habits 159

due to playing the advergame, or was it because the REFERENCES


advergame enhanced the unhealthy products? In the case of
the latter, a marketing opportunity may exist in the Acar A. 2007. Testing effects of incidental advertising exposure in
promotion of healthy eating by using fun and play strategies online gaming environments. Journal of Interactive Advertising
8(1): 45–56.
that employ new techniques such as advergames in order to
Acuff D, Reiher RH. 1997. What Kids Buy and Why. The
increase the interaction of children with these products Psychology of Marketing to Kids. Free Press: New York.
through the game experience. Alvy LM, Calvert SL. 2008. Food marketing on popular children’s
web sites: A content analysis. Journal of the American Dietetic
Nutritional knowledge Association 108(4): 710–713.
Bardhi F, Rohm AJ, Sultan F. 2010. Tuning in and tuning out:
No significant differences were found in what concerns media multitasking among young consumers. Journal of
children’s nutritional knowledge. Children in this age group Consumer Behaviour 9(4): 316–332.
have a very good understanding of what is healthy and Birch LL. 1980. Effects of peer models’ food choices and eating
unhealthy food. Indeed, these findings are aligned with the behaviors on preschoolers’ food preferences. Child Develop-
nutritionist’s expectations because such concepts and ment 51(2): 489–496.
Contento I. 2008. Nutrition education: linking research, theory,
nutritional contents are introduced early in elementary and practice. Asia Pacific Journal of Clinical Nutrition 17(1):
Portuguese school. However, as explained in the dual‐ 176–179.
process model, having nutritional knowledge does not Cooke LJ, Wardle J. 2005. Age and gender differences in children’s
guarantee that children will follow a balanced diet or food preferences. The British Journal of Nutrition 93(5):
will not be persuaded by advertisers to make less healthy 741–746.
Coon KA, Goldberg J, Rogers BL, Tucker KL. 2001. Relationships
choices. between use of television during meals and children’s food
consumption patterns. Pediatrics 107(1) e7: 1–9.
Culp J. 2010. Characteristics of food industry web sites and
LIMITATIONS AND INSIGHTS FOR “advergames” targeting children. Journal of Nutrition Educa-
FURTHER RESEARCH tion and Behaviour 42(3): 197–201.
Domel SB, Baranowski T, Davis H, Leonard SB, Riley P,
Baranowski J. 1993. Measuring fruit and vegetable preferences
One limitation of this study was the short‐term evaluation. among 4th‐ and 5th‐grade students. Preventive Medicine 22(6):
Even though an effect was found on children’s snack 866–879.
selection, the children answered the questionnaire immedi- Donohue TR, Henke LL, Donohue WA. 1980. Do kids know what
ately after playing the game. Medium‐term and long‐term TV commercials intend? Journal of Advertising Research 20(5):
51–56.
effects of advergames on food choices were not addressed Edwards JSA, Hartwell HH. 2002. Fruit and vegetables—attitudes
herein and could constitute an insight for future research. and knowledge of primary school children. Journal of Human
One possibility would be to increase the time lag between Nutrition and Dietetics 15(5): 365–374.
exposure and measurement in order to account for medium‐ Galst JP. 1980. Television food commercials and pro‐nutritional
public service announcements as determinants of young
term effects. Another suggestion would be to evaluate the children’s snack choices. Child Development 51(3): 935–938.
long‐term effects by using a longitudinal design, which Gawronski B, Bodenhausen GV. 2006. Associative and proposi-
could also introduce repeated exposure effects. The present tional processes in evaluation: an integrative review of implicit
study did not examine any changes before and after and explicit attitude change. Psychological Bulletin 132(5):
treatment. Instead, it only made inferences about possible 692–731.
differences caused by exposure to different visual stimuli Goldberg ME, Gorn GJ, Gibson W. 1978. TV messages for snack
and breakfast foods: do they influence children’s preferences?
(food type). Therefore, it can be interesting to address this Journal of Consumer Research 5(2): 73–81.
nutritional assessment comparison before and after game Gorn GJ, Goldberg ME. 1982. Behavioral evidence of the effects of
exposure to better evaluate positive or negative changes. televised food messages on children. Journal of Consumer
The focus here was on elementary‐school‐age children Research 9(2): 200–205.
within 7 and 8 years old. However, interesting conclusions Greig A, Taylor J, MacKay T. 2007. Doing Research with
Children, 2nd edn. Sage Publications, Inc.: London.
may arise from children in different cognitive development
Harrison K. 2005. Is “fat free” good for me? A panel study of
stages. television viewing and children’s nutritional knowledge and
reasoning. Health Communication 17(2): 117–132.
ACKNOWLEDGEMENT Hart KH, Bishop JA, Truby H. 2002. An investigation into school
children’s knowledge and awareness of food and nutrition.
Journal of Human Nutrition and Dietetics 15(2): 129–140.
The authors thank the Editor David Marshall and 2 Hastings G, McDermott L, Angus K, Stead M, Thomson S. 2006.
anonymous JCB reviewers for their constructive feedback. The extent, nature and effects of food promotion to children: a
review of the evidence. Technical paper prepared for the World
Health Organization. Electronic Resource.
Hernandez MD, Chapa S. 2010. Adolescents, advergames and
BIOGRAPHICAL NOTES snack foods: effects of positive affect and experience on
memory and choice. Journal of Marketing Communications 16
Mónica Dias was a master student at NOVA SBE and works at
(1): 59–68.
WeDo technologies in Singapore.
Lee M, Choi Y, Quilliam ET, Cole RT. 2009. Paying with food:
Luísa Agante is a Visiting Assistant Professor of Marketing at content analysis of food advergames. The Journal of Consumer
NOVA SBE. Affairs 43(1): 129–154.

Copyright © 2011 John Wiley & Sons, Ltd. J. Consumer Behav. 10: 152– 160 (2011)
DOI: 10.1002/cb
160 M. Dias and L. Agante

Lee M, Youn S. 2008. Leading national advertisers’ uses of Roedder, DL. 1981. Age differences in children’s responses to
advergames. Journal of Current Issues and Research in television advertising: an information‐processing approach.
Advertising 30(2): 1–13. Journal of Consumer Research 8(2): 144–153.
Livingstone S, Haddon L. 2009. EU Kids Online: Final Report. EU Selman RL. 1980. The Growth of Interpersonal Understanding.
Kids Online: LSE, London. (EC Safer Internet Plus Programme Academic Press: New York.
Deliverable D6.5). Signorielli N, Staples J. 1997. Television and children’s concep-
Lobstein T, Davies S. 2009. Defining and labelling ‘healthy’ and tions of nutrition. Health Communication 9 (4): 289–301.
‘unhealthy’ food. Public Health Nutrition 12(3): 331–340. Singleton N, Rhoads DS. 1984. An assessment of the nutrition
Lobstein T. 2010. 11th International Congress on Obesity. 11-15 education of students in grades 3–12. Journal of the American
July 2010, Stockholm, Sweden. Dietetic Association 84(1): 59–63.
Lytle L, Eldridge A, Kotz K, Piper J, Williams S, Kalina B. 1997. Story M, French S. 2004. Food advertising and marketing
Children’s interpretation of nutrition messages. Journal of directed at children and adolescents in the US. Interna-
Nutrition Education 29(3): 128–136. tional Journal of Behavioral Nutrition and Physical Activity 1
Macklin CM. 1985. Do young children understand the selling intent (1): 3.
of commercials. Journal of Consumer Affairs 19(2): 293–304. Sullivan SA, Birch LL. 1990. Pass the sugar, pass the salt:
Mallinckrodt V, Mizerski D. 2007. The effects of playing an experience dictates preference. Developmental Psychology 26(4):
advergame on youth children’s perceptions, preferences, and 546–551.
requests. Journal of Advertising 36(2): 87–100. Sullivan SA, Birch LL. 1994. Infant dietary experience and
Mathiot L. 2010. Child consumption of fun food: between acceptance of solid foods. Pediatrics 93(2): 271–277.
deviating practice and re‐appropriating food‐use. Young Con- Thomson DM. 2010. Marshmallow power and Frooty treasures:
sumers 11(2): 108–116. disciplining the child consumer through online cereal
McCrindle M. 2006. Word Up. McCrindle Research: Sydney. advergaming. Critical Studies in Media Communication 27(5):
McGinnis JM, Gooman JA, Kraak VI. 2006. Food Marketing to 438–454.
Children and Youth. Threat or Opportunity? The National Tufte B, Rasmussen J. 2005. Children and adolescents use of the
Academies Press: Institute of Medicine, Washington, D.C. Internet—with a focus on tweens. In Tufte B, Rasmussen J,
Moore ES. 2006. It’s Child Play: Advergaming and the Online Christenses AB (eds). Frontrunners and Copycats. Copenhagen
Marketing of Food to Children. Kaiser Family Foundation: Business School Press: Copenhagen.
Menlo Park, CA. Tufte B, Rasmussen J. 2011. Children and the Internet. In Marshall
Moses LJ, Baldwin DA. 2005. What can the study of cognitive D (ed). Understanding Children as Consumers. Sage: London;
development reveal about children’s ability to appreciate and 184–201.
cope with advertising? Journal of Public Policy and Marketing. UNICEF Evaluation Office. 2002. Children Participating in
24(2): 186–201. Research, Monitoring and Evaluation (M&E)—Ethics and Your
Nairn A, Fine C. 2008. Who’s messing with my mind? The Responsibilities as a Manager. Evaluation Technical Notes, No.
implications of dual‐process models for the ethics of advertising 1. UNICEF: New York; 1–11.
to children. International Journal of Advertising. 27(3): 447–470. Wardle J, Herrera M‐L, Cooke L, Gibson EL. 2003. Modifying
Neuborne E. 2001. For Kids on the Web, It’s an Ad, Ad, Ad, Ad children’s food preferences: the effects of exposure and reward
World: How to Help Your See the Sales Pitches Behind Online on acceptance of an unfamiliar vegetable. European Journal of
Games. Business Week, August 13. Available at http://www. Clinical Nutrition 57(2): 341–348.
businessweek.com/print/magazine/content/01_33/b3745121. Wardle J, Sanderson S, Gibson EL, Rapoport L. 2001. Factor‐
htm?chan=mz [accessed on 23 September 2010]. analytic structure of food preferences in four‐year‐old children
Nicklaus S, Boggio V, Chabanet C, Issanchou S. 2005. A in the UK. Appetite 37(3): 217–223.
prospective study of food variety seeking in childhood, Webber K, Story M, Harnack L. 2006. Internet food marketing
adolescence and early adult life. Appetite 44(3): 289–297. strategies aimed at children and adolescents: a content analysis
Noble C, Corney M, Eves A, Kips M. 2000. Food choice and school of food and beverage brand web sites. Journal of the American
meals: primary schoolchildren’s perceptions of the healthiness of Dietetic Association 106(9): 1463–1466.
foods and the nutritional implications of food choices. Interna- Wellman H, Johnson C. 1982. Children’s understanding of food
tional Journal of Hospitality Management 19(4): 413–432. and its functions: a preliminary study of the development of
Parker BJ. 2003. Food for health. The use of nutrient content, concepts of nutrition. Journal of Applied Development
health, and structure/function claims in food advertisements. Psychology 3(2): 135–148.
Journal of Advertisement 32(3): 47–55. WHO. 2010. Population Based Prevention Strategies for Childhood
Pempek TA, Calvert SL. 2009. Tipping the balance. Use of Obesity. Report of a WHO Forum and Technical Meeting,
advergames to promote consumption of nutritious foods and Geneva, 15–17 December 2009.
beverages by low‐income African American children. Archives Williams J, Wake M, Hesketh K, Maher E, Waters E. 2005.
of Pediatrics & Adolescent Medicine 163(7): 633–637. Health‐related quality of life of overweight and obese
Piaget J, Inhelder B. 1972. The Psychology of the Child. Basic children. Journal of the American Medical Association 293
Books: New York. (1): 70–76.
Podsakoff PM, Mackenzie SB, Lee J‐Y, Podsakoff NP. 2003. Winkler T, Buckner K. 2006. Receptiveness of gamers to
Common method biases in behavioral research: a critical review embedded brand messages in advergames: attitudes towards
of the literature and recommended remedies. The Journal of product placement. Journal of Interactive Advertisement 7(1):
Applied Psychology 88(5): 879–903. 24–32.
Rexha D, Mizerski K, Mizerski D. 2010. The effect of availability, Wise K, Bolls PD, Kim H, Venkataramn A, Meyer R. 2008.
point of purchase advertising, and sampling on children’s first Enjoyment of advergames and brand attitudes: the impact of
independent food purchases. Journal of Promotion Management thematic relevance. Journal of Interactive Advertising 9(1):
16(1): 148–166. 27–36.

Copyright © 2011 John Wiley & Sons, Ltd. J. Consumer Behav. 10: 152–160 (2011)
DOI: 10.1002/cb

You might also like