Journal of Public Health | Vol. 29, No. 4, pp. 358 –367 | doi:10.1093/pubmed/fdm067 |
Advertising of food to children: is brand logo recognition
related to their food knowledge, eating behaviours and
food preferences?
C. A. Kopelman1, L. M. Roberts2, P. Adab3
1
Birmingham University Medical School, University of Birmingham, Birmingham B15 2TT, UK
Department of Primary Care and General Practice, University of Birmingham, Birmingham B15 2TT, UK
3
Department of Public Health and Epidemiology, University of Birmingham, Birmingham B15 2TT, UK
Address correspondence to L. M. Roberts, E-mail: [email protected]
2
A B S T R AC T
Background There remains controversy about the contribution of food advertising targeted at children to the epidemic of childhood obesity in
the UK. The aim of this study is to explore the relationship between the ability to recognize brand logos featured in promotional campaigns of the
food industry and eating behaviours, food knowledge and preferences in children aged 9– 11 attending six primary schools in Birmingham, West
Midlands.
Methods A ‘20 flashcard’ brand logo quiz assessed children’s brand logo recognition ability; a self-completed questionnaire collected information
on children’s socio-demographic characteristics, eating behaviours, food knowledge and preferences (n ¼ 476).
Results Children demonstrated both high brand logo recognition abilities with 88.4% (420/476) recognizing at least 16/20 brand logos in the
quiz and high levels of poor diet. No strong correlation was found between higher brand logo recognition ability and poorer eating behaviours,
food knowledge and preferences.
Conclusion Although many children are familiar with commonly presented logos of food products, brand awareness does not appear to be a
major influence on the consumption of a poor diet amongst children. The regulation or restriction of food advertising to children is unlikely to
have a significant impact on obesity rates among children unless combined with measures to address other detrimental influences.
Keywords advertising, children, obesity, diet, food preference
Introduction
The rapidly increasing prevalence of childhood obesity in
the UK over the past decade has occurred against a backdrop of an ever developing and expanding food industry
where advertising to children has become accepted and
actively promoted.1,2 This observation has fuelled the considerable debate in recent years over the possibility of a
causal link between food advertising and the childhood
obesity epidemic. Food advertising campaigns, of which
95% are for nutritionally poor foods, comprise the largest
category of products advertised on children’s television in
the UK; 7 out of 10 adverts on children’s weekday ITV television are for food products.3 Furthermore, it has been well
documented that children are both receptive and responsive
to advertising campaigns.4 – 6
The growing concern over the influence that food promotion has on a child’s eating behaviour has led to the
358
publication of two key reports in the UK. A systematic
review of all the evidence available by the Food Standards
Agency (FSA) in 2003, and a report by the Office of
Communications (Ofcom) in 2004.7,8 Both concluded that
promotion by the food industry has a detrimental impact on
children’s food preferences, eating behaviour and the purchase requests they make to their parents. These conclusions
were drawn from evidence that demonstrated the volume
and persuasive content of food advertisements, the recognized link between television, diet and obesity, the experimental findings that exposure to adverts has a direct impact
C. A. Kopelman, Medical Student
L. M. Roberts, Senior Lecturer
P. Adab, Senior Lecturer
# The Author 2007, Published by Oxford University Press on behalf of Faculty of Public Health. All rights reserved.
FO O D A DV E RT I S I N G A N D CH IL D H O O D O B E S I T Y
on children’s food choices and self-reported (by parents and
children) influences on eating behaviour.
There are, however, major limitations to these two reports
and these restrict the confident acceptance and application
of their findings.9 The FSA review is dominated by studies
undertaken in North America between 20 and 30 years ago
and is thus of questionable generalizability to the current
environment and UK setting. Furthermore, both reports
concentrate principally on the effects of television advertising only; this is not reflective of the present day multitude
of food promotion methods and mediums that children are
exposed to.
The food industry has always contested the claim that
they are contributing to the childhood obesity crisis.10
Instead, they argue that they are simply promoting specific
brand preference in an already well-established market. The
current study was undertaken, given the limitations of the
existing research, to investigate this claim in a robust and
methodological manner. The aim was to establish whether
there is a relationship between children’s ability to recognize
brand logos featured in promotional campaigns of the food
industry, and their food knowledge, eating behaviours and
food preferences.
In the UK, support for legislation that restricts or regulates food advertising to children is growing and comes
from consumer groups, the media and public health organizations.11 In November 2006, in acknowledgement of
public opinion, Ofcom imposed restrictions on the television advertising of food and drink products high in fat, salt
and sugar during television programmes with particular
appeal to children.12 The scale and implications of the
current childhood obesity epidemic warrants both prompt
and effective action. A clearer evidence base is needed to
evaluate the existing decisions regarding food advertising to
children to ensure legislation is effective in reducing the
current obesity crisis.
Methods
The study was conducted during February and March 2005.
Study objectives
The main objectives of the study were: (i) to explore the
eating behaviours, food knowledge and preferences of
primary school children; (ii) to assess the ability of
primary school children to recognize brand logos that are
currently featured in food promotional campaigns; (iii) to
relate differences in individuals’ abilities to recognize
brand logos with their eating behaviours, food knowledge
359
and preferences; (iv) to establish the role of gender, socioeconomic group and ethnicity in children’s brand logo recognition and their eating behaviours, food knowledge and
preferences.
Study population
The sample population consisted of children in National
Curriculum years 5 and 6 (children aged 9 –11). It has been
found that children start to develop self-care activities
including food choices between the ages of 3 and 8 and that
these are stable by ages 9 – 11.13
The study sample was drawn from six primary schools
situated in South Birmingham, West Midlands: five were
state schools and one was an independent school. A convenience sampling technique was employed to recruit the
schools into the study. A letter of invitation to participate
was sent to all primary schools in South Birmingham that
fulfilled specified criteria (more than 200 pupils; of secular
denomination) and whose contact details were accessible on
the Internet. All schools that agreed to take part were
recruited into the study.
The study group consisted of all children present in years
5 and 6 on the day of the study. No exclusion criteria were
employed; all children were allowed to take part unless their
parents had explicitly expressed a wish otherwise. Ethical
approval was not required. However, prior approval was
obtained from the Student sub-committee of the South
Birmingham LREC to ensure best practice.
Study design
The study was a cross-sectional, questionnaire-based survey.
Children’s brand logo recognition ability was assessed by
a short quiz. The quiz consisted of a series of 20 flashcards
each displaying a coloured image of a frequently advertised
brand logo of food or drink (Table 1). The children were
required to write down for each logo foods they believed it
represented (e.g. burgers, chips, etc.). The logos included in
the quiz had been selected based on a review of commonly
advertised food products in children’s magazines, on community billboards and during children and adult television
viewing times.
Table 1 The 20 logos used in the brand logo recognition quiz
MacDonald
Kellogs
Haribo
McVities
Pizza Hut
Robinsons
Starbursts
Mr Kipling
KFC
Ribeena
Muller
Bisto
Walkers
Evian
Dairylea
Birds eye
Pringles
Kinder
Jacobs
McCains
360
J O U RN A L O F P U B L I C H E A LTH
A questionnaire was used to collect information on
the children’s socio-demographic characteristics (section 1),
their eating behaviours (section 2) and their food knowledge
and preferences (section 3). The quiz and questionnaire
responses were anonymized but linked.
The questions used in sections 1 and 2 were taken from a
validated questionnaire used in the Relachs study 2001.14
This was a large cross-sectional survey investigating the
health of 2800 children from three areas of East London.
The children’s socio-economic status was assessed by postcode using the Index of Multiple Deprivation (IMD)
2004.15 This reports deprivation at a small area level (continuous geography areas of approximately 1500). Parental
employment status and ethnicity were also recorded.
Section 3 of the questionnaire was modelled on a scale
developed in America by Calfas et al. 16 It used a series of
10 matched pairs of food types, one being healthy and the
other unhealthy (Table 2). The children were required to
record for each food pair which food type she/he believed
to be the healthiest and the food type she/he would prefer.
The test was validated by Calfas and found to have high validity and test –retest reliability for both preference and
knowledge tests (Table 3).
Both the quiz and questionnaire were piloted and modified accordingly, prior to the data collection process taking
place.
The quiz and questionnaire were administered to the children in their classrooms under the supervision of the
Table 2 The 10 matched pairs of food types
1. Orange juice versus orange squash
2. Water versus coca-cola
3. Bran flakes versus frosties
4. An apple versus a fruity lollipop
5. A cereal bar versus a chocolate bar
6. Pasta versus pot noodle
7. Jacket potato versus French fries
8. Cheese and tomato sandwich versus cheese and tomato pizza
9. Roast chicken versus chicken burger
10. Strawberry yoghurt versus strawberry mousse
Table 3 Validation scores of the preference and knowledge tests15
Test –retest reliability score
Validity
Preference test
R ¼ 0.70, P , 0.01
66%, P , 0.001
Knowledge test
R ¼ 0.72, P , 0.01
Not available
researcher and the class teacher. The children were informed
of the anonymity of their answers, were given time to ask
questions and were guided question by question through the
quiz and questionnaire to increase the completion rate. The
class teacher ensured silence during testing to prevent collusion between the children.
Statistical analysis
The brand logo recognition ability of the children was
assessed by summing up the total number of correct
answers given in the quiz, to form a score out of 20.
Eating behaviours were determined by responses to the
following.
† Frequency of eating breakfast before school (ate breakfast
everyday versus did not)
† Total portions of fruit and vegetables consumed per day
(consumed at least five portions versus did not)
† Daily consumption rates of crisps, biscuits, sweets, fried
food and fizzy drinks (high consumption: at least once a day
versus low consumption: less than once a day)
All response categories were formed using nationally
approved standards.8
Knowledge of healthy foods was assessed by summing
up the total number of correctly identified ‘healthy’ food
types out of the 10 food pairs, to form a knowledge score
out of 10. Similarly, food preferences were assessed by
summing up the total number of healthy food types
preferred out of the 10 food pairs, to form a total healthy
preference score out of 10.
Data analysis was performed using SPSS-13.0 computer
statistical software package. Comparative analysis using
either the Mann –Whitney or Kruskal – Wallis tests was used
to compare median brand logo scores of children with contrasting food knowledge, eating behaviours and food preferences. Non-parametric tests were used due to the
non-normal distribution of the brand logo scores in the
sample population.
Chi-squared analysis for categorical variables was used to
explore relationships between gender, ethnicity, socioeconomic status (determined by postcode deprivation
score15) and parental employment status, and the brand logo
recognition abilities and eating behaviours, food knowledge
and preferences of the children.
Results
Four hundred and seventy-six children took part in the
study and completed both the quiz and the questionnaire.
FO O D A DV E RT I S I N G A N D CH IL D H O O D O B E S I T Y
361
n
%
Male
254
53.4
Female
222
46.6
White
308
65.4
Asian
65
13.8
Black
50
10.6
Mixed race
36
7.6
Chinese, Malaysian or other
12
2.5
Of these children, 254 were boys (53.4%) and the average
age of the children was 10.02 (SD ¼ 0.72). Overall 65.4%
of the sample population were white, 13.8% Asian, 10.6%
black, 7.6% of mixed race and 2.5% Chinese, Malaysian or
of other ethnic origin; 8.7% of the children lived in an area
of deprivation score 1, 44.4% in an area of deprivation
score 2 or 3 and 45.8% in an area of deprivation score 4. It
should be noted that a higher deprivation score is indicative
of an area of greater deprivation and thus lower socioeconomic status.15 These findings are summarized in
Table 4 and are similar to the general population of
Birmingham, and many inner city areas in the UK.17 – 19
Children’s eating behaviours
Table 4 Characteristics of the children
Gender
Ethnicity
Deprivation scorea
1
28
8.7
2
73
22.7
3
73
22.7
4
147
45.8
a
Higher score is indicative of greater deprivation and thus lower
socio-economic status.15
Over a quarter (120/469) of children in the study reported
that they did not eat breakfast everyday before going to
school (Table 5). Breakfast eating was less common among
children living in an area of high deprivation compared with
those living in an area of low deprivation (x2 ¼ 18.3, P ¼
0.03) and among children whose parents were unemployed
Table 5 Eating behaviours classified by gender, deprivation score, parental employment status and ethnicity
Eats breakfast
Eats five portions
High
High
High consumption
High consumption of
everyday, % (n)
fruit and vegetables
consumption of
consumption of
of fried food, % (n)
fizzy drinks, % (n)
per day, % (n)
crisps, % (n)
sweets, % (n)
74.4 (349)
56.9 (260)
72.1 (334)
70.9 (326)
51.8 (237)
60.2 (274)
Male
75.9 (189)
55.8 (134)
72.5 (179)*
72.5 (176)
53.0 (128)
63.8 (153)
Female
72.5 (158)
57.7 (124)
72.0 (154)*
68.8 (148)
50.7 (109)
56.3 (120)
1
96.4 (27)**
64.3 (18)
50.0 (14)
67.8 (19)
26.9 (7)***
42.3 (11)***
2
81.7 (58)**
59.7 (43)
61.7 (45)
63.9 (46)
43.9 (32)***
38.9 (28)***
3
84.9 (62)**
60.6 (43)
77.8 (56)
77.5 (55)
37.7 (26)***
58.6 (41)***
4
67.1 (98)**
56.7 (80)
77.8 (112)
70.2 (101)
63.5 (92)***
79.5 (100)***
Overall
Gender
Deprivation score
Parental
employment
status
Unemployed
61.0 (36)**
51.7 (30)
83.1 (49)
86.2 (50)***
77.6 (45)*
69.0 (40)***
Employed
75.8 (244)**
60.3 (188)
69.5 (221)
66.1 (209)***
44.6 (141)*
55.4 (175)***
White
73.6 (226)
55.2 (164)
73.9 (223)
74.0 (222)***
50.5 (151)
61.6 (185)*
Mixed
77.1 (27)
55.9 (19)
77.2 (27)
57.2 (20)**
58.8 (20)
57.6 (19)*
Black
66.0 (33)
62.5 (30)
75.0 (36)
68.1 (32)**
58.3 (28)
65.9 (31)*
Asian
80.6 (50)
61.9 (39)
63.5 (40)
71.4 (45)**
50.6 (32)
58.0 (36)*
Other
91.7 (11)
58.3 (7)
36.4 (4)
33.3 (4)**
41.7 (5)
9.1 (1)*
Ethnicity
*P , 0.001.
**P , 0.05.
***P , 0.01.
362
J O U RN A L O F P U B L I C H E A LTH
compared with those whose parents were employed (x2 ¼
13.4, P ¼ 0.04). No significant difference was observed by
gender or ethnicity.
Over half (260/457) of the children reported eating five
or more portions of fruit and vegetables per day. No factor
under investigation was associated with consumption of the
recommended portions of fruit and vegetables. Such a high
proportion of children reporting these levels of fruit and
vegetable consumption per day is not surprising as all
schools were participating in the government’s 5-a-day
initiative at the time of study conduction. This initiative, in
addition to providing each child with one portion of fruit
per day, would have raised children’s general awareness of
the recommended national standards inevitably impacting
on both real and self-reported behaviour.
Overall, children displayed high consumption rates of the
specified less healthy foods and drinks with 72.1% (334/
463) consuming crisps, 70.9% (326/460) sweets, 51.8%
(237/458) fried food, 60.2% (274/455) fizzy drinks and
56.3% (255/453) biscuits at least once a day. Children from
areas of high deprivation were found to be significantly
more likely to consume high levels of fried food (x2 ¼ 27.5,
P , 0.01) and fizzy drinks (x2 ¼ 34.2, P , 0.01) in comparison to children from areas of lower deprivation.
Children whose parents were unemployed were found to be
significantly more likely to consume high levels of sweets
(x2 ¼ 33.1, P ¼ 0.03), fried food (x2 ¼ 41.1, P , 0.01) and
fizzy drinks (x2 ¼ 10.3, P , 0.01) compared with those
whose parents were employed. Levels of consumption of
sweets and fizzy drinks varied significantly by ethnicity: the
highest consumption of sweets being in the white population (x2 ¼ 12.9, P ¼ 0.01) and the highest consumption of
fizzy drinks in the black population (x2 ¼ 13.1, P ¼ 0.01).
significantly more likely to have a poorer knowledge of
healthy foods in comparison to children with employed
parents (x2 ¼ 11.3, P 0.01). Finally, children of mixed,
black and Asian ethnic origins, respectively, were significantly
more likely to have a preference for the less healthy foods in
comparison to children of white and ‘other’ ethnic origins
(x2 ¼ 13.6, P , 0.01).
Children’s brand logo recognition ability
The children demonstrated a high brand logo recognition
ability with 88% (420/475) recognizing at least 16 out of
the 20 brand logos in the quiz and only 2.3% (11/475)
recognizing five or less. Figure 1 demonstrates the numbers
of children correctly identifying each of the brand logos.
The total median scores on the brand logo quiz were
significantly higher among girls in comparison to boys
(Z¼ 23.14, P , 0.01) and among the white and mixed
ethnic groups in comparison to the other ethnicities
(x2 ¼ 44.7, P , 0.01). No significant difference in brand
Table 6 Food knowledge and preferences categorized by gender,
deprivation score, parental employment status and ethnicity
Overall
Children scoring 6 or
Children scoring 6 or
greater in food
greater in food
preference testa
knowledge testb
% (n)
P-value
% (n)
P-value
27.1 (114)
—
95.4 (417)
—
Gender
Male
23.9 (53)
Female
31.1 (61)
0.173
93.0 (213)
0.040
98.1 (202)
Deprivation score
1
24.0 (6)
Children’s food knowledge and preferences
2
24.6 (16)
0.929
97.1 (68)
100 (26)
Although overall children demonstrated a good knowledge
of which foods were ‘healthy’ or ‘unhealthy’ they showed a
definite preference for the less healthy options out of the
ten food pairs (Table 6). The majority of children (95.4%)
scored six or more in the knowledge test compared with
only 27.1% in the food preference test. The discrepancy
between food preference and knowledge was present and
significant at the individual level. Ninety-four percent of
children preferring five or less of the healthier options were
able to correctly identify six or more of the healthier
options (x2 ¼ 4.3, P ¼ 0.04).
Boys were significantly more likely than girls to have a
poorer knowledge of what foods were healthy (x2 ¼ 6.4,
P ¼ 0.04). Children whose parents were unemployed were
3
29.2 (19)
96.9 (62)
4
26.9 (36)
96.5 (137)
0.459
Parental employment
status
Unemployed
24.1 (13)
Employed
29.4 (85)
0.274
87.5 (49)
0.003
91.1 (296)
Ethnicity
White
31.5 (87)
Mixed
16.1 (5)
0.009
94.5 (273)
96.8 (30)
Black
20.0 (8)
95.6 (43)
Asian
14.0 (8)
98.3 (57)
Other
50.0 (6)
100 (11)
0.675
a
Higher score indicating more healthy preferences.
b
Higher score indicating greater knowledge of healthy eating behaviours.
363
FO O D A DV E RT I S I N G A N D CH IL D H O O D O B E S I T Y
eating behaviours or the food preferences of the children
appeared to affect brand logo recognition ability (Table 8).
Discussion
Main findings of this study
Fig. 1 Numbers of children correctly identifying each brand logo in the
brand logo recognition quiz (total number completing ¼ 476).
logo recognition ability was found by deprivation score or
parents’ employment status (Table 7).
Associations between brand logo recognition ability
and eating behaviours, food knowledge and
preferences
Greater recognition rates of the brand logos was significantly associated with high snacking of crisps (Z ¼ 22.33,
P ¼ 0.02), low snacking of biscuits (Z ¼ 22.10, P ¼ 0.04)
and a better food knowledge in terms of what foods are
‘healthy’ or ‘unhealthy’ (Z ¼ 23.55, P , 0.01). No other
The results of the study confirm the findings of previous
studies that the diet of children aged 9– 11 living in the UK
is poor and may be a factor contributing to the childhood
obesity epidemic.20 For the purpose of the study, ‘unhealthy’
eating behaviour referred to the consumption of foods or
drinks, high in fat and/or sugar, at least once a day. This
definition stemmed from the belief that there is only such a
thing as an ‘unhealthy’ diet, not an ‘unhealthy’ food and is
in line with the definition of unhealthy foods included in the
recent Ofcom consultations.
The majority of the children in the study reported eating
sweets and crisps and drinking fizzy drinks at least once a
day. Previous research has implicated the consumption of
such high levels of these food and drink types as contributors to overweight and obesity in children.11,21,22 The
poorest eating behaviours were reported at consistently
higher rates among children of low socio-economic status,
children with unemployed parents and children from minority ethnic groups. This confirms the previous reports
Table 7 Brand logo recognition scores classified by gender, deprivation score, parental employment status and ethnicity
Percentage of children with brand logo score (n)
Median brand logo
P-value
recognition score (IQR)
0 –5
Overall
6 –10
11 –15
16 –20
2.3 (11)
2.3 (11)
6.9 (33)
88.4 (420)
19.0 (2)
Male
2.8 (7)
2.8 (7)
8.3 (21)
86.2 (219)
19.0 (3)
Female
1.8 (4)
1.8 (4)
5.5 (12)
90.0 (199)
20.0 (2)
1
0 (0)
3.6 (1)
3.6 (1)
92.9 (26)
19.0 (2)
2
0 (0)
1.4 (1)
1.4 (1)
97.3 (71)
20.0 (1)
3
5.5 (4)
0 (0)
2.7 (2)
91.8 (67)
20.0 (1)
4
0.7 (1)
2.7 (4)
9.6 (14)
87.0 (127)
20.0 (2)
Unemployed
3.3 (2)
3.3 (2)
10 (6)
83.3 (50)
19.0 (3)
Employed
0.9 (3)
1.2 (4)
5.2 (17)
92.6 (300)
19.0 (2)
White
1.0 (6)
1.3 (4)
5.9 (18)
90.9 (279)
20.0 (1)
Mixed
2.8 (1)
0 (0)
2.8 (1)
94.4 (34)
20.0 (1)
Black
2.0 (1)
10.0 (5)
22.0 (11)
66.0 (33)
17.5 (5)
Asian
3.1 (2)
1.5 (1)
3.1 (2)
92.3 (60)
19.0 (3)
Other
0 (0)
8.3 (1)
8.3 (1)
83.3 (10)
19.0 (4)
—
Gender
0.002
Deprivation score
0.137
Parental employment status
0.389
Ethnicity
,0.001
364
J O U RN A L O F P U B L I C H E A LTH
Table 8 Brand logo recognition score by eating behaviours, food knowledge and preferences
Mean brand
Median brand
logo score
logo score (IQR)
P-value
Frequency of eating breakfast before school
Everyday
18.24
19.0 (2)
3– 4 days
18.17
20.0 (1)
1– 2 days
17.52
19.0 (3)
Rarely/never
17.63
20.0 (3)
Four or less
18.26
19.0 (2)
Five or more
18.09
19.0 (2)
High consumptiona
18.32
19.0 (2)
Low consumptionb
17.95
19.0 (3)
0.472
Portions of fruit and vegetables consumed per day
0.165
Crisps
0.02
Sweets
High consumption
18.16
19.0 (2)
Low consumption
18.44
19.0 (2)
0.79
Fried food
High consumption
17.98
19.0 (2)
Low consumption
18.57
19.0 (2)
0.226
Fizzy drinks
High consumption
18.34
19.0 (2)
Low consumption
18.33
19.0 (2)
0.898
Biscuits
High consumption
18.13
19.0 (2)
Low consumption
18.42
20.0 (2)
0– 5
18.27
19.0 (2)
6– 10
18.25
19.0 (2)
0– 5
15.38
16.5 (7)
6– 10
18.32
19.0 (2)
0.036
Total ‘healthy’ food types preferred
0.578
Total number of correctly identified ‘healthy’ food types
0.008
a
Item consumed at least once a day or more.
b
Item consumed less than once a day.
identifying that these children are at greatest risk of becoming obese.23
Overall the study found that most of these young children
demonstrated a good knowledge of what food/drinks were
healthy/unhealthy but nevertheless their preferences were
towards the less healthy varieties. This supports the conclusions of previous studies indicating that an absence of
nutritional knowledge is not the major explanation for the
poor diet displayed by UK children.13,24
The high brand logo recognition ability demonstrated by
the children in the study is striking and provides good evidence that children living in the UK are very aware of food
promotion by the food industry. The study’s findings of a
significantly higher brand logo recognition ability among
girls in comparison to boys have been observed by previous
studies and are believed to be a reflection of young girls’
overall better verbal ability and general awareness.25
Although the study confirmed a high brand logo recognition in the children, it did not demonstrate a close relationship between this and the children’s reported eating
behaviours, food knowledge and preferences. Interestingly,
children from black ethnic origins were identified as having
one of the poorest eating behaviours out of all the groups
of children, in spite of the lowest brand logo recognition
ability overall. It is acknowledged that this may in part have
been confounded by socio-cultural differences.
In conclusion, the study does confirm that poor eating
patterns are frequent in young children, many of whom do
FO O D A DV E RT I S I N G A N D CH IL D H O O D O B E S I T Y
not regularly eat breakfast. Additionally, the findings demonstrate an important influence of socio-economic and
parents’ employment status in young people’s food choice.
365
parent response to children’s request for food, need to be
investigated further.
Limitations of this study
What is already known on this topic
The National Diet and Nutrition Survey of young people
aged 4 –18 (2000) demonstrated that a high proportion of
young people in the UK are consuming nutritionally poor
diets that are predisposing them to the development of
obesity.17 Previous research has identified correlations
between the increasing advertising of foods with high fat,
salt or sugar content on children’s television and the rising
prevalence of obesity among children.7,8 In addition, empirical studies have demonstrated a direct relationship between
exposure to food advertising and over-consumption of food
products in children.26,27 However, the role of existing
research in determining and evaluating public policy is
restricted by limitations to their methodologies.
What this study adds
This study adds to the existing body of evidence, data that
are both valid and relevant to the present day and the UK
setting. Its major advantages over previous research lie in its
use of a UK population of children and the non-contrived
method used to assess the children’s exposure to food
advertising. The study used children’s brand logo recognition
ability as a proxy measure of children’s overall receptivity to
food advertising. The logos used in the quiz were from a
multitude of different promotional methods and mediums,
thus measuring the child’s ‘real-life’ exposure to food advertising. There are no comparable studies using such methodologies. The timeliness of this research is important given
the prioritization of childhood obesity as a public health
issue and the debate over the role of legislation to reduce
advertising as a means of tackling this issue.
The findings of the current study indicate that although
food promotion may have an influence on young children
in terms of their brand awareness, no evidence was found
to suggest a direct casual link between food advertising
and the poor diet being consumed by our children. It is
thus unlikely that the single action of regulating or restricting food advertising to children will have an immediate or
major impact on the spiralling prevalence of obesity in
children.
The study’s findings of the importance of socio-economic
status and ethnicity on a child’s food choice highlight the
areas of further research needed. The role that food adverts
play on parental selection of foodstuffs for children, and the
Measuring the effects of advertising on children is notoriously difficult. Brand awareness is just one way in which
advertising can influence children, and the design of this
study enabled this to be assessed in a naturalistic setting. It
is acknowledged that there may be other factors not considered that might have impacted on the children’s brand
logo recognition ability such as their parents’ purchasing
patterns.
One of the main limitations of the methodology used in
this study was its failure to measure and adjust accordingly
for the IQ of the children. The children’s ability to recognize
and record their answers was inevitably linked to their
reading and writing skills, and thus intelligence was a potential confounder to the results. However, steps were taken to
ensure that such effects were minimal: special-needs children
were supervised at all times by their allocated teaching assistant and all recognizable spellings of food types (including
phonetics) were accepted.
It is acknowledged that a greater range in the children’s
brand logo recognition ability may have been observed if
more logos had been used in the quiz. However, the aim
was to measure children’s receptivity to advertising; 20 logos
therefore allowed a balance between obtaining an overall
measure of receptivity to promotional media, while maximizing the completion rates.
The reliance of the study on the self-reporting of the children’s eating behaviours and food preferences is a potential
source of bias in the study design. More accurate methods
of data collection would have included food diaries and
observation studies. However, such methods are both
expensive and time-consuming and would have thus greatly
reduced the sample size. They also inevitably involve the
children’s parents into the data collection process, introducing an arguably greater source of social acceptability bias
than the children themselves.
Finally, due to time and financial constraints a convenience sampling technique was used to recruit schools into
the study. Such sampling techniques may have resulted in
a bias towards including the more highly motivated
schools with regards to healthy lifestyle initiatives in the
study. Although such effects were not controlled for in
the study design, the good overall representation of children from all ethnic groups and socio-economic classes in
the study increases the generalizability of the findings.
Furthermore, if such poor eating behaviours are found in
366
J O U RN A L O F P U B L I C H E A LTH
children from highly motivated schools, it is likely that the
current study has underestimated the size of the overall
problem.
8 Office of Communications. Childhood obesity—food advertising in
context. London: Office of Communications, 2004.
Acknowledgments
The authors would like to thank Mr Daniel Alton, Medical
Student, University of Birmingham for his assistance with
data collection and also all the children, parents and school
staff who supported the project through their involvement.
C.A.K. was responsible for the design of the study, data collection and analysis and the drafting of the manuscript.
L.M.R. supervised the design and conduction of the study
and was involved in the critical revision of the manuscript.
P.A. was involved in the conception of the study, supervision
during conduct of study and fieldwork and critical revision
of the manuscript.
Competing interests
All authors declare that there are no competing interests.
Funding
The study was conducted to fulfil the requirements of
an intercalated BMedSc Public Health and Epidemiology
degree. Claire Kopelman received financial support from the
Medical School, University of Birmingham and the Arthur
Thompson Trust. Funders did not contribute to the design,
conduction or reporting of findings.
9 Ashton D. Food advertising and childhood obesity. J R Soc Med
2005;97(2):51 – 2.
10 ISBA. Food advertising and children: position report (online).
January 2004 (cited 23 October 2004). http://www.isba.org.uk/
public_documents/food-advertsing.pdf
11 International Association of Consumer Food Organisations.
Broadcasting bad health: why food marketing to children needs to
be controlled. London: The Food Commission, 2003.
12 Office of Communications. Television advertising of food and
drink products to children—statement and further consultation
(online). 17 November 2006 (cited 24 May 2007). http://www.
ofcom.org.uk/consult/condocs/foodads_new/summary/
13 Kennedy C. Examining television as an influence on children’s
health behaviours. J Pediatr Nurs 2000;15(5):272– 81.
14 Institute of Community Health Sciences. Health of young people in
East London: the relachs study. London: Institute of Community
Health Sciences, 2001.
15 Office of the Deputy Prime Minister. Indices of deprivation
2004—summary (revised) (online). 2004 (cited 10 March 2006).
http://www.odpm.gov.uk/stellent/groups/odpm_urbanpolicy/
documents/page/odpm_urbpol_028470-01.hcsp
16 Calfas KJ, Sallis JF, Nader PR. The development of scales to
measure knowledge and preferences of diet and physical activity
behaviour in 4- to 8-year-old children. J Dev Behav Pediatr
1991;12(3):185– 90.
17 Birmingham City Council. Profile for Birmingham (online). 2003
(cited 24 September 2007). http://www.birmingham.gov.uk/Media?
MEDIA_ID=84931
18 National Statistics. Ethnicity (online). 13 February 2003 (cited 24
September 2007). http://www.statistics.gov.uk/cci/nugget.asp?id=
273
References
1 Anonymous. Selling
2002;360:959.
7 Hastings G, Stead M, McDermott L et al. Review of research on
the effects of food promotion to children (online). 22 September
2003 (cited 5 October 2005). http://www.food.gov.uk/multimedia/
pdfs/foodpromotiontochildren.pdf
to-and
selling
out-children.
Lancet
2 McLellan F. Marketing and advertising: harmful to children’s health.
Lancet 2002;360:1001.
3 The Food Commission. The Children’s Nutrition Action Plan
(online). 2001 (cited 6 September 2005). http://www.foodcomm.
org.uk/PDF%20files/childrens_Nutrition_Action_Plan.pdf
19 Birmingham City Council. Index of Multiple Deprivation 2004
(online). 2004 (cited 24 September 2007). http://www.birmingham.
gov.uk/Media?MEDIA_ID=97715
20 Gregory J. National Diet and Nutrition Survey: young people aged
4– 18 years, Volume 1—findings. London: The Stationary Office,
2000.
4 Huhman M, Potter LD, Wong FL et al. Effects of a mass media
campaign to increase physical activity among children: year-1 results
of the VERB campaign. Pediatrics 2005;116(2):e277 – 84.
21 Ludwig DS, Peterson KE, Gortmaker SL. Relation between
consumption of sugar-sweetened drinks and childhood obesity:
a prospective, observational analysis. The Lancet 2001;357:
505 – 8.
5 King R, Bickman L, Nurcombe B et al. The impact of a poster
advertising campaign in buses on young people’s awareness and
knowledge of a telephone counselling service. Health Promot J Austr
2005;16(1):74– 7.
22 Berkey CS, Rockett HRH, Field AE et al. Activity, dietary intake,
and weight changes in a longitudinal study of preadolescent and
adolescent boys and girls (online). Pediatrics 2000;105(4). http://
www.pediatrics.org/cgi/content/full/105/4/e56
6 Lovato C, Linn G, Stead LF et al. Impact of tobacco advertising
and promotion on increasing adolescent smoking behaviours.
Cochrane Database Syst Rev 2003;Issue 3:Art No. CD003439.
23 Strauss RS, Knight J. Influence of home environment on the development of obesity in children (online). Pediatrics 1999;103(6).
http://www.pediatrics.org/cgi/content/full/103/6/e85
FO O D A DV E RT I S I N G A N D CH IL D H O O D O B E S I T Y
367
24 Brownell KD, Cohen LR. Adherence to dietary regimes.2:
components of effective interventions. Behav Med 1995;20(4):
155 – 64.
26 Halford JCG, Gillespie J, Brown V et al. Effect of television advertisements for foods on food consumption in children. Appetite
2004;42:221– 5.
25 Pine KJ, Nash A. Barbie or Betty? Preschool children’s preference
for branded products and evidence for gender linked differences.
J Dev Behav Pediatr 2003;24(4):219 – 24.
27 Halford JCG, Boyland E, Hughes G et al. Beyond-brand effect
of television (TV) food advertisements/commercials on caloric intake
and food choice of 5–7-year-old children. Appetite 2007;49:263– 7.