This paper was presented at the Association for Education in Journalism and
Mass Communication in Toronto, Canada, August 2004.
        If you have questions about this paper, please contact the author
directly. If you have questions about the archives, email
[log in to unmask] For an explanation of the subject line, send email to
[log in to unmask] with just the four words, "get help info aejmc," in the
body (drop the "").
(Oct 2004)
Thank you.
Elliott Parker
************************************************************************

Food For Thought: A Content Analysis of Food Advertising during Prime-Time
Television


Wei-Na Lee
Associate Professor
The University of Texas at Austin
Department of Advertising
1 University Station A1200
Austin, TX 78712-1092
Tel. 512-471-8149
<[log in to unmask]>


Eliana Shiao Tseng
Doctoral Student
The University of Texas at Austin
Department of Advertising
1 University Station A1200
Austin, TX 78712-1092
Tel. 512-471-1101
<[log in to unmask]>


Sejung Marina Choi
Assistant Professor
The University of Texas at Austin
Department of Advertising
1 University Station A1200
Austin, TX 78712-1092
512-471-3359
<[log in to unmask]>


Key words: food advertising, nutrition content

Paper submitted to the 2004 Annual Convention of the Association for
Education in Journalism and Mass Communication

The authors would like to acknowledge the help of Alan Magee in the
completion of this project.
 Food For Thought: A Content Analysis of Food Advertising during Prime-Time
Television

Abstract
Today almost two-thirds of American adults are classified as overweight.
While there are several known causes of obesity, food advertising on
television has received a significant amount of criticism for its role in
fueling the rise in obesity. The study reported in this paper examined the
amount and general characteristics of food advertising and the actual
nutritional content of the products advertised during prime-time
programming on major networks. Results show that food advertising composed
one quarter of the overall advertising during prime-time television. In
general, food products advertised in prime-time television were unhealthy
according to their actual nutritional content. Meanwhile, not surprisingly,
the advertising messages for these products tended to employ
taste/flavor/smell as the key promotional appeal instead of nutritional
value. Implications of observations made in this study and suggestions for
future research are provided.



 Introduction
Obesity in America has recently become one of the most popular topics
discussed in research studies and public media outlets (Hellmich 2003). The
coverage of this issue is well worth the exposure it receives. Today,
almost two-thirds (64.5%) of American adults, age 20 years and older are
overweight (American Obesity Association 2004). According to the Journal of
American Medical Association, 30.5% of Americans are obese, up 7.6% from
1992 (Raeburn, Forster, Foust, and Brady 2002). This increase in
unhealthiness is alarming and it has brought the issue of obesity to the
forefront of American popular press (Hellmich 2003).
This great concern about the rise in obesity is due primarily to the fact
that many adverse health effects have been associated with obesity.
According to the American Obesity Association (2004), obese individuals are
at risk of developing arthritis, cancer, cardiovascular disease, diabetes,
urinary stress incontinence, among several other severe health conditions.
These factors together could lead to an increase of deaths from obese
effects, which already numbered 300,000 in 2002, and is 180,000 cases more
than alcohol and drug abuse deaths combined (Grimm 2003).
Genetics, behavior, and environment have been identified as the key factors
contributing to the rise of obesity in America (Center for Disease Control
and Prevention 2004). Among the behavioral factors, the lack of physical
activity and poor eating habits are considered the main culprits. With the
advancement of technology and the proliferation of media options, Americans
are spending more time sitting in front of a computer/TV screen rather than
engaging in physical activities. It has been documented that an average
American adult these days watches between 4 and 4.5 hours of television per
day (Horgen, Choate and Brownell 2002). This tendency to live a sedentary
lifestyle, coupled with increased food portion size and availability of
snacks high in salt, sugar, and calories, are said to have produced an
increasing number of obese adults.
Among the environmental factors, food advertising on television has
received the most criticism for its role in fueling the rise in obesity
(Hellmich 2003). Such criticism is no coincidence, as food advertising is a
major player in national television. In 2002, food advertising represented
a little over $ 6 billion in ad expenditure, with over $ 2 billion on
network television alone. These numbers are nearly double the amount of
restaurants ad expenditure ($3.7 billion in total ad expenditures and $1.2
billion on television, respectively). Together, the combined expenditure of
food and restaurants advertising add up to close to $ 10 billion, which is
more than half the amount of advertising expenditure on the automobile
industry ($ 16.3 billion), the top domestic advertiser  (AdAge.com 2004).
Research by scholars and health experts have thus far produced divergent
conclusions and further propelled obesity into a much debated national
health issue. The goal of this study is to provide a baseline assessment of
the nature of food advertising and set the stage for future research in
understanding the relationship between food advertising and obesity.
Specifically, the following research questions are addressed in the study:
What is the amount of food advertising for the general population during
prime-time television? What proportion of food ads is for "unhealthy food"
versus for other types of food? Among those deemed as "unhealthy food" ads,
what types of executional elements are employed? A two-stage content
analysis of prime-time television commercials of the four major networks
(ABC, CBS, NBC, and FOX) during one week in March of 2003 was carried out
to answer the above research questions.

Background Information
        The Center for Disease Control and Prevention (CDC) defines obesity as
resultant from energy imbalance between the number of calories consumed and
the number of calories used. In other words, overweight and obesity involve
not getting enough exercise while taking in too many calories (Center for
Disease Control and Prevention 2004).  Currently, roughly two-thirds
(64.5%) of American adults, age 20 years and older are overweight, 30.5%
are obese, and 4.7% are severely obese. These percentages reflect an
increase from the 1988-1994 National Health and Nutrition Examination
Survey (NHANES) by the CDC, which reported that 56% of U.S. adults were
overweight, 23% obese and 2.9% severely obese (American Obesity Association
2004). These numbers suggest that there is a rising trend of obesity in the
country.
The American Obesity Association have identified several causes to obesity,
such as lack of physical activity, frequent intake of high caloric food, an
abundance of larger than recommended serving size of today's food, among
others (American Obesity Association 2004). Among these causes, sedentary
lifestyle as a result of modernization is considered a major factor that
contributes to the rise of obesity. With the advancement of technology,
there is an increase of inactivity at home and at work. The computerization
of the workplace has reduced the demand for physical activity during
working hours. While at home, the use of modern appliances, such as
dishwashers and microwaves also reduces the amount of manual labor.
Modernization has also led the population to rely increasingly more on
non-traditional foods, such as ready-made or convenience foods, which also
contribute to obesity (American Obesity Association 2004). Another related
factor is the increased serving size of food today compared to the past.
Larger serving sizes contain more calories, which result in consumers
exceeding the average daily serving of calories on a constant basis. For
example, muffins today weigh almost a half a pound each, compared to 1.5
ounces on average in the 1950's (Raeburn et al. 2002).
Some think that another driving force of the obesity problem is the
increase in television viewing combined with the accessibility to
entertainment. The access to hundreds of television channels and the rise
of DVD movies has all been suggested as possible links to the decrease in
physical activity (Raeborn et al. 2002). According to a 2003 survey by the
Television Bureau of Advertising (2003), 79.8% of the American adults
considered television the most exciting medium while 81.8% found it the
most influential.
There is a general assumption among some communication scholars that "The
sheer volume of viewing hours suggests that television can have a powerful
influence on our perception and behavior" (Powers 1996, p. 9). As the
amount of television viewing time increases, it is suggested, so does the
influence of its content (Schneider 2003). A content analysis of
commercials from 1992 and 1998 by Byrd-Bredbenner and Grasso found that
soft drinks were advertised frequently. Furthermore, the intake of soft
drinks had an inversely related relationship with the intake of other
products such as milk, fruits, and vegetables which, while important
dietary supplements, were almost never advertised (Byrd-Bredbenner and
Grasso 2000).
While food advertising expenditures have risen steadily during the past 10
years to over $ 6 billion in 2002, the diversification of food being
advertised has also increased. A closer scrutiny indicates that the top
foods being advertised are not necessarily health conscious. Less
advertising expenditures are being spent on advertising for foods that are
better for the overall health (Goode 2003). According to TNS Media
Intelligence/CMR, in 2002 the food industry spent more than $750 million on
soft-drink ads, $570 million on cereals and $545 million on candy and
mints. At the same time, the industry spent about $51 million on fruit and
$ 47 million on vegetables advertising (Martin 2004). In terms of fast food
chains, the target of many recent public blames, McDonald's spent $465
million in advertising during the first three quarters of 2003, compared
with $238 million from Wendy's and $221 million from Burger King (Martin
2004).
Some studies have agreeably shown that commercials placed during prime-time
television programming promote non-nutritious food and that there were very
few advertisements for healthy foods such as fruits and vegetables (Powers
1996). In their longitudinal study, Francis and Birch showed that
advertising directly influenced increased snacking on high fat and sugary
foods (2002). Furthermore, Hitchings and Moynihan (1998) discovered that
the food ads that children recalled correlated with the types of foods that
they consumed. Although Bolton (1983) found that the long term effects of
food advertising to be minimal, she did conclude that food advertisements
have short term effects on dietary influences.
         Critics of this reasoning, such as Lvovoich (2003), believe that obesity
is such a multi-faceted problem with numerous factors that there is not
really a direct relationship between television food advertisements and
obesity. He suggests the lack of exercise that is missed from time spent
watching television as a stronger link to obesity rather than the food
commercials viewed while watching television (2003). In addition, Powers
states that food types eaten in the television shows themselves have a
stronger effect than ads for food products in adolescents (1996). Young
(2003) postulates that advertising affects consumption only at the brand
level but not the product category level. This is because people already
know from what food product category they are going to consume.
Consequently, it is argued that advertising affects only the choice of
brands within those pre-selected categories.
        Young asserts that advertising is only a piece of the puzzle and the rest
of the factors causing obesity need to be looked at in closer retrospect.
He states that (p. 8, 2003):
"…the route from advertising to obesity is a torturous one – from
advertising being a dietary influence of sorts, and diet itself being only
one element in the balance between energy in and energy out. Yet again
advertising seems to have been effectively scapegoated as the villain of
the piece."
Although food advertising has received a great deal of attention from the
media, much of the focus is on the role of advertising in childhood obesity
(Kaiser Family Foundation 2004). However, studies have shown that increases
in the availability of unhealthy food to the adult population, combined
with easy access to a wide array of entertainment media may also contribute
to the rise of adult obesity in America (Jeffery and French 1998). A
careful inspection reveals that food advertising expenditure has been
disproportionately concentrated on television (Ippolito and Pappalardo
2002). Meanwhile, 81.8% of the adult population considers television as the
most influential advertising medium (Television Bureau of Advertising
2003). All of these factors combined make it necessary to have an objective
assessment of the amount and nature of food advertising on television that
is targeted at the adult population. The outcome of this assessment will
then serve as a baseline reference for future studies on the role of food
advertising in obesity. Therefore, a content analysis study was designed
and carried out as a first step toward this goal.
The Study
This study seeks to examine the amount, general characteristics and
nutritional information of food advertising during general population
programming on major networks. The sample of ads was obtained from tapings
of prime-time (7-10pm Central Time) television on four major networks (ABC,
CBS, NBC, and FOX) during the week of March 3 to March 7, 2003. These
networks and time slots were chosen because of the large viewing audience,
the vast array of programming, and their availability to population nationwide.
        A total of 60 hours of prime-time television programming and commercials
from four major networks were taped. Out of the 60 hours of television, a
total of 1,379 commercials were obtained. Two coders coded the ads
independently and were instructed to discharge incomplete spots, i.e., the
first ad if the tape started with an ad and the last ad if the tape ended
with an ad. Coders also excluded the coding of announcements of other
television programs, tie-in promotions or give-aways related to the
channel, public service, and promotion of movies.
The coding of the ads occurred in two stages. During the first stage of
coding, the main goal was to obtain information on the presence of the ads
from prime-time television. Therefore, duplication was not removed from the
data so that all occurrences of the product categories analyzed could be
considered. During this stage, each ad was coded for product category,
brand name, network and associated program type.
During the second stage of the study, only food related ads were selected
as the sample and coded. The category of food ads included drinks (juice,
coffee, etc.), food products (meat, fruits, microwave foods, canned goods,
etc.), gum and snacks (candies, cookies, etc.), fast food (Mc Donald's,
Burger King, etc.) and family restaurants (Olive Garden, Johnny Carinos,
etc). These food categories were adapted from Kunkel and Gantz's study
(1992). Except restaurants, all other food related products (food, drink
and snacks) in the sample were then coded for executional elements and
actual nutritional content (ANC).
In coding the executional elements, each ad was assigned a main promotional
theme that best represented the communication strategy used to persuade the
target audience: fun/happiness (positive association provided by the use of
the product), taste/flavor/smell (any association related to taste or
flavor of the product), appearance (improved appearance as the main reason
for having the product), new/innovative (introduction of a new product or
flavor), unique (product differentiation), quantity/size/amount (product
packaging), quality, health/nutrition (description of the nutrition related
features of the product), personal gain (any additional personal
satisfaction for having the product), convenience (improvement of speed
while consuming the product), texture (how it feels when using the
product), product in social context (social situations surrounding the use
of the product, such as peer status, popularity, etc.), and other.
Categories of themes employed in this study were adapted from previous
content analysis investigations (Kunkel and Gantz, 1992; Alexander,
Benjamin, Hoerrner, and Roe, 1998).
Ads were also coded according to their use of disclaimers, humans, settings
and character role. Appendix A details how food related ads were further
coded. Intercoder reliability was calculated based on Perreault and Leigh
formula (1989), also shown in the Appendix. An overlap of coded ads between
the two coders was used in the calculation. The average intercoder
reliability among all coding categories was 0.87, ranging from 0.74 to 1.0.
Coding disagreements were discussed and resolved before analysis continued.
Next, the goal of the study was to determine the amount of healthy versus
unhealthy products advertised on television. In order to do so, each ad
needed to be judged based on an objective standard. Therefore, the
guidelines established by Kuribayashi, Roberts and Johnson (2001) were
adopted. According to this set of guidelines, a food product was considered
unhealthy if it fell in one or more of the following criteria: (1) more
than 30% of its calories comes from fat per serving, (2) it contains more
than 360 mg of sodium per serving, (3) it contains more than 35mg of
cholesterol per serving, or (4) more than one third of its calories per
serving come from sugar.
In order to code the ANC of each product, coders visited grocery stores and
recorded the nutritional information printed on each product's label. For
products that were advertised with more than one flavor or variety without
any indication of specific flavor, an average ANC was calculated. This
guideline was adapted from Kuribayashi et al. (2001) in an effort to obtain
better accuracy of the nutritional information of all the products appeared
in TV commercials. The codesheet for ANC can be found in Appendix B.
Results and Discussion
The Amount of Food Advertising
        Out of the total number of 1,379 ads analyzed, 25.8% of them were of food
related ads versus 74.2 % of ads related to all other product categories.
Fast food restaurants had the highest presence at 40%, followed by an
almost equal number of food products (69 or 19.4%) and gum and snacks (67
or 18.9%). Family restaurants and drinks had 44 (12.4%) and 33 (9.3%) ads
respectively. This finding is consistent with past research that shows a
significant presence of food advertising on television. Furthermore, fast
food restaurants were found to have the highest count, among all the
sub-categories. This may help explain why the fast food industry has been
the target of finger-pointing by the public. Table 1 below presents the
detail information on the overall distribution of the ads.
Table 1
Food Products Advertised During Prime-Time Programming

Frequency
Percent
Food Related Ads
355
25.8
Non-Food Ads
1024
74.2
Total Ads
1379
100
Food Ads
Frequency
Percent
Fast Food Restaurants
142
40.0
Food Products
69
19.4
Gum and Snacks
67
18.9
Family Restaurants
44
12.4
Drinks
33
9.3
Total Ads
355
100

        Among those 355 food related ads, roughly one third (115 ads) was unique
ads. Of those unique food ads, 32 (27.8%) of them were for fast-food
restaurants, 31 (27%) for gum and snacks, 30 (26.1%) for food products, 14
(12.2%) for drinks, and 8 (7%) for family restaurants. In the drinks
category, half of the ads were for artificial drinks such as sodas, which
was equivalent to 6.1% (7) of all the unique food ads. Processed drinks,
such as liquid yogurts, caffeinated drinks, and coffee derivatives were
each represented at 1.7% of all unique food ads (2 each). Natural fruit
juice were the least advertised drink type, with 0.9% (1) of all unique
food ads. In terms of food products, cereals and breakfast (15) composed
half of the ads in that category and 13% of all the unique food ads,
followed by microwave or convenience foods, with 11.3% (13) of the total
number of unique food ads. In the gum and snacks category, cookies and
chips composed 14.8% (17) of the unique food ads, while gum and candies
were at 10.4% (12). Fast food restaurants and family restaurants composed
27.8% (32) and 7.0% (8) of the unique food ads respectively. Restaurants
were excluded from the remainder of the analysis since they were
advertising for food outlets rather than for a specific food product. Thus,
out of the 115 unique food ads, 75 ads were further analyzed according to
their executional elements and ANC. Table 2 provides the frequency
distribution of unique ads.
Table 2
Unique Ads Frequency Distribution

Frequency
Percent

Fast food restaurants
32
27.8
Gum and snacks
31
27.0
Food products
30
26.1

Drinks
14
12.2

Family restaurants
8
7.0

Total
115
100.0




Executional Elements of Food Ads
Among the sub-sample of 75 unique food ads, the most commonly employed
theme was of taste/flavor/smell at 36% (27), followed by new/innovative (12
or 16%) and health/nutrition (11 or 14.7%). Table 3 presents the detailed
breakdown of themes used in those ads. Past research has shown that there
is a relationship between the theme/appeal used and the type of product
advertised (Kunkel and Gantz 1992). Consistent with the literature, this
study found that taste/flavor/smell was by far the primary theme employed
in the promotion of food products (16 or 53.3% within the category).
An interesting finding concerning the themes was the use of new/innovative
appeal. While it was the second most employed promotional appeal in the
sample (16%), it was the most used promotional theme within the category of
gum and snacks (9 or 29% within the category), suggesting a product line
extension or even a product differentiation promotional strategy for that
category. For example, the main theme of Dentyne Ice ads referred to the
introduction of a new flavor of the product.
Another interesting finding on the use of themes was that
quantity/size/amount was scarcely used. Literature has suggested that,
compared to the past, the increased serving size of food today might be one
of the causes of obesity. It is therefore worth noting that, as an appeal,
quantity/size/amount was not reflected in the content of the food ads. It
is possible that such a trend could occur in restaurant ads where
quantity/size/amount might be used as a competitive advantage.
Table 3
Themes Used in Food Related Product Advertisements