Food Advertising Targeted at School-Age Children: A Content Analysis

RESEARCH BRIEF

Food Advertising Targeted at School-Age Children: A Content Analysis Sara C. Folta, PhD1, Jeanne P. Goldberg, PhD, RD1, Christina Economos, PhD1, Rick Bell, ScD1,2, Rachel Meltzer, MS1 ABSTRACT Objective: To determine whether the contents of food and beverage advertisements are associated with physical activity and athletic ability more often than those for toys and games, and to describe persuasive techniques used in advertising food and beverages to children. Design: A content analysis of advertisements during 31 hours of school-age children’s television programming. Analysis: Chi-square tests were used to examine differences in depictions of physical activity. Types of persuasive techniques were tabulated and, within each advertisement, categorized as implicit or explicit. Results: Food and beverage ads depicted children engaged in physical activity and associated the advertised product with athletic ability significantly more than toy and game ads. Food was most often associated with fun and good times (75%), pleasant taste (54.1%), being hip or cool (43.2%), and feelings of happiness (43.2%). Implications for Research and Practice: These findings raise concern that greater levels of physical activity and athletic ability in food advertising, in which the product is frequently associated with fun, may promote overconsumption, especially of calorie-dense, nutrient-poor foods. Further research would elucidate whether this concern is warranted. On the other hand, since food advertisements are presumably effective, health educators can use these techniques to formulate messages for nutritious foods. This concept should be tested with well-designed interventions. Key Words: childhood obesity, food, advertising (J Nutr Educ Behav. 2006;38:244-248)

INTRODUCTION Concern that young children cannot comprehend television advertising and that it could have undue manipulative effects on them was first raised in the 1970s.1 Advertising targeting children is emerging once again as a public policy issue, this time in the context of the childhood obesity epidemic. Many studies have shown that children who spend more time with media, particularly television, are more likely to be overweight.2 Television viewing may displace less sedentary activities3 or lead to increased energy intake, since the consumption of greater amounts of food overall, and specifically high-fat and high-sugar foods, may be en1 Gerald J. and Dorothy R. Friedman School of Nutrition Science and Policy, Tufts University, Boston, Mass 2 Natick Research Center, Natick, Mass.

Author for correspondence: Sara C. Folta, PhD, Gerald J. and Dorothy R. Friedman School of Nutrition Science and Policy, Tufts University, 150 Harrison Ave, Boston, MA 02111; Phone: (617) 636-3423; Fax: (617) 636-3781; E-mail: [email protected]

©2006 SOCIETY FOR NUTRITION EDUCATION doi: 10.1016/j.jneb.2006.04.146

couraged through advertising.4 Many studies indicate that advertising influences children’s food preferences and consumption patterns.5 The World Health Organization considers food marketing aimed at children to be a probable causative factor contributing to the obesity epidemic.6 Even 30-second messages have been shown to influence children’s food preferences.7 Our first goal in undertaking a content analysis of television advertising targeting school-age children was to determine whether physical activity and athletic ability were associated with the product more often in food ads than in those for toys and games. Advocacy groups and academics have suggested that the food industry deflects attention from its possible role in the obesity epidemic by emphasizing physical activity.8,9 Others have suggested that associating physical activity and athletics with nutrientpoor foods may mislead children to believe that these foods are healthful.10 However, to our knowledge, the association between physical activity and food advertising has never been quantified. To explore these claims, we compared the frequency of moderate to vigorous physical activity in food

Journal of Nutrition Education and Behavior ● Volume 38, Number 4, July/August 2006

ads with that in toy and game ads. The logic is that to play with a toy or game is no more likely to involve physical activity than eating or drinking. Our second goal was to explore the nature of the persuasive techniques used in advertising to children. Previous research has indicated that advertising for food products is most often associated with fun and happiness.1,11 However, very few content analyses of children’s television advertising in the United States have been published since the early 1990s. Of 5 studies identified that were published since 1995,4,11-14 only one examines the persuasive techniques that were used.11 More in-depth and current research is warranted to determine which persuasive techniques to sell food to children have persisted since the most recent analyses were conducted; to examine a broader range of persuasive techniques used to sell food; and to identify any new strategies that may have emerged in recent years. Although we do not have direct evidence of the effectiveness of the advertisements evaluated in the present study, we do know that advertising to children does influence their choices.5 For that reason, health educators may want to adopt some of the strategies used by the commercial sector to promote more healthful foods. This information may also help inform media literacy training to enable children to critically evaluate media messages to understand their implicit purpose and impact. Finally, as the debate continues about whether the United States should implement policies that limit or change the nature of food advertising targeting children, as has been done in other countries, a systematic study of advertising would help inform policy decisions.

STUDY PROCEDURES A total of 31 hours of children’s programming was videotaped during one week in late September 2003 in the metropolitan Boston area, on stations and time slots that were most popular with children, according to Nielsen data.15,16 Children’s programming is increasingly found on cable rather than network television,17 and in the United States, most households with children have cable television.18 Therefore, two cable stations (Nickelodeon and The Cartoon Network) and one network station (WB) were chosen. Time slots included weekdays from 7:00-10:00 AM and 3:00-6:00 PM, and Saturday mornings from 7:00 AM until 1:00 PM. Tuesday, Wednesday, and Thursday were chosen as representative weekdays. For each weekday morning and afternoon time slot, one of the three stations was taped. A random procedure was used to determine which station would be taped on which day. All three stations were taped on Saturday. On WB, children’s programming went only until 5:00 PM on weekdays and to 12:00 PM on Saturdays, and there was no children’s programming in the morning time slot. Two coders (the first and last authors) pretested the initial coding instrument using a sample of the advertise-

245

ments, and then refined it. To determine the final interrater reliability, all variables were coded by both coders for a sample of 57 food, toy, and game ads viewed during one of the time slots. This number represented 47% of the unique food, toy, and game ads viewed during the entire 31 hours. Variables that did not achieve inter-rater reliability (Cohen’s kappa19 or Spearman’s rho) of 0.8 were excluded from final analysis. Inter-rater reliability statistics were calculated using Pram software (version 0.4.5, Skymeg Software, www.geocities.com/skymegsoftware/pram.html). A third coder, blinded to the hypotheses, coded all variables that compared food and toy ads. Inter-rater reliability for all three coders was acceptable (⬎0.8) for these variables. For this study, physical activity was defined as an activity done by a child or childlike character who would raise his or her heart rate by, for example, running, jumping, or playing a sport. A product was considered to be associated with increased athletic ability if consuming or using it resulted in improved physical performance. A list of persuasive techniques used in advertising to children was generated from a literature search of previous content analyses, and a preliminary coding instrument was developed. The instrument was expanded and refined in an iterative process by coding ads observed in 5 hours of children’s programming recorded a month prior to the actual data collection period. Persuasive techniques were further coded according to whether the associations were explicit or implicit. An association was considered explicit if it was stated explicitly (“if you eat this, you will be stronger”) or if using or consuming the product directly resulted in the characteristic (the child eats the cereal and is able to perform an athletic feat that he or she could not do immediately before). An association was considered implicit if words, images, or actions alluded to the association, but it was not stated explicitly20 or shown. Only advertisements deemed to be targeted toward school-age children, ages 6-12 (rather than preschool children, adolescents, or adults), were used in the final analysis of the ads. Ads were considered to be targeted toward preschool children because they were slower paced and more repetitive, and consistently featured preschool children. Ads targeting adolescents included more “adult” themes (such as romance) and featured older adolescents or adults in their early 20s. To test hypotheses related to physical activity in food ads compared to toy and game ads, chi-square statistics were used.

FINDINGS There were a total of 987 advertisements and promotions in the 31 hours of children’s programming, or an average of approximately 32 per hour. Because ads and promotions were 15 to 30 seconds each, this number represents approximately 12 minutes of nonprogram content per hour. Pro-

246

Folta et al/FOOD ADVERTISING TARGETED AT SCHOOL-AGE CHILDREN: A CONTENT ANALYSIS

motions, defined as announcements for upcoming shows, movies, concerts, and contests, accounted for 28% of the 987 advertisements. Excluding promotions, there were 711 ads, of which 39% were for toys and games, and 35% were for food and restaurants (Table 1). Of the 711 total, 183 unique advertisements were shown. Therefore, on average, ads were repeated approximately 4 times each; however, the number of times an ad was repeated was highly variable and ranged from 1 to 18 times during the taping period. Of the 183 unique ads, 121 were for food, toys, or games. Of these, 100 were targeted toward school-age children. Breakfast cereals were the most heavily advertised foods (10 ads of 37, or 27% of unique food ads) (Table 2). We calculated percentage sugar content of all the cereal advertised as the ratio of grams of sugars to grams of product per serving. Based on this calculation, all cereals ranged between 33-47% sugar by weight. According to the ingredients list, virtually all sugars were added (rather than natural). Restaurant meals (such as McDonald’s Happy Meal™) were the next most prevalent (19%) category of ads, followed by sweet snacks and desserts (13%), other breakfast foods (11%), and juice drinks (11%). Of the 4 juice drinks advertised, 3 contained 10% real fruit juice and one contained none. Dairy products and restaurants each accounted for 8% of the foods advertised; 3% of the ads were for savory snacks. There were no ads for fruits or vegetables or for carbonated beverages during the sample period.

Physical Activity In Food Versus Toy and Game Advertising Our first hypothesis was that physical activity would be depicted more often in food ads compared to toy and game ads. Eighteen of 37 (48.6%) food ads showed some type of physical activity by a child or child-like animated character that would raise his or her heart rate, whereas only 13 of 63 (20.6%) toy and game ads depicted this type of physical

TABLE 1. Types of Advertisements Shown during 31 Hours of Children’s Programming (N ⫽ 711)

Type of Advertisement Toys/games Food/restaurant CD/DVD/Video Services Home/bath products Electronics Web site PSA Local ad Clothes/shoes Total

N 274 247 63 29 28 20 15 13 8 6 711

% of Total Advertisements 39 35 9 4 4 3 2 2 1 1 100

TABLE 2. Types of Products Shown in Food/Restaurant Advertisements targeted at School-Age Children (N ⫽ 37)

Type of Food Product Breakfast cereal Restaurant, specific meal Sweet snack/dessert Other breakfast food (waffles, toaster pastries) Juice drinks Dairy Restaurant, no specific meal Savory snack Total

N 10 7 5 4

% of Total Food Advertisements 27 19 13 11

4 3 3 1 37

11 8 8 3 100

activity. The difference was significant (␹2 ⫽ 8.55, P ⬍ .007). Our second hypothesis was that products would be associated with increased athletic ability more often in food ads than in toy and game ads. The product was associated with increased athletic ability in 13 of 37 (35.1%) food ads, and in only 8 of 63 (12.7%) toy and game ads. This was also a significant difference (␹2 ⫽ 7.07, P ⬍ .01).

Persuasive Techniques Used In Food Advertisements All 37 food ads were analyzed for the persuasive techniques used. The foods being advertised were most often associated with fun and good times, pleasant taste, being hip or cool, and feelings of happiness (Table 3). They were also associated with toys being given away with the product, athletic ability, innovation or newness, friendship or social success, magical or superhuman abilities, and deceiving or tricking adults. Less often, food products were associated with convenience (10.8%; N ⫽ 4); deceiving or tricking older kids or siblings (10.8%; N ⫽ 4); turning a bad situation into a good one (5.4%; N ⫽ 2); fun or interesting packaging (5.4%; N ⫽ 2); and nutrition or healthfulness (5.4%; N ⫽ 2). Physical attractiveness, physical comfort, annoying a younger sibling, being able to play on fun equipment, being able to play with the food, and love were each associated with the product in only one ad. No ads associated the product with either academic success or pleasing a parent. In most cases, the association with the product was implied. Associations were more likely to be explicit when the message included deceiving or tricking adults, a pleasant taste, or offers for free toys. When the product was associated with deceiving or tricking older kids or siblings, half of the associations were implicit and half were explicit.

DISCUSSION This study confirms the anecdotal observation that food advertisers link their products with physical activity. Phys-

Journal of Nutrition Education and Behavior ● Volume 38, Number 4, July/August 2006

TABLE 3. Ten Most Common Qualities with which Food Advertisers Associate their Products

Product Associated With: Fun/good times

N 28

Pleasant taste

20

Being hip or cool Feelings of happiness

16 16

Toys being given away with product Athletic ability

13

Innovation/newness Friendship/social success

12 11

13

Magical/superhuman abilities 10 Fooling/getting the better of adults

8

% Assoc. Implicit or With Explicit? 75.7 Implicit–82% Explicit–18% 54.1 Implicit–45% Explicit–55% 43.2 Implicit–100% 43.2 Implicit–75% Explicit–25% 35.1 Implicit–8% Explicit–92% 35.1 Implicit–22% Explicit–78% 32.4 * 29.7 Implicit–90.9 Explicit–9.1 27 Implicit–90% Explicit–10% 21.6 Implicit–37.5% Explicit–62.5%

*Did not achieve satisfactory inter-rater reliability.

ical activity was depicted in one of every two food ads, but in just one of every five toy or game ads. Likewise, food was associated with increased athletic ability in more than one in three food ads, but in just one in eight toy and game ads. These findings raise several concerns. At best, these portrayals of physical activity could encourage children to become more active. At worst, advertisers could be using them to divert the emphasis on food as a contributor to obesity. Further study is warranted to determine children’s perceptions of this relationship and of the effect, either positive or negative, that these ads may have on behavior. Focus groups in which children view and respond to the ads would help to answer this question. Another approach would be to conduct copy testing research, which is often used by the Federal Trade Commission to help determine if advertising is deceptive.21 This type of research involves exposing a representative sample of the target population to advertisements and then surveying the sample to determine their perceptions of the ad. We found that food advertising accounted for 35% of all advertisements shown. Past studies have found that food ads account for over 50% of all ads targeted toward children.14 A number of factors could explain the discrepancy. Food advertisers may be reducing the number of food ads targeted toward children in response to increasing pressure. Many of the studies that showed a higher percentage of food ads were done before CDs, DVDs, video, electronics, and the Web were commonplace. It is also possible that given a limited number of advertising minutes per hour, these newer items have “crowded out” food, toy, and game

247

advertising to some extent over the past decade. Finally, this discrepancy may reflect the time of year that was sampled. Sampling occurred in early fall (September), and a number of other studies have found that toy advertising increases significantly in the “pre-Christmas” season.22 It is possible that this increase had already begun when sampling took place. The most common persuasive technique that advertisers used was to associate their product with fun and good times. Apparently, this has been the most popular technique since the early days of television.11 Foods are also commonly associated with a pleasant taste. This finding is not surprising, considering that taste is a key determinant of food choice.23 A limitation of this study is that sampling took place only in the fall. However, this time period is reasonably “typical” in that it was not during or near any programming, such as sweeps week, which would have a major effect on advertising. It is possible, as mentioned above, that toy advertising was higher in this sample due to its proximity to the pre-Christmas season. Another limitation is that only three television stations were sampled, one of which had no programming for children in the morning. However, the stations were chosen because they were rated as the most popular with children, based on Nielsen media data.

IMPLICATIONS FOR RESEARCH AND PRACTICE Health educators can use some of the persuasive techniques identified in this study to formulate messages for nutritious foods targeted to school-age children. For example, based on these findings, a public service announcement (PSA) that promotes a healthful food such as broccoli to schoolage children would associate the food with fun and happiness, and would imply that it tastes good. Well-designed interventions to test whether these techniques will work to increase consumption of more desirable foods are warranted. Educators can also use this information to help design media literacy training for children. For example, such training could help make children aware that food is often associated with fun and happiness in order to make it seem more desirable. Older children have the ability to critically evaluate these types of associations once they are aware of them. Although a nutrient analysis of the advertised foods was not conducted, our study corroborates earlier findings that advertised foods are poor in nutritional quality.4 The foods advertised were high in calories, fat, and/or sugar. There were no ads for fruits or vegetables. The net result is that children, who watch between 20,000 and 40,000 advertisements per year,5 are probably getting a skewed picture of foods and diets. A number of studies have established the link between exposure to television advertising and food preference and choice. The link between preference,

248

Folta et al/FOOD ADVERTISING TARGETED AT SCHOOL-AGE CHILDREN: A CONTENT ANALYSIS

choice, and obesity has yet to be explored. Studies in which food advertising is removed from television programming to which children are usually exposed in a natural environment will help clarify its specific role in the etiology of obesity. There has been increasing debate about whether limits should be placed on food advertising targeted toward children. Just as it was inappropriate for cigarette advertisers to portray smoking in the context of a fun, active lifestyle, it may be inappropriate for food advertisers to portray less healthful foods in this context, especially in those that are targeted toward children. Therefore, it seems reasonable to consider limitations on food advertising targeted toward children.

9. 10.

11.

12.

13.

14.

15.

REFERENCES 1. Kunkel D, Gantz W. Children’s television advertising in the multichannel environment. J Commun. 1992;42(3):134-152. 2. Caroli M, Argentieri L, Cardone M, Masi A. Role of television in childhood obesity prevention. Int J Obes. 2004;28:S104-S108. 3. Epstein LH, Paluch RA, Consalvi A, Riordan K, Scholl T. Effects of manipulating sedentary behavior on physical activity and food intake. J Pediatr. 2002;140(3):334-339. 4. Gamble M, Cotugna N. A quarter century of TV food advertising targeted at children. Am J Health Behav. 1999;23(4):261-267. 5. Story M, French S. Food advertising and marketing directed at children and adolescents in the US. Int J Behav Nutr Phys Act. 2004;1(3). Available at: http://www.ijbnpa.org/content/1/1/3. 6. Report of a Joint WHO/FAO Expert Consultation. Diet, Nutrition and the Prevention of Chronic Diseases. Geneva: World Health Organization; 2003. 7. Borzekowski DLG, Robinson TN. The 30-second effect: An experiment revealing the impact of television commercials on food preferences of preschoolers. J Am Diet Assoc. 2001;101:42-46. 8. Dalmeny K, Hanna E, Lobstein T. Broadcasting Bad Health: Why Food Marketing to Children Needs to be Controlled. Washington, DC: The

16. 17.

18.

19. 20.

21. 22.

23.

International Association of Consumer Food Organizations; July 2003. Schor J. Born to Buy. New York: Scribner; 2004. Center for Science in the Public Interest. Pestering Parents: How Food Companies Market Obesity to Children. Washington, DC: Center for Science in the Public Interest; November 2003. Alison A, Benjamin L, Hoerrner K, Roe D. “We’ll be back in a moment”: A content analysis of advertising in children’s television in the 1950s. J Advertising. 1998;27(3):1-9. Taras H, Zive M, Nader P, Berry CC, Hoy T, Boyd C. Television advertising and classes of food products consumed in a paediatric population. Int J Advertising. 2000;19(4):487-494. Kuribayashi A, Roberts M, Johnson R. Actual nutritional information of products advertised to children and adults on Saturday. Children’s Health Care. 2001;30(4):309-322. Byrd-Bredbenner C. Saturday morning children’s television advertising: A longitudinal content analysis. Fam Consum Sci Res J. 2002; 30(3):382-403. Nielsen Media Research. 2000 Report on Television. New York: Nielsen Media Research; 2000. Friedman W. The Biz: WB, Nickelodeon lead hot kid ratings. Advert Age. 2002;73(50):19. Jordan A, Jamieson K. The State of Children’s Television: An Examination of Quantity, Quality, and Industry Beliefs. Philadelphia, Pa.: The Annenberg Public Policy Center of the University of Pennsylvania; 17 June 1996. Roberts DF, Ulla FG, Rideout VJ, Brodie M. Kids and media the new millennium. Menlo Park, Calif.: Kaiser Family Foundation; November 1999. Cohen J. A coefficient of agreement for nominal scales. Educ Psychol Meas. 1960;20(1):37-46. Kotz K, Story M. Food advertisements during children’s Saturday morning television programming: Are they consistent with dietary recommendations? J Am Diet Assoc. 1994;94:1296-1300. Andrews J, Maronick T. Advertising research issues from FTC versus Stouffer Foods Corporation. J Public Policy Mark. 1995;14(2):301-309. Kunkel D. Children and Television Advertising. In: Singer D, Singer J, eds. Handbook of Children and the Media. Thousand Oaks, Calif.: Sage Publications, Inc; 2001. Glanz K, Basil M, Maibach E, Goldberg J, Snyder D. Why Americans eat what they do: taste, cost, convenience, and weight control concerns as influences on food consumption. J Am Diet Assoc. 1998; 98(10):1118-1126.