eAppendix 1. Food and Fitness and Youth, Education, and Society shared methodology and respondent types
eAppendix 2. Detailed analysis procedures
eTable. Measures of school-based commercialism
Terry-McElrath YM, Turner L, Sandoval A, Johnston LD, Chaloupka FJ. Commercialism in US Elementary and Secondary School Nutrition EnvironmentsTrends From 2007 to 2012. JAMA Pediatr. 2014;168(3):234-242. doi:10.1001/jamapediatrics.2013.4521
Copyright 2014 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.
Schools present highly desirable marketing environments for food and beverage companies. However, most marketed items are nutritionally poor.
To examine national trends in student exposure to selected school-based commercialism measures from 2007 through 2012.
Design, Setting, and Participants
Annual nationally representative cross-sectional studies were evaluated in US public elementary, middle, and high schools with use of a survey of school administrators.
School-based commercialism, including exclusive beverage contracts and associated incentives, profits, and advertising; corporate food vending and associated incentives and profits; posters/advertisements for soft drinks, fast food, or candy; use of food coupons as incentives; event sponsorships; and fast food available to students.
Main Outcomes and Measures
Changes over time in school-based commercialism as well as differences by student body racial/ethnic distribution and socioeconomic status.
Although some commercialism measures—especially those related to beverage vending—have shown significant decreases over time, most students at all academic levels continued to attend schools with one or more types of school-based commercialism in 2012. Overall, exposure to school-based commercialism increased significantly with grade level. For 63.7% of elementary school students, the most frequent type of commercialism was food coupons used as incentives. For secondary students, the type of commercialism most prevalent in schools was exclusive beverage contracts, which were in place in schools attended by 49.5% of middle school students and 69.8% of high school students. Exposure to elementary school coupons, as well as middle and high school exclusive beverage contracts, was significantly more likely for students attending schools with mid or low (vs high) student body socioeconomic status.
Conclusions and Relevance
Most US elementary, middle, and high school students attend schools where they are exposed to commercial efforts aimed at obtaining food or beverage sales or developing brand recognition and loyalty for future sales. Although there have been significant decreases over time in many of the measures examined, the continuing high prevalence of school-based commercialism calls for, at minimum, clear and enforceable standards on the nutritional content of all foods and beverages marketed to youth in school settings.
School-based commercialism includes sponsorships, exclusive pouring rights contracts, incentive programs, advertising, fund-raising, and branded products sold in schools (eg, fast foods available during lunch).1- 3 Top food and beverage companies spent nearly $186 million in 2006 on youth-directed in-school marketing, most of which involved payments made or items provided through competitive venue contracts.4 In 2006, the Federal Trade Commission and the Department of Health and Human Services5 issued industry self-regulation recommendations for nutrition-related marketing to children. By the end of 2007, industry members had taken steps to encourage better nutrition,4 including the Children’s Food and Beverage Advertising Initiative and the Alliance for a Healthier Generation’s School Beverage and Competitive Food Guidelines.6,7 The Children’s Food and Beverage Advertising Initiative addresses in-school marketing only through sixth grade (thus not affecting most middle school or any high school environments).8 The Alliance for a Healthier Generation’s School Beverage and Competitive Food Guidelines address nutrient requirements and calorie restrictions but not commercialism. By 2009, in-school marketing expenditures had significantly decreased but were still at $149 million and primarily focused on carbonated and noncarbonated beverages.9
School-based commercialism policy is limited. Some states and school districts have addressed specific commercialism components10; however, few districts have relevant comprehensive policies. In 2011, only 11% of elementary, 10% of middle, and 9% of high school public school districts in the United States explicitly prohibited unhealthy food and beverage marketing.11
Estimates of student exposure to school-based commercialism include single-year estimates,4,10,12,13 separate elementary and secondary school estimates,14- 16 and single-venue estimates.17 School-based commercialism may vary significantly by grade and student body racial/ethnic distribution and socioeconomic status (SES).4,15 Updated estimates across the grade spectrum are needed. The present study provides estimates of exposure to specific school-based commercialism forms for nationally representative samples of US public elementary, middle, and high school students from 2007 to 2012. Three research questions were examined. What are the current prevalence estimates of school-based commercialism exposure? Is there evidence of significant change over time? Is there evidence of significant differences by student body SES or racial/ethnic composition?
This study used 6 years of data (2007-2012) from 2 parallel surveys conducted through the Bridging the Gap research program. Elementary school data were collected through the Food and Fitness (FF) study, which is conducted at the University of Illinois at Chicago (detailed methods are available elsewhere18); middle and high school data were collected through the Youth, Education, and Society (YES) study, which is conducted at the University of Michigan (detailed methods are available elsewhere19). Both projects received ethics approval from their university institutional review boards. The studies were deemed exempt; however, the elements of consent were included in both the invitation letter and the questionnaire. For both studies, questionnaires with a modest monetary incentive were mailed to sampled school administrators in the spring. The mean of the FF response rates was 60%; means of the YES response rates were 72% without replacement for nonresponding schools and 86% with replacement. Detailed methods are reported in the Supplement (eAppendix 1).
Several survey items pertained to school-based commercialism. Pilot testing of surveys was conducted with a convenience sample of principals. Detailed reliability and validity studies were not conducted; however, participants reported no difficulties in completing the measures. Full descriptions of all commercialism measures are found in the Supplement (eTable).
YES school demographics were reported by survey respondents; FF school demographics were obtained from the National Center for Education Statistics public school data files.20 Student body race/ethnicity was coded as predominately (≥66%) white, majority (>50%) African American/black, majority (>50%) Hispanic, and other racial/ethnic composition. Student body SES was coded according to the percentage of students eligible for free and reduced-price lunch as follows: low (≥40% eligible), medium (15%-39% eligible), and high (<15% eligible). Elementary school total enrollment was categorized as a dichotomy of up to 500 compared with 501 or higher (very few elementary schools had >1000 students). Middle and high school total enrollment was categorized as 0 to 500 students, 501 to 1000, 1001 to 1500, and 1501 or more students. Additional measures included population density, region, and grade surveyed.
Analyses were conducted using SAS, version 9.2 (SAS Institute, Inc). Data were weighted to adjust for differential probability of school selection and estimated target grade enrollment. Weighted results represent the percentage of all target grade students in public schools with specified commercialism measures. Analysis procedure descriptions are reported in the Supplement (eAppendix 2).
Table 1 provides school characteristic descriptive statistics. Table 2 and Table 3 provide trend analysis results and grade comparisons.
The percentage of students attending schools with exclusive beverage contracts (EBCs), incentives, and profits decreased significantly from 2007 to 2012 for all grades. By 2012, only 2.9% of elementary school students attended schools with existing EBCs (vs 10.2% in 2007); 49.5% of middle school and 69.8% of high school students attended schools with EBCs in 2012 compared with 2007 rates of 67.4% and 74.5%, respectively. More students attended schools receiving EBC profits (for elementary schools, total revenue generated from vending beverage sales; for secondary schools, profit from contract beverage sales receipts) than incentives (cash awards, equipment, supplies, or other donations obtained once total beverage sales exceeded a specified amount) (3.5% vs 0.9% for elementary; 31.3% vs 14.9% for middle; and 53.4% vs 24.7% for high school students in 2012). In schools receiving any EBC profits, mean per-student 2012 profit was $1.75 for elementary, $1.54 for middle school, and $4.18 for high school students (differences between elementary and middle school were not significant). For all beverage-related measures, exposure significantly increased with grade level.
School-based food vending commercialism was less extensive than that for beverage vending; still, 24.5% of middle school and 51.4% of high school students attended schools with company-sold food vending in 2012. Significant change over time was observed only for 2 measures (and only at the high school level): (1) any companies selling food vending items and (2) schools receiving food vending sales profits. More students attended schools receiving food vending sales profits than incentives (13.9% vs 5.8% for middle school, 35.6% vs 9.3% for high school students). Mean per-student profit from food vending sales (if any) was not significantly different by grade level in 2012 ($1.25 for middle and $1.87 for high school students). Food vending commercialism was generally more extensive at the high school than middle school level, although less so than for beverage measures.
At the elementary school level, food coupons were used more often than posters or other advertisements. In 2012, 63.7% of elementary school students attended schools with food coupons compared with 5.8% attending schools with posters or other advertisements for soft drinks, fast food, or candy (however, 5.8% was a significant increase from 2007, when only 3.3% of elementary school students attended schools with posters or other advertisements).
In 2012, the form of visual advertising or other marketing experienced by the largest percentage of secondary students was school event sponsorships (10.4% of middle, 21.1% of high school students), followed by beverage supplier advertising (5.3% of middle, 13.5% of high school students), coupons (5.5% of middle, 5.3% of high school students), and posters or other displayed materials (2.3% of middle, 2.1% of high school students). Less than 1% of middle or high school students attended schools with textbook cover or menu advertising. Middle school student exposure to beverage supplier advertising, coupons, and textbook cover or menu advertisements decreased significantly since 2007. High school student exposure to school event sponsorships and posters or other displayed materials also decreased. Exposure to coupons and posters or other displayed materials was not significantly different for middle and high school students; exposure to all other forms of visual advertising or other marketing was significantly higher for high school than for middle school students.
Schools where fast food was available to students at least once a week in 2012 were attended by 10.2% of elementary, 18.3% of middle, and 30.1% of high school students. In 2012, 1.3% of elementary, 9.0% of middle, and 19.3% of high school students attended schools where access to such products occurred daily. Significant decreases over time in such exposure were observed for middle school but not elementary or high school students.
In 2012, more than 70% of elementary and middle school students and nearly 90% of high school students attended schools with any of the commercialism measures examined. Significant decreases in exposure were observed for middle school students since 2009.
Student body characteristic analyses required adequate sample sizes to obtain reliable estimates and were limited to measures for which at least 10% of students in a grade attended schools with such commercialism in 2012. Three measures were examined at the elementary level: coupon use, any fast food availability, and any commercialism. The following measures were examined for both middle school and high school: any EBCs, EBC incentives or profits, company-based food vending, food vending profits, sponsorships, any fast food availability, and any commercialism. Two additional measures were examined only for high school: EBC advertising and daily fast food availability. Models included all years available and simultaneously controlled for student body race/ethnicity, SES, school enrollment, region, population density, and year (high school models also controlled for grade).
No significant differences by student body racial/ethnic distribution in the likelihood of coupons, fast food availability, or any commercialism were observed at the elementary school level. The unadjusted percentage of elementary school students attending schools with food coupons was significantly lower in high-SES schools (57.0%) than in mid-SES (72.5%) (odds ratio [OR], 1.90; 95% CI, 1.41-2.57; P < .001) or low-SES (72.2%) schools (OR, 1.86; 95% CI, 1.40-2.48; P < .001). Allowing fast food was more likely for students in high-SES elementary schools (26.7%) than in mid-SES (13.0%) (OR, 0.45; 95% CI, 0.30-0.69; P < .001) or low-SES (8.6%) schools (OR, 0.26; 95% CI, 0.17-0.42; P < .001). No significant SES differences were observed for the any commercialism measure.
At the secondary level, significant differences by student body racial/ethnic distribution were observed for only 2 measures for both middle and high school students (any EBCs and any EBC profits) and 3 additional measures for high school students (company-based food vending, sponsorships, and any commercialism) (Table 4). Where significant differences were observed, exposure for students attending predominately white schools was significantly greater than for students attending majority African American/black schools. Evidence of significantly higher exposure for students in predominately white schools compared with majority Hispanic schools was also found for any EBCs at the middle school level and sponsorships at the high school level.
Differences by student body SES were observed for 3 secondary school commercialism measures related to beverage vending (Table 5). Middle and high school students attending high-SES schools were significantly less likely than students in mid-SES or low-SES schools to have an EBC. The EBC incentives and profits were significantly more likely for high school students attending low-SES compared with those attending high-SES schools.
To examine potential interactions between student body race/ethnicity and SES with any EBCs, middle and high school models were evaluated including centered continuous measures for percentage of white students and percentage of students eligible for free and reduced-price lunch followed by models including an interaction between the 2 centered measures. For middle school, both direct effects and the interaction term were significant and positive; for high school, both direct effects were significant and positive but no significant interaction was observed. Thus, for middle and high school students, the likelihood of attending a school with an EBC increased with the percentage of white students attending the school and also independently increased as the percentage of students eligible for free and reduced-price lunch rose. For middle school students, EBCs were most likely in schools with both a higher percentage of white students and a higher percentage of students eligible for free and reduced-price lunch.
This study examined 2007-2012 commercialism trends in schools attended by nationally representative samples of US elementary and secondary school students. Although some measures showed significant decreases over time (especially beverage vending measures), most students attended schools with one or more commercialism types. Overall, commercialism increased significantly with grade level. For some measures, there was consistent evidence for higher exposure for students attending predominately white schools or schools with lower student-body SES.
The findings are subject to limitations. Data are based on school administrator responses to self-administered questionnaires, raising possibilities of reporting error, social desirability bias, or both. To minimize social desirability bias, schools and respondents were guaranteed confidentiality. To minimize response error, directions called for different questionnaire segments to be completed by personnel most knowledgeable about the subject matter. Respondents may not have considered branded items as product advertising if the main communication focused on another issue (eg, physical activity promotion posters sponsored by food or beverage companies); thus, reported commercialism levels may be underestimated. Limitations notwithstanding, these analyses provide nationally representative school-based commercialism exposure trends for US public elementary and secondary school students.
The most prevalent commercialism type was different for elementary vs secondary school students. Two-thirds of elementary school students attended schools using food coupons (vs approximately 5% of secondary school students). In contrast, 49.5% of middle and 69.8% of high school students attended schools with EBCs (vs 2.9% for elementary school students). Children and adolescents clearly influence parental purchasing decisions, but access to disposable income is markedly lower for children. Allowances (if present) generally increase with age, and adolescents can seek employment to earn wages. Thus, it is not surprising that the most prevalent commercialism for elementary school students—food coupons—does not involve immediate purchases but rather is easily delivered to parents with purchasing power. Conversely, the most prevalent commercialism form for older students—EBCs—requires disposable income and immediate purchasing ability.
Neighborhood fast food density, as well as outdoor food and nonalcoholic beverage and overall television food and beverage advertising, has been shown to be especially high for African Americans in the United States.21- 24 In contrast, the present study found that EBC-related commercialism, companies selling food-vending items, and sponsorships were significantly lower in majority African American/black high schools compared with predominately white high schools. Research14 has shown overall vending machine availability to be significantly higher in predominately white high schools than in majority African American/black high schools; if vending machines are less prevalent, commercialism associated (as least in part) with such venues is also likely to be lower. We are unaware of research investigating corporate sponsorship activity based on the racial/ethnic composition of the receiving entity that could be used to evaluate the current findings.
Schools present highly desirable marketing environments; students are captive audiences arranged in presegmented age groups in relatively uncluttered marketing environments.25- 28 Students are in schools to learn, and schools and teachers are viewed as trusted role models, lending added credibility to marketing efforts.27,28 School-based commercialism attempts not only to increase direct sales but also to increase brand recognition and future brand loyalty.25,26,29 Positive brand associations also increase when companies appear to give back to communities through participation in fund-raisers or incentive programs.1,27
School-based commercialism presents highly desired income for cash-strapped schools and districts. Approximately 44% of total 2011 US education expenditures for public elementary and secondary schools came from state funds,30 but state education funding cuts have resulted in 35 states having 2012 education funding below 2008 levels.31 School districts must make up the difference, and one possibility is additional revenue through school-based commercialism.
Several psychological, educational, and health concerns have been raised regarding school-based commercialism.28,32,33 Saturating the school environment with marketing conflicts with the need to train individuals to become critical thinkers with the ability to reflectively evaluate a consumer society.34 Also, most foods and beverages marketed in any venue toward children and adolescents are high in calories, sugar, salt, and fat, and are low in essential nutrients35; data indicate that the same is true for in-school marketing.4 In-school marketing for less-healthy foods and beverages clearly counteracts educational programming aimed at developing good nutritional habits.1
The Federal Trade Commission4 has called for companies to voluntarily end in-school promotion of nutritionally poor products, and potential local policy options to limit marketing of obesogenic foods and beverages have been suggested.26 Others have called for school commercialism prohibitions unless compelling evidence is provided showing that the intended marketing causes no harm to children28 or for statutory action requiring schools to be free from commercial promotions targeting children under all circumstances.36 In 2007, Maine passed the first state legislation limiting the marketing of foods of minimal nutritional value in public kindergarten through 12th-grade school campuses.37 A 2010 evaluation38 of the law’s implementation indicated that marketing of these foods remained widespread in high schools more than 2 years after the ban was implemented; the authors noted the need for stronger nutrition standards, improved communication with school administrators, implementation assistance, industry cooperation, and enforcement.
Efforts to improve school nutrition environments and reduce student commercialism exposure have left many school decision makers concerned about revenue loss.1,39,40 Reviews of available revenue-related studies examining the consequences of school nutrition guideline and/or marketing practice changes indicate that most schools do not experience overall revenue loss41; school-based commercialism benefits are often less than expected.2 However, districts that are more heavily involved in school-based commercialism have fewer financial resources.2 Such patterns are consistent with those identified in the present study that found more involvement with beverage vending–related commercialism among schools with the most economically disadvantaged student populations.
The recently published US Department of Agriculture interim final rule42 governing school competitive nutrition environments provides standards for all competitive venue foods and beverages served and sold in schools participating in the National School Lunch Program and School Breakfast Program. The new standards should significantly improve the nutritional quality of foods and beverages sold in school competitive venues; however, they do not address in-school marketing. This leaves open an interesting opportunity for companies to continue to extend brand recognition and loyalty efforts as well as promote less-healthy items. For example, vending machines may be stocked with items meeting the most recent Dietary Guidelines for Americans,43 but the visual advertising on the machine may still promote items of low nutritional value.
The current analyses show that in 2012, most US elementary, middle, and high school students attended schools where they were exposed to commercialism aimed at obtaining food or beverage sales or developing brand recognition and loyalty for future sales. Although there were significant decreases over time in many of the measures examined, the continuing high prevalence of school-based commercialism supports calls for, at minimum, clear and enforceable standards on the nutritional content of all foods and beverages marketed to youth in school settings.
Accepted for Publication: September 6, 2013.
Corresponding Author: Yvonne M. Terry-McElrath, MSA, Institute for Social Research, University of Michigan, Room 2344, PO Box 1248, Ann Arbor, MI 48106 (firstname.lastname@example.org).
Published Online: January 13, 2014. doi:10.1001/jamapediatrics.2013.4521.
Author Contributions: Ms Terry-McElrath had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Terry-McElrath, Johnston, Chaloupka.
Acquisition of data: Turner, Sandoval, Johnston, Chaloupka.
Analysis and interpretation of data: Terry-McElrath, Turner, Chaloupka.
Drafting of the manuscript: Terry-McElrath, Sandoval.
Critical revision of the manuscript for important intellectual content: Terry-McElrath, Turner, Johnston, Chaloupka.
Statistical analysis: Terry-McElrath.
Obtained funding: Johnston, Chaloupka.
Administrative, technical, or material support: Turner, Sandoval, Johnston.
Study supervision: Chaloupka.
Conflict of Interest Disclosures: None reported.
Funding/Support: The FFS and YES are part of a larger research initiative funded by the Robert Wood Johnson Foundation, entitled Bridging the Gap: Research Informing Policy and Practice for Healthy Youth Behavior.
Role of the Sponsor: The Robert Wood Johnson Foundation had no role in the design and conduct of the study; collection, management, analysis, or interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer: The views expressed in this article are those of the authors and do not necessarily reflect the views of the funder.