THE EFFECT OF CELEBRITY ATHLETE MODELS

IN FOOD ADVERTISING ON THE PERCEIVED

HEALTHINESS OF FOOD PRODUCTS

By

ALLEN E BARROW III (TREY)

Bachelor of Arts Journalism/Advertising

University of Oklahoma

Norman, Oklahoma

2003

Submitted to the Faculty of the

Graduate College of the
Oklahoma State University
in partial fulfillment of
the requirements for
the Degree of
MASTER OF SCIENCE
May, 2009
THE EFFECT OF CELEBRITY ATHLETE MODELS IN

FOOD ADVERTISING ON THE PERCEIVED

HEALTHINESS OF FOOD PRODUCTS

Thesis Approved:

Dr. Jami Fullerton

Thesis Adviser

Dr. Lori McKinnon

Dr. Stan Ketterer

Dr. A. Gordon Emslie

Dean of the Graduate College

ii
 TABLE OF CONTENTS

Chapter Page

I. INTRODUCTION ......................................................................................................1

Health Trends ...........................................................................................................3

Federal Trade Commission Involvement .................................................................4
The Use of Celebrity Spokespersons .......................................................................5
Michael Phelps as a Celebrity Athlete .....................................................................6
Eli Manning as a Celebrity Athlete ..........................................................................7
Theoretical Framework ............................................................................................8
Significance of the Study .......................................................................................10
Methodology ..........................................................................................................10
Overview ................................................................................................................11

II. REVIEW OF LITERATURE..................................................................................12

The Process of Advertising ....................................................................................12

Persuasion Theory..................................................................................................13
Source Credibility in Advertising ..........................................................................14
Elaboration Likelihood Model ...............................................................................16
Celebrity Athlete Models in Advertising ...............................................................20
Food Marketing Efforts..........................................................................................21
Health Claims in Advertising.................................................................................24
Government Regulations .......................................................................................28
Summary ................................................................................................................29

III. METHODOLOGY ................................................................................................31

Research Objective ................................................................................................31

Variables and Hypotheses ......................................................................................32
Experiment Design.................................................................................................34
Procedure ...............................................................................................................35
Advertising Treatments ..........................................................................................36
Research Instruments .............................................................................................37
Respondents ...........................................................................................................38
Data Collection, Processing and Analysis .............................................................39

iii
Chapter Page

IV. FINDINGS.............................................................................................................40

Respondent Profile .................................................................................................41

Health Perception of the Product (HPP) ................................................................42
Health Concern (HC) .............................................................................................42
Attitude Toward the Ad .........................................................................................43
Purchase Intent .......................................................................................................43
Michael Phelps Drug-use in the Media ..................................................................44

V. CONCLUSION ......................................................................................................45

Summary ................................................................................................................46
Discussion ..............................................................................................................47
Implications............................................................................................................49
Limitations .............................................................................................................50
Population ........................................................................................................50
Respondent.......................................................................................................50
Sample Size ......................................................................................................51
Advertisement Type .........................................................................................51
Experimental Conditions .................................................................................51
Treatments........................................................................................................52
Type of Celebrity .............................................................................................52
Future Research .....................................................................................................52
Children............................................................................................................52
Celebrity Athletes ............................................................................................52
Advertising Type .............................................................................................53
Conclusion .............................................................................................................53

REFERENCES ............................................................................................................55

APPENDIX A: RESEARCH INSTRUMENTS ..........................................................61

A1: Michael Phelps Instrument and treatment.......................................................61
A2: Eli Manning Instrument and treatment ...........................................................72
A3: Non-celebrity Male Model Instrument and treatment ....................................83

APPENDIX B: IRB DOCUMENTS ...........................................................................92

IRB Approval .........................................................................................................92
Approved Student Recruitment Flyer ....................................................................93
Approved Faculty Email ........................................................................................94
Approved Faculty Script ........................................................................................95
Approved Student Questionnaire Email ................................................................96
Approved Sign-Up Sheet .......................................................................................97

iv
 CHAPTER I

INTRODUCTION

Obesity among Americans has risen dramatically in recent years (Mikhailovich &

Morrison, 2007). Statistics from the Centers for Disease Control and Prevention (CDC)

indicate that since the 1970’s obesity has steadily increased among people of all ages,

races and socioeconomic backgrounds (CDC, 2006). The effects of obesity have created

dire health consequences for society. If this trend continues, young people today will

face worse health conditions later in life than any previous generation (Larimore & Flynt,

2005).

Factors contributing to obesity are widely debated but focus on two basic causes –

not enough exercise and a high-calorie diet (Colditz, 1999). Previous research has

explored a variety of environmental factors that influence increasing overweight trends

among Americans. These are commonly thought to include increased portion sizes,

eating out more frequently, increased consumption of sugary sweetened drinks, increased

television and electronic games, changing labor markets and a fear of crime, which

prevents outdoor exercise (Hardus, van Vuuren, Crawford & Worsley, 2003).

1
 Many have blamed an abundance of fast-food and snack food advertising for the

increase in calorie intake among young people (Lynn & Novosat, 2008). Kids today face

almost double the amount of television commercials than children in the 1970’s, with

American children averaging 40,000 commercials in the year 2000 (Walsh, 2008).

Recognizing the seriousness of the obesity problem, the government, consumer

groups and parents have put pressure on the food industry to adjust their advertising

practices (Seiders & Petty, 2007). The use of certain celebrity athletes in advertisements

is particularly concerning as they may be seen as “healthy” role models. The Federal

Trade Commission (FTC), in its recent report on food advertising, recognized celebrities

used in food marketing, including such athletes as David Beckham, Alex Rodriguez,

LeBron James and Derek Lee (FTC, 2008). However, the report does not make specific

recommendations for further research, on the use of celebrity athletes and their effect on

consumer perceptions of food products. Therefore, empirical research is important to

understand the potential impact that the presence of the celebrity athlete model has on

food choices.

While the outcry is loudest over concern for the health nutrition of children,

statistics show that all Americans are affected by obesity and weight-related illnesses. At

this time it is still largely in the marketers’ control to self-govern their advertising and

make the necessary adjustments based on the recommendations by the Federal Trade

Commission. Therefore, the purpose of this study is to determine how the use of a

celebrity athlete model in food advertising affects the perception of healthiness of the

advertised food product.

2
Health Trends

According to statistics maintained by the Centers for Disease Control and

Prevention (2006), in the years ranging from 2003-04 just over 17 percent, or 12.5

million, children and adolescents ages 2 - 19 in the United States were categorized as

overweight. Nearly half of young adults and children do not exercise on a regular basis.

Similarly, 32 percent of adults were categorized as obese, with another 5 percent defined

as being extremely obese. The statistics also indicate a lack of exercise or activity for

adults, with at least 60 percent of adults falling short of the recommended physical

activity. Additionally, dietetic survey research showed an increase in calories consumed

among Americans between the 1970’s and late 1990’s (Nestle, 2007).

The statistics for both boys and girls show a significant increase in the prevalence

of obesity between 1994 and 2004 (CDC, 2006). According to various sources, obesity

typically develops from an imbalance between calories consumed and calories used

through physical activity (Robinson, 2001; Colditz, 1999). The implications of less

physical activity are documented by increased cases of type II diabetes, hypertension,

stroke, cancers and other conditions (Colditz, 1999). For young people experiencing

overweight symptoms it is more severe because, as statistics indicate, they are more at

risk to be obese as adults. The development of obesity at earlier ages, such as childhood,

allows for many years of exposure to health risks and complications. This increased

exposure creates a higher probability that further health issues will develop throughout a

person’s lifetime (Dietz, 1998). While the increases in obesity statistics are most

prevalent in the United States, the trends are evident in other countries and regions

around the world (Ebbeling, Pawlak & Ludwig, 2002).

3
 Research predicts that based on current rates, there is a potential that 100% of

Americans will be overweight or obese by the year 2040 (Heaner, 2007). The statistics

indicate the alarming rate at which children are susceptible to obesity at earlier ages than

previous generations. Today’s generation of children are said to be the first generation to

have a shorter life expectancy than their parents (Olshansky, Passaro, Hershow, et al,

2005).

Federal Trade Commission Involvement

In 2006, the Federal Trade Commission (FTC, 2008) initiated a review of the

food industry’s advertising practices to children. The FTC analyzed the marketing

practices of 44 major food companies in the United States. According to the study, food

companies spent roughly $1.6 billion on advertising to children and adolescents in 2006

(FTC, 2008). The FTC study set out to analyze how food companies targeted children

through the media. Additionally, the FTC wanted to know the potential consequences of

the food advertising on children and how advertising could increase poor dieting habits

among young people. The study found that children were still regularly exposed to a

variety of food advertising messages through the media. The report recommended

several areas of concern which needed future research; however, the use of celebrity

athletes in food advertising was not an area the report noted for further research.

Due to government pressure, food companies slowly began to change their

advertising practices. This government review highlighted the ongoing public debate

between food advertising and sedentary lifestyle to see how each contributed to obesity.

4
The Use of Celebrity Spokespersons

The use of celebrities in advertising is a common way for marketers to identify

and communicate with a target audience (Kim & Na, 2007). The power of the role

model, whether an athlete or celebrity, can alter consumer attitudes and perceptions

toward a brand or product (Bush, Martin & Bush, 2004). As society becomes

increasingly aware of issues causing obesity, it becomes critically important to

understand how powerfully these role models can influence people’s health perceptions

of a food product.

The celebrity spokesperson has historically been used by advertisers. Some

estimates indicate that about 25 percent of advertisements feature a well-known celebrity

endorser (Amos, Holmes & Strutton, 2008). There are a variety of objectives that can be

accomplished with the presence of a role model in an advertisement. Most commonly

celebrities in advertising create an identity for a brand or product that relates to or is

inferred by the spokesperson (Farrell, Karels, Monfort & McClatchey, 2000).

Additionally, with improved communication technology over the recent decade,

celebrities have become more prevalent in society (Bush, Martin & Bush, 2004). The

celebrity athlete’s image can be leveraged as an expert endorsement. While this depends

on the industry and the tone of the advertisement, an expert opinion made by a role model

in an advertisement can have the power to influence the perceptions and attitudes of an

audience.

Celebrity athletes are those who, for the purpose of this research, are described as

publicly well-known athletes who have an image that appeals to consumers, and,

therefore, marketers. They are typically known for their performance on their respective

5
field or court. However, in addition to being a professional athlete, the most popular

sports athletes are sought after by corporations for million-dollar endorsement contracts.

These contracts can range from soft drinks to underwear (Bush, Martin & Bush, 2004). It

is not uncommon for one celebrity to have multiple endorsement contracts at one time.

This could suggest in some sense that celebrity athletes have a well rounded image or

expertise because their opinions are relevant to consumers outside of their specific sport.

For celebrity athletes, the food and beverage category represents great opportunity for

endorsement contracts. These endorsements alone can bring the athletes greater publicity

and financial gain than from sports performance alone.

Michael Phelps as a Celebrity Athlete

The Olympic Summer Games of 2008 in China produced some of the highest

Olympic network ratings for the sponsor network NBC (ESPN Online, 2008). This was

largely driven by the swimmer Michael Phelps. Phelps was publicized for winning more

gold medals in one Olympic Games than any previous Olympic athlete. Headlines

suddenly referenced the endorsement success that Phelps would bring to brands as his

recognition around the world increased (Walker, 2008).

Suddenly sponsors were leveraging their brands to incorporate the endorsement of

Phelps. At the same time the public was embracing the media attention regarding Phelps.

This attention involved information about everything from his family to his training

regiment. Information spread about his 12,000 calorie diet (Henley, 2008) as well as his

McDonald’s breakfast, which could consist of four Egg-McMuffins. One could argue

that it is relatively easy to understand that the intensity of Phelps’ training program

6
allowed him to burn more calories than the average person. Additionally, it could be

argued that people know four egg McMuffins at one meal is not considered a nutritious

breakfast. However, the question can be raised whether or not young people have the

ability or motivation to make those connections after the exposure of such endorsement

advertising.

In contrast to the success Phelps’ enjoyed after the Olympic Summer Games,

news rapidly spread in January 2009 regarding a photo captured at a University of South

Carolina party of Michael Phelps smoking marijuana (Nashawaty, 2009). In the

following days as Phelps publicly admitted to the drug accusations, corporate sponsors

quickly began to let contracts expire or reexamined marketing campaigns with his

presence. Although his image was altered from the hero he was just days and months

earlier, he is recognized for his athletic ability by people around the world.

Eli Manning as a Celebrity Athlete

Super Bowl XLII Champion and Most Valuable Player Eli Manning leveraged his

marketability in February of 2008. After following the success of his older brother,

Peyton Manning, on the football field, Eli would rapidly receive endorsement contracts

similar to those held by his older brother. Endorsement deals for Manning include Kraft

Oreos, Reebok and Citizen Watch (Janoff, 2008).

The season Eli Manning won the Super Bowl included many weeks of national

media attention during the New York Giants season. Manning has, in contrast to Phelps,

remained out of the negative media spotlight. Whereas Phelps faced issues with his DUI

at age 19, Manning has been noted as a less publicized celebrity in his personal life. For

7
marketers this could be a safer celebrity endorsement bet. However, it is difficult to

determine how consumers perceive these young athletes as they grow up in front of the

public. Both Phelps and Manning are young white male athletes in their early 20s who

are well-known in the public through their respected sports. Increasingly, both have a

strong presence in more mainstream life through advertising endorsements.

The use of athletes, or in this case a celebrity athlete, is nothing new to

advertising. There are studies that analyze the effects of the role model or celebrity

athlete in advertising (Bush, Martin & Bush, 2004). However, there is a lack of literature

regarding the influence a celebrity athlete has on the perception of nutritional value of the

products they endorse. Coming off the recent summer Olympic Games and Michael

Phelps newly acquired endorsement deals, several of which are in the food and beverage

category, it is timely and important to understand the perception of the nutritional value

that a celebrity athlete can create through advertising.

Theoretical Framework

Persuasion is usually defined as attitude change resulting from exposure to

information from others (Olson & Zanna, 1993). The research by Carl Hovland

(Hovland, Janis & Kelly, 1953) with the U.S. Army during World War II was

instrumental in demonstrating how people develop attitudes through exposure to

mediated communications such as films. When applied to advertising, the goal of

persuasion is to change consumers’ attitudes toward a particular product, brand, person or

idea that ideally will lead to buying the product, voting for a political candidate or

8
accepting an idea, like “Click-it or Ticket.” In an attempt to persuade consumers,

advertisers use many techniques including celebrity models in their advertisements.

The Elaboration Likelihood Model (ELM) is an extension of persuasion theory

that explains how people interpret messages. Developed by Petty and Cacioppo (1979),

the model suggests that there are two routes to persuasion people can take on a

continuum when processing a message. The central route involves a high level of

elaboration or thought. It is believed to occur when a message topic has personal

relevance to the receiver. The receiver then analyzes the characteristics of the message,

such as the quality of the argument, and considers previous assumptions regarding the

topic. The second route, known as the peripheral route, involves less elaboration by the

receiver on the presented topic. In this situation, the message receiver will likely have

less personal relevance with the topic, and therefore does not engage in detailed thought

or analysis of the argument. However, receivers displaying the peripheral route are more

likely to be influenced by the peripheral cues, such as color, design or models in the ad

(O’Keefe, 2002).

The persuasion theory, the Elaboration Likelihood Model will inform this study.

The theory will be used to predict and possibly explain how the presence of a celebrity

athlete in an ad might convince consumers that a food product is healthy in the absence of

any explicit health claims.

9
Significance of the Study

The purpose of this study is to determine how the use of a celebrity athlete model

in food advertising affects the perception of healthiness of the advertised food product

among college students.

If the research finds that the celebrity athlete has an effect on nutrition perception,

it may be beneficial for regulators and the industry. This knowledge could help set

standards to develop stricter guidelines for the use of celebrity athletes in food

advertising.

Methodology

The methodology for this study is a simple post-test only experimental design

using an experimental (treatment) group and a controlled variable group. Participants

were randomly assigned one of three different print ads for a breakfast cereal. One

treatment featured celebrity endorser Michael Phelps, another treatment featured celebrity

endorser Eli Manning, and the final treatment featured a generic male student. Other than

the cutline of the celebrity athlete’s name, treatment ads are identical and contain no

explicit health claims (see Appendix A). After viewing the ad, participants completed a

questionnaire measuring attitude toward the ad, health perceptions, source-credibility of

the celebrity endorsers, health concern and purchase intent.

The independent variables are the presence of a celebrity athlete in a food cereal

ad. The dependent variable is the participant’s perception of the health value of the food

product, attitude toward the ad, the source-credibility of the endorser and purchase intent.

The final dependent variable, which was also explored, is the respondent’s overall health

10
concern. The study used a sample of university students from Oklahoma State

University.

Overview

Chapter 2 provides a review of the past literature. This provides analysis of the

research conducted in relation to the celebrity role model and health perceptions of food

advertising.

Chapter 3 provides the method used to conduct the experiment design. This

includes the hypothesis, research questions and measurement scale. Additionally,

information regarding the sample and procedures are provided.

Chapter 4 includes a detailed analysis of the experimental results. Details

regarding the demographics of the sample are provided with measurement analysis of the

health perceptions.

Chapter 5 highlights relevant findings related to the research topic. There is also

detailed information regarding the limitations of this study, as well as recommendations

for future research in this area. Included are implications and recommendations for

marketers and regulators that impact the advertising of food products.

The final section of the thesis contains an appendix with the experimental

treatments.

11
 CHAPTER II

REVIEW OF LITERATURE

The following chapter provides a detailed overview of the existing literature

regarding the concepts of advertising, persuasion, source credibility, nutritional labeling,

and the effects of food advertising. The previous research in the related subjects offers

context for this study. Ultimately, the body of research will inform the study herein.

The Process of Advertising

Advertising is an indirect type of persuasion carried on the mass media and paid

for by an identified sponsor. It uses a variety of information, product benefits or

emotional appeals to create positive impressions (Rossiter & Percy, 1997). Effective

advertising leverages consumer insights to develop the most accurate tone and message

tailored to an audience. Ultimately the goal of advertising is to persuade the audience to

buy a product, vote for a candidate or adopt an idea (Severin & Tankard, 2001). The

repetitive nature of advertising typically forces audiences to view the advertisement

multiple times and ultimately creates an impression. Over time advertising can either

12
change or reinforce people’s attitudes toward the advertisers’ goals, such as increasing

sales of a product (Altman, Avery, et al, 2003).

It has been noted by previous academic researchers (Fullerton & Kendrick, 2006)

that unlike most industries of business, consumers feel compelled to openly discuss their

opinions regarding the success or failure of an advertisement – everyone is an expert on

advertising. However, the scientific research in the area of advertising and consumer

behavior explores in depth how ads affect people. Research has shown that people are

willing to alter their preferences, attitudes and behaviors after just one exposure to an ad

(Lowery & DeFleur, 1995).

Persuasion Theory

Persuasion is usually defined as attitude change resulting from exposure to

information from others (Olson & Zanna, 1993). This can be done through a typical

conversation or through mass communication; such is the case with advertising.

During World War II, the U.S. military recognized the need to persuade young

soldiers to fight in a foreign war. Toward that end, numerous films were made to

motivate soldiers and to hopefully change their attitudes toward the enemy and the Allies.

A Yale researcher, Carl Hovland, was responsible for measuring the effectiveness of the

military films known as the Why We Fight films (Lowery & DeFleur, 1995). Results

from Hovland’s military research led to his “magic keys of persuasion” (Lowery &

DeFleur, 1995). The keys of persuasion attempt to identify the factors that influence

effective persuasion, such as source credibility, which is an important element in this

study.

13
Source Credibility in Advertising

Carl Hovland (1951) discovered early in his work with persuasion that there was a

strong influence based on the credibility of the source. The marketer’s goal is to establish

credibility for a brand or product with the consumer through the consumer’s

identification with the spokesperson (Choi & Rifon, 2007). Research on the topic of

source credibility has uncovered several factors that influence people’s willingness to

give credibility to a source, such as expertise, trustworthiness and attractiveness

(Ohanian, 1991).

Ohanian (1991) researched source credibility of various celebrities among college

students. The study measured the impact that a celebrity’s attractiveness, expertise and

trustworthiness have on an audience and the purchase intent of an advertised product.

Four treatments were created using celebrities Linda Evans, John McEnroe, Madonna

and Tom Selleck. The celebrities were then matched to products that a person at the time

would have associated them with, such as tennis rackets for John McEnroe. A total of

578 respondents viewed one of the four treatments with a questionnaire. The results

found that there was a significant relationship between physical attractiveness and

trustworthiness, but the responses indicated that the perceived expertise of the endorser

with the product significantly related to the purchase intention of the product.

Amos, Holmes and Clinton (2008) analyzed the existing literature from previous

decades to determine the collective findings regarding source credibility in advertising.

The results highlighted the celebrity trustworthiness, celebrity expertise, and celebrity

14
attractiveness. Conversely, negative celebrity information had the most detrimental

effect on consumer purchase intent, brand attitudes, and attitude toward the ad.

The process of social learning was developed by Bandura in research conducted

in 1965 (Lowery & DeFleur, 1995). The experiments studied a series of aggressive acts

toward a Bobo doll, which were viewed by groups of children. One group was shown the

aggressive acts followed by rewards for such behavior, while another group was shown

punishment for the acts of aggression toward the doll. Finally, a third group was shown

no consequences for the acts of aggression. The results demonstrated a high and uniform

degree of learning whereby the students who witnessed punishment for the actions were

less likely to treat the doll in an aggressive manner. The research shows that children are

able to learn aggressive behavior through observation and are able and willing to

reproduce such actions when there are little consequences for doing so.

The role of the mass media in the learning process detailed by Bandura (1986)

revealed how role models influence people as they are exposed to new ideas, causes or

products. Initially this process assumed that for a person to experience a change about an

idea or product, the change would likely occur through a change agent. The change agent

(Rogers, 1963) was thought to be someone whom the person interacted with regularly,

such as a classmate, family member or personal salesman. However, as research in the

effects of role models in mass communication became more prominent, the concept was

adapted to recognize that there are vicarious role models created in the media. The

audience can develop a relationship with this role model even though there had been no

interaction other than through the means of mass communication. The outcome leads to

the opportunity for the role models to transfer their image to other products or brands

15
through advertising. The vicarious role model theory accounts for the broad recognition

of some characters and celebrities to mass audiences.

Elaboration Likelihood Model (ELM)

The Elaboration Likelihood Model, developed by Petty and Cacioppo (1979),

suggests that there are variations in the way that persuasion affects the receiver. This

effect is determined by how the receiver of a message is willing to engage in elaboration,

or critical thinking of the message topic. Based on the degree of willingness there can be

two types of influence.

One is the central route to persuasion which involves a great deal of elaboration

and critical thinking by the receiver. This high level of thinking is used because the

message receiver has preexisting knowledge of the topic or the topic has relevance to the

receiver. The second type of elaboration, known as the peripheral route, involves less

elaboration or critical thinking than the central route. With the peripheral route the

receiver uses simple cues to decide his position on the topic. The peripheral cues, such as

source credibility and the appeal of the ad, can be more influential in the receiver’s

attitudes than the argument value (O’Keefe, 2002).

Two factors that affect the opportunity for the Elaboration Likelihood Model to

take place are the ability and motivation of the message receiver to engage in elaboration

of the subject (O’Keefe, 2002). First, message receivers must have the ability to

comprehend the intention of the message. If the receiver does not have the ability to

understand the topic, there is little opportunity to create a change in perception. Second,

the receiver must have the motivation to receive the message and make an effort to

16
interpret the basic meaning of the message. There is very little opportunity for high

elaboration to occur unless both factors are present.

Research by Petty, Cacioppo and Goldman (1981) builds on persuasion theory by

exploring the relationship between an individual’s personal involvement with the topic

and the effect on perception. The research tested how undergraduate students are

persuaded by varying types of arguments based on their level of personal relevance to the

topic. The results found that subjects with a high level of personal relevance to the topic

were strongly influenced by the quality of the argument. Inversely, subjects who had a

low level of personal relevance to the topic showed more influence to the peripheral cues,

such as the credibility of the information source. The findings confirmed the principles

of ELM by demonstrating the effects that personal relevance has on how an individual

will perceive similar messages. Findings from this research are important for marketers

who attempt to target messages to consumers who have varying levels of personal

involvement with their product.

The elements of ELM are thought to be an example of a dual process approach of

information comprehension and interpretation (O’Keefe, 2002). The argument strength

and level of involvement are identified by the previous research by Petty and Cacioppo

(1984) as the factors determining central and peripheral routes. The present research

will draw from the principles developed by Petty and Cacioppo (1979) and study the

effect of the peripheral route. The research purpose is consistent with previous research

where the low involvement respondents are influenced more strongly by the peripheral

effects of the ad.

17
 The ELM research by Petty, Cacioppo, and Schumann (1983) is important to

marketers who seek to design ads that effectively engage the consumer into interpreting a

message as the marketer desires. The researchers performed an experiment to measure

attitude change based on argument strength. In the study, a total of 160 male and female

college undergraduates were exposed to magazine ads under varying degrees of product

involvement. The advertisements were presented to the students in the form of booklets,

one containing the advertising stimuli and the second containing the dependent measures.

The results found that the argument quality, which in this study was related to Edge

Razors, was an important factor when high personal relevance was present in the

respondent. However, the manipulation of the celebrity endorser in a low involvement ad

produced significant attitude change. The study concludes that there cannot be an

overwhelming emphasis on the central or peripheral routes in advertising as each person

responds differently depending on their personal relevance to the topic.

Research conducted by Frewer, Howard, Hedderley and Shepherd (1997)

explored the Elaboration Likelihood Model and food risks in communication methods.

The primary goal of the experiment was to measure the effectiveness of the source

credibility in risk communication. The research was conducted with 160 participants.

Source credibility is seen to be a major influence in people’s beliefs and perceptions of

information credibility. The research by Frewer et al. (1997) explored the personal

relevance that the source credibility produces when the topic relates to a possible risk.

The two topics used by the researchers were the hazards of excessive alcohol use and

microbiological food-borne risk. The research exposed experimental groups to

communication from a trusted source (medical doctors) and a non-trusted source

18
(government officials). The results showed that source credibility and message content

did affect the perception of the message; however the ELM did not indicate significant

elaboration toward risk based on the source alone.

In a study by Jones, Sinclair and Courneya (2003), the authors researched the

effects that advertising health and exercise promotion can have on behaviors. The study

specifically examines the Elaboration Likelihood Model to measure the respondents’

willingness to engage in extensive elaboration toward the topic. Previous research in

weight loss (Kreuter, Bull, Clark & Oswald, 1999) show that the persuasion theory can

be effective in influencing individuals to engage in health and exercise activities after

exposure to the messages. Because of this success, the authors (Jones, Sinclair &

Courneya, 2003) used the ELM framework to understand the effectiveness of persuasion.

Previous research exercise campaigns (Tversky & Kahnesman, 1981) demonstrated the

prospect theory, in that individuals respond differently depending on how communication

messages are framed, specifically relating to how behavior can result in gains or losses.

Therefore, the research by Jones, Sinclair and Corneya (2003) sought to determine how

the principles of ELM affect behavior and intent to exercise based on the presence of a

credible source in the message. A total of 192 students participated in the experiment.

The students were randomly assigned a reading from a credible source, a medical doctor,

or non-credible source, a student. The results indicated support for the ELM theory and

the prospect theory in that positively framed messages from a credible source were the

most effective communication to increase exercise motivation in students.

19
Celebrity Athlete Models in Advertising

The use of a celebrity athlete in advertising is common because of the publicity

the athlete receives through the media. The athlete’s recognition has the potential to be

extremely effective for marketers. A study by Bush, Martin and Bush (2004) measured

the effects of the sports celebrity on the behavioral intentions of teenagers. Specifically,

the researchers wanted to know how the presence of the sports celebrity would influence

the word-of-mouth among friends about the product, willingness to change brands

because of the athlete, and brand loyalty. The results of a sample of 218 teenagers

surveyed found that the presence of a sports celebrity in the advertising of a product led

to increased positive word-of-mouth and brand loyalty. The authors suggest that through

the study it can be confirmed that sports celebrities are role models for young adults.

Research suggests that the similarity between the role model and the product is

necessary. Experimental research such as Kim and Na (2007) suggest that participants

view the role model more favorably when matched with a congruent product.

Lockwood and Kunda (1997) analyze an alternative impact that role models and

superstars can impose on an audience. In the study, the authors proposed that because of

prior research that indicates that people can be either inspired or discouraged by the

presence of a successful role model, it is important to understand how this affects

emotions. When an audience becomes discouraged by the presence of a role model,

research suggests that it could be due to an overwhelming feeling of inability to

accomplish similar success as the superstars presented.

While the studies demonstrate an increase in behavioral change, empirical

research does not detail how the celebrity athlete serves as source credibility in food

20
advertising specifically. Research indicates that effective use of a celebrity in advertising

should have a strong link between the product being sold and the celebrity (Kim & Na,

2007; McCracken, 1989). In a culture driven by thinness and perfect body image

(Almond, 2000), it is difficult to understand why Americans are trapped in an obesity

epidemic, while yet they feel such a strong connection to world-class athletes.

Food Marketing Efforts

Companies that produce and market food products to children have been under

pressure in recent years regarding their advertising practices. Food advertisements to

children frequently run during Saturday morning cartoons or during the early afternoon

after school (Batada, Seitz, Wootan & Story, 2008). Many of these advertisements took

the form of the show that kids would have been watching, such as using cartoon

characters. The characters, many of which would become icons to young children and

even enter popular culture, would connect with children and reinforce the product they

were selling, such as Trix Rabbit or Ronald McDonald. Research has shown that the

types of products companies were advertising are commonly sugary-sweet food or

beverage products (Batada, Seitz, Wootan & Story, 2008).

Research by Batada and Wootan (2007) sought to assess the foods advertised in

affiliation with Nickelodeon media and characters. The authors recognized that while the

childhood obesity epidemic has been created by numerous factors, food marketing efforts

influence many important food choices that children make. The decisions include how

they interpret food choices, preferences, their diets, and their health. In the study, Batada

and Wootan (2007) analyzed television ads, magazine ads and grocery store packaging.

21
The results found a large portion of advertisements contained poor nutritional quality,

with television at 88 percent and magazines at 76 percent. Additionally, 60 percent of

Nickelodeon characters were found on food packaging of poor nutritional items. The

promotional partnerships with restaurants found that 94 percent were rated as poor

nutritional status. The significance of the percentages found in the study indicated a

negative trend in food marketing efforts.

A more recent content analysis study by Warren, Wicks, Wicks, Fosu and Chung

(2008) found a higher number of nutritional appeals in food advertising. The sample was

exposed to cable and broadcast television food advertisements, during the peak time

when school-aged children are viewing – 2:00 p.m. to 10:00 p.m. Seventy-seven

randomly selected days were recorded between January and May of 2006. Coders were

used to categorize ads as child-targeted, if the ads fell within four areas. The categories

were based on programming, visual cues or voice-overs, verbal references and specific

marketing promotions to children. The results found, similar to previous studies, that

low-nutrition products were the most frequent advertised to children. The ads were

effective in using cues that would appeal to children specifically, including messages that

could alter taste and mood effects. The concept of nutritional content was found to be

used third most frequently. The authors noted that the findings signaled an encouraging

change by marketers. However, it can be argued that children are not interested in the

direct nutritional messages in advertising, and, therefore, it is important for future

research to analyze how children peripherally process nutritional cues.

King and Hill (2008) explored the effect that unhealthy food advertising has on

309 British children. The experiment focused on how food magazine ads impacted the

22
children’s mood, hunger, food choice and product recall. Booklets were created with ads

randomly placed throughout. The booklets were presented to the children as school-

based media literacy exercise. The children’s body weight, height and body satisfaction

were measured one week after the exposure. There were no effects on the food choice or

preference of the respondent during analysis. However, results found that there was a

greater impact in the less healthy food products than the healthier products. The authors

note that further research is needed to determine the effectiveness of alternative

marketing channels such as television. This study indicates that children are influenced

by ads for less healthy food products.

The dietary restraint theory used by Warren, Strauss, Taska and Sullivan (2005)

analyzes the circumstances and factors that affect restrained eaters. These are people,

who are typically women in the studies, who constantly try to refrain from breaking a

strict set of dieting guidelines. Because of this strict diet, these restrained eaters are at

risk of becoming temporarily disinhibited (Warren et al, 2005). At these times, emotional

and/or environmental factors can overwhelm the person causing a splurge of poor eating.

Based on the research by Warren et al (2005), the mass media is suspected to cause many

of the emotional factors. The experiment had female participants view emotionally

involved movies with commercials placed at a specific time during the movie, depending

on the sample group. Specifically the commercials were designed for one treatment

group to be exposed to diet advertisements or images of models representing the ideal-

thin. The second treatment group was offered a more neutral commercial tone as to not

remark or influence food or weight issues. Research indicated that people’s moods and

behavior vary dramatically depending on the participants’ environment. The women

23
participants in Warren et al (2005), research indicated that both high and low restraint

eaters were influenced by the mass media images highlighting the ideal-thin. As the

authors of the study suggest, it is important to evaluate our existing assumptions

regarding regulatory issues in advertising due to the effects that have been shown to

occur in research. Additionally, modern day issues, social responsibility and media

pressures put the emphasis of dieting on almost all people. This includes both males and

females in demographics that include children as young as kindergarten to adults.

Because this topic has permeated throughout society, future research is important

to determine how a more general group, including both men and women, react to food

advertising. The topic will continue to be a pressing issue in the future because of the

dramatic consequences facing those with overweight health issues.

Health Claims in Advertising

In recent decades there has been public concern that the health claims in

advertising can be misleading to consumers (Andrews, Burton & Netemyer, 2000). The

effects that nutritional claims in advertising produce are important because of the

abundance of claims on food products. Recent diet trends created a consumer culture that

draws people to the food or product option they believe will be healthier in comparison to

alternative products (Baltas, 2001).

Consumers have regularly been exposed to messages from a variety of sources

recommending an increase in vegetables, decrease in alcohol, or other types of messages

suggesting an action to better the health of the individual (Klassen, Wauer, Cassel, 1990).

The authors note this has been an ongoing occurrence for the past 30 to 40 years.

24
However, their research focuses specifically on the increase of health claims in the

1980’s. The 1980’s are important in regards to how health claims became prominent in

advertising. During this time period it became appealing for consumers to see that food

products were “better” for them than another product or type of food. At times, specific

health issues were the focus of the claims. In order to maintain the truthfulness of

advertising for the sake of the consumer, the Food and Drug Administration took an

active role in establishing guidelines. The research sought to explore whether the amount

of weight-loss claims targeted at women had changed over the prior three decades

leading up to the time of the research. Specifically looking at women’s magazines, the

results indicated that there was a significant increase in manufactured product health and

weight-loss claims during the 1980’s. The findings are important to track the impact that

health claims have over time, as well as the frequency that marketers are using an issue

such as weight-loss to influence consumers.

Baltas (2001) built on previous research to measure the effects that nutritional

information has on consumer choice. The research specifically reviews packaging

nutrition information and allows for marketers and policy makers to measure the

influence nutritional information has on purchasing behavior. Research on the topic is

important because of the number of products that list the nutritional information and the

increase in consumers who are willing to purchase healthier products. The author

recognized that consumers could be potentially confused by the technical information

associated with nutritional content. The recommendations from the review show that

there are differences between consumers based on what they are looking for in a food

product. The advertising message is said to create expectations of the healthfulness of

25
the product. From the expectation, it can either be confirmed or rejected based on the

nutritional values listed on the product.

Andrews, Netemeyer and Burton (1998) expanded on previous research focused

on nutrition label and packaging claims to include the advertising claim effect. The

purpose of the experiment was to measure to what extent nutritional content claims

generalize health perceptions across different disclosures, ad claim types, and nutrition

knowledge impact (Andrews, Netemeyer & Burton, 1998). The research was important

at the time because of FTC rulings (FTC, 1994) that attempted to build unity in food

labeling regulations between The Food and Drug Administration (FDA) and the FTC.

As consumers were drawn to more nutritious health products, health claims became more

prominent in marketing efforts (Andrews, Netemeyer & Burton, 1998). The experiment

measured how terms such as, “low-fat” or “healthy” affected the individual’s perception

of the food product as to lower risk for specific health diseases. Additionally, the

nutritional knowledge of the respondent was tested because of the hypothesis that

consumers with greater nutritional knowledge would have less favorable opinions of the

product without nutrient quality or content. The results found that those respondents with

high nutritional knowledge rated the advertised brand as being higher in fat content than

the respondents with low nutritional knowledge. Respondents with low nutritional

knowledge were more willing to accept the health claim or cues from the treatment.

Results also demonstrated an overgeneralization effect of the health claims whereby the

respondents transferred the health claims to other unmentioned health benefits of the food

product. Nutritional knowledge found mixed results, but suggested that high-knowledge

consumers are better able to interpret health information than low-knowledge consumers.

26
 A study by Adams and Geuens (2007) researches the effect that health slogans

have on the consumers’ perception of the food product. The study used an experimental

method to test 310 adolescents in Ghent, Belgium. The authors note that, based on

previous research from Andrews, et al. as discussed above, consumers could potentially

discount a positive health claim as simply an attempt to sell based on the marketers’

desired outcome. Groups of school children were randomly assigned to different

treatments. Treatments were designed to contain either a generally considered healthy

product (Cornflakes cereal) or a generally assumed unhealthy product (cookies). Both

products were then paired with healthy and unhealthy advertising slogans. The

measurements included the Health Perception of the Product (HPP), and Health Concern

(HC). The dependent variable was measured through the Attitude Toward the

Advertisement, and Attitude Toward the Product and Purchase Intention. The results

showed little difference in gender, which did not support the authors’ hypothesis. While

the results did not indicate a significant difference between claim types of healthy versus

unhealthy, the results were positive in both taste quality and the health slogan. The

authors suggest that this indicates that children were not as skeptical of advertising as

previous research suggested, and the respondents were more persuaded by advertising

than previously thought. The authors discuss the possibility that the children do not take

the persuasive advertising tactics into account when forming attitudes of a food product.

The results also found that the most effective ad claims were those that were placed on a

product that is already generally perceived as healthy. This reinforcement of the healthy

claim on an assumed product could pose a problem for consumers due to the importance

that companies place on food taste at any cost, even nutrition. Therefore, if consumers

27
rely on a health claim to determine the food products they purchase, there could be a

difference between the actual and perceived health qualities of the product. The present

study partially replicates the research by Adams and Guens (2007).

Government Regulations

The Federal Trade Commission released recommendations for the food and

beverage industry following a study on the food marketing practices to children and

adolescents (FDA, 2006; FDA, 2008). Released in July of 2008, the Commission had

obtained data from 44 major food and beverage marketers, who accounted for $1.6

billion in promotion dollars for their products in 2006. The 2006 Federal Trade

Commission’s review of the marketing practices of food companies was not the first time

the government had mandated regulations in response to public outcry (Seiders & Petty,

2007).

The FTC Food Marketing to Children Report (2008) explores the current status of

food marketing practices of major companies. The study examines the many different

methods used to promote brands and products to children. Examples are seen in the

product packaging, event marketing, online promotions, movie cross-promotion and

celebrities. The study names several sports celebrity superstars who endorse food

products in the media. As mentioned earlier, the report names such celebrity athletes as

LeBron James and Alex Rodriguez.

The FTC Report recommends several specific areas for further research and

improvement. For example, the report notes several companies’ efforts to improve on the

nutritional label packaging that has been designed for consumers to make more informed

28
choices that fit their needs when selecting food items. Because of the high level of

misinterpretation or lack of use of the nutritional labels noted by industry and academic

researchers, companies have developed their own methods of developing logos or claims

to assist the consumer. The report states the following, "Companies should conduct

research on the effectiveness of various labeling devices to determine how consumers

interpret such labeling and to identify those devices most effective at conveying

meaningful, truthful information".

While the FTC Report does not recommend specific research for the use of

celebrity athletes in advertising, it is important to understand due to the effect that a role

model can create on an audience. As the obesity concern increases and society seeks to

slow the frightening health trends, it becomes crucial to explore and better understand

how the common practice of using a role model or celebrity athlete to endorse a food

product affects the consumer’s perception of the product’s health quality in the absence

of any other health messages.

Summary

The advertising methods of food marketers are under close watch by the

government, consumer watch group and parents because of the potential contribution

advertising has in the obesity trends (Seiders & Petty, 2007). There is little empirical

research regarding celebrity athletes in food adverting. This study hopes to add to the

research body summarized in this chapter in an effort to build on the existing mass

communication theories and their role in food marketing.

29
 Based on the previous research, celebrity endorsers are frequently used in

advertising messages and have the ability to affect buyer behavior (Bush, Martin & Bush,

2004). The present study will replicate parts of the research by Adams and Guens (2007)

on the effects of health slogans indicated some influence in the perception of the food

product. Additionally, the celebrity endorser can serve as a peripheral cue in advertising

through the Elaboration Likelihood Model, which suggests that the level of personal

relevance can alter a person’s willingness to accept an advertising message (Petty &

Cacioppo, 1984).

Empirical research is important to understand the potential impact that the

presence of the celebrity athlete model has on food choices. Therefore, this researcher

hopes to determine the effects of a celebrity athlete in food advertising.

30
 CHAPTER III

METHODOLOGY

This chapter details the research objectives, independent variables, dependent

variables, hypotheses tested and the experimental design used to analyze the research

questions at hand. Additionally in this chapter are the elements represented in the

measurement scales, advertising treatments and sampling methods. Finally, the chapter

discusses the statistical analysis of the data collection and data processing.

Research Objective

The objective of this study was to better understand how the use of celebrity

athlete models in food advertising affects the perception of the healthiness of advertised

food among college students.

This study aimed to build on the past research in the area of food advertising and

health perceptions of food advertisements. Specifically, the framework and design

partially replicated a study by Adams and Geuens (2007), which attempted to determine

the effects that advertising slogans have on food product health perceptions.

31
 The results from the present research may be beneficial to advertising regulators

by offering insight into the potential consequences and impact of food advertising

featuring celebrity athletes. Due to the negative health implications of obesity, this

research is important for consumers, the government and the food industry.

Variables and Hypotheses

The independent variable in this experiment was the presence of a celebrity

athlete model in the ad. Two celebrity athletes were tested -- Olympian Michael Phelps

in one treatment, and Super Bowl MVP Eli Manning in another. Both athletes appear in

a cereal print ad. The dependent variables were health perceptions of the product (HPP),

the health concern (HC), attitude toward the ad (AttAd) and purchase intent (PI). The

treatments using the celebrity athletes measured their source-credibility (Ohanian, 1991).

Previous empirical research suggests that consumers rely on the health claims

made in the advertising of food products (Andrews, Netemeyer & Burton, 1998; Baltas,

2001). The research debates the effects that health claims have on consumer behavior

when choosing food options. Further, the Elaboration Likelihood Model (ELM) suggests

that celebrities in ads may serve as a peripheral cue enhancing the attributes of a product

being advertised (Petty & Cacioppo, 1984).

Therefore, it was hypothesized that consumers will rate food products healthier

when there is a celebrity athlete pictured in the ad.

H1: Health perception of the food product (HPP) will be higher for the ads

containing Michael Phelps and Eli Manning than for the ad featuring the

unknown model.

32
 The Elaboration Likelihood Model (Petty & Cacioppo, 1984) framework can be

applied to the present experiment. The dependent variable of health concern (Adams &

Guens, 2007) indicates the respondent’s personal relevance toward nutritional value in

food choices. A high level of personal relevance should engage the respondent in

detailed elaboration regarding the actual health value of the food product advertised.

Based on the theory, if the respondent engages in a significant amount of elaboration

regarding the product’s nutritional value, he or she will rate the food product less healthy

than those not concerned with the nutritional value of foods. Those respondents with low

personal relevance toward health concern (HC) would have been influenced by the

presence of Phelps or Manning in the ad and will rate the product’s nutritional value

higher. Therefore,

H2: There will be an inverse relationship between health concern and the

health perception of the food in the ads containing Michael Phelps or Eli

Manning.

The attitude toward the ad (Adams & Geuens, 2007; Gresham & Shimp, 1985)

explores the effects that a celebrity endorser can have on the appeal of an ad or product.

The use of a role model has been shown to be effective in creating positive attitudes

among consumers toward the ad.

H3: Attitude toward the ad mean scores will be higher for the ads containing

Michael Phelps and Eli Manning than for the ad featuring the unknown

model.

33
 The positive attitudes toward the ad are related to the purchase intent of the

product being advertised (Mitchell & Olson, 1981; Mitchell, 1986). Therefore,

H4: There will be a positive relationship between attitude toward the ad and

purchase intent in the ads featuring Phelps or Manning.

The research relating to the source credibility of a celebrity endorser (Amos,

Holmes and Clinton, 2008) demonstrates how negative information regarding a celebrity

can have a negative impact on perception of the advertising featuring the celebrity, and

therefore decrease purchase intent. The media coverage highlighting Phelps drug-use just

weeks before the research collection could be considered negative information.

Therefore,

H5a: Participants will rate the source credibility of Michael Phelps lower

than the source credibility of Eli Manning.

H5b: Participants will rate attitude toward the ad lower for the ad containing

Phelps versus Manning.

H5c: Participants will rate purchase intent lower for the ad containing

Phelps versus Manning.

Experiment Design

The experiment used a basic post-test only with control variable group design. It

explored the relationship between the independent variable of the celebrity athlete and the

dependent variables of the health perceptions of the product (HPP), health concern (HC),

attitude toward the ad (AttAd), purchase intent (PI).

34
Procedure

The researcher recruited students from Oklahoma State University undergraduate

journalism classes to participate in the research project by contacting professors.

Professors who were willing to participate were supplied with a script to read to students

explaining the purpose of the research and highlighting that their participation was

completely voluntary. Additionally, a signup sheet was provided to pass around to those

students willing to participate. The only information students were asked to offer was an

email address where the survey link was sent via email.

Three online questionnaires were developed on the Internet survey site, Survey

Monkey. One questionnaire contained the experimental ad with Michael Phelps and the

other contained and identical ad featuring Eli Manning. Finally, the control ad consisted

of the same ad with a non-celebrity generic male student. Otherwise the questionnaires

were identical, except for the source credibility scale questions, which asked the opinions

of the respondent in regards to the celebrity athlete model in the ad that they had just

seen.

Students who agreed to participate in the research were randomly assigned to one

of the treatment ads or to the control ad. Students were then emailed a link to one of the

three questionnaires according to their assignment.

The first page of the questionnaire explained the survey and informed the

participants of their voluntary participation. Additionally, it reminded the students that

they could close out of the survey at any point, if they desired. Following the questions

regarding their opinions of the advertising treatments, participants were asked

demographic and basic lifestyle questions. The questions were in the format of semantic

35
differential, Likert rating scale, multiple choice questions for demographic details, or

participant entered text responses. All responses were automatically tabulated and loaded

into a database for future analysis.

The procedure was first pre-tested among a small group of college students to

make sure the questionnaire was clear to the respondents and the data collection was

working properly.

Advertising Treatments

The print advertising treatments were designed by replicating features of a Total

Raisin Bran cereal box cover. Cereal was selected as the food product for the experiment

because it is a category that is widely common among households (Adams & Geuens,

2007). The price of a box of cereal is accepted as relatively affordable for most people,

and cereal is frequently associated with both children and adults.

The cereal box cover was altered in Adobe Photoshop to create print ads using

similar claims and graphics produced on the original cereal box design. The brand name

of the cereal was changed to the generic name “Crunchy Flakes” and claims were

adjusted to be non-health specific. Images including a purple full color background, the

cereal bowl, and the milk and cereal were used from the original cereal box scan to

maintain consistency with what people would typically see on the store shelf or in

advertising. The headline of the advertisement was “A Great Way to Start Your Day!”

Next to the “Crunchy Flakes” title, the term “New” was added to stay consistent with our

purpose of asking the opinions of the research respondents. The copy was modified to

read, “Crunchy Wheat & Bran Flakes with Plump, Juicy Raisins”. The three variations

36
of the advertising treatment maintained consistent copy and layout design elements,

except for the image of Michael Phelps, Eli Manning and the generic male student.

Additionally, both of the celebrity athletes had a cut line with their name and

achievement. For Phelps the cut line read, “Olympian Michael Phelps”. Manning’s cut

line read, “Super Bowl MVP Eli Manning”. There was no cut line for the generic male

student.

The control advertising treatment was an exact replication of the celebrity athlete

treatment advertisements; however, the image of a celebrity athlete was replaced with the

non-celebrity model. All graphics and claims otherwise remained consistent. All

slogans, colors and space layouts were otherwise identical.

The researcher determined that it was important to have the entire electronic

image on the questionnaire screen so the participants did not have to scroll down the

page. Therefore, the ad was reduced in size but proportions of a standard print ad were

maintained so as not to distort the advertisement.

Research Instruments

Perceptions of the healthiness of the cereal were measured using Adams and

Geuens’ (2007) health perception of the product (HPP), and the health concern (HC)

scale. The likeability of the ads were measured by the attitude toward the ad scale

(Gresham & Shimp, 1985). The questionnaires containing Phelps and Manning used

Ohanian’s (1991) source-credibility Scale.

The HPP and the attitude toward the advertisement were measured on a 5-point

semantic differential scale. The HPP contained six items. There were five items related

37
to the attitude toward the advertisement. Both were anchored with opposing statements

at each end of the scale. Health concern was measured by nine items on a five point

Likert scale ranging from 1=strongly disagree to 5=strongly agree.

The source credibility scale used to measure the respondent perceptions of both

Phelps and Manning contained 16 items. Respondents viewing the experimental ad with

Michael Phelps were asked if the respondent was aware of the media coverage

surrounding Phelps that occurred just weeks before the research launched.

The demographic questions captured information regarding the participants’ age,

gender, race, university, student classification, whether they eat breakfast or not and their

exercise frequency. The questionnaire containing Eli Manning had a total of 43 items,

while the questionnaire containing Phelps had 44 due to the additional question

surrounding the drug use. The non-celebrity male model questionnaire had a total of 27

items, because it did not include the source credibility scale.

Respondents

Students were recruited for this study from several undergraduate journalism

courses at Oklahoma State University. The respondents represented a non-random

representative sample of university students. However, the courses were journalism

classes due to the convenience of contacting the professors teaching those courses. Also,

college students are easier to sample than young children because of strict regulations

when working with children under the age of 18. The sample of college students was

chosen to maintain a fairly homogenous group under study.

38
 College-aged students are an important demographic for government regulators

and marketers to better understand. Because of their age, they stand to gain the most

from changes in behavior that could result in more nutritious eating and, therefore, a

lesser risk of the repercussions of poor health. Today’s college students have grown up

in the recent decades when health concerns have gained attention in popular culture. For

this reason, college students will have a unique perspective on the health quality of

advertising that contains food products. Additionally, this group has been exposed to a

greater number of advertisements than children in previous generations. The amount of

potential spending that college students represent makes them of great importance to

marketers who are constantly attempting to find ways to sell them products and brands

(Bush, Martin & Bush, 2004).

Data Collection, Processing and Analysis

The data collected from the three questionnaires were combined to create one data

set. The Internet site organized the responses and allowed the data to be downloaded into

an Excel file. The Excel file was then loaded into the SPSS system for detailed analysis

and statistical tests. The variable scales were reverse coded so that they could be

averaged to produce one score for each variable.

ANOVA’s were used to test hypothesis one and three. A Pearson Correlation Coefficient

was used to test hypothesis two and hypothesis four. T-tests were used to test hypotheses

five a, five b and five c.

39
 CHAPTER IV

FINDINGS

The following chapter details the findings of the experiment to measure the health

perception of food products that celebrity athletes may create in food advertising. The

experiment measured the respondent’s health perception of the product, attitude toward

the ad, source credibility of the celebrity athletes, health concern and purchase intent.

The respondents were also asked whether or not they ate breakfast, how often they

exercised and demographic questions.

Three variations of an advertisement for a fictional cereal maintained consistent

copy and layout design elements, except for the image of the model. One ad featured

Michael Phelps, one Eli Manning and one a generic male student. Both of the celebrity

athletes had a cut line with their name and achievement. For Phelps the cut line read,

“Olympian Michael Phelps”. Manning’s cut line read, “Super Bowl MVP Eli Manning”.

There was no cut line for the generic male student. The treatments were randomly

assigned to a group of mass communication students at Oklahoma State University.

Three measurement scales were used in the experiment questionnaires to measure

the health perception of the product, attitude toward the ad, and health concern of the

respondent. A fourth scale was present for the celebrity athlete treatments to measure the

40
source-credibility of the athlete. The health perception of the product (HPP) was

designed by Adams and Geuens (2007). The HPP contained six items on a five point

semantic differential scale. The responses were measured with five indicating the most

positive opinion of the product and one indicating the most negative. The attitude toward

the ad scale was adapted from Gresham and Shimp (1985). The scale contained five

items, also on the five point semantic differential scale with five indicating the most

positive response and one indicating the most negative response. The health concern

(HC) scale, also Adams and Geuens (2007), contained 10 items on a Likert-type scale.

The HC was measured on a five point scale with a score of five indicating the most

positive response “strongly agree” and one indicating the most negative response

“strongly disagree”. The source credibility scale (Ohanian, 1993) measured the

respondents’ opinions of the celebrity athletes’ attractiveness, expertise and

trustworthiness. The scale, used for both Phelps and Manning, contained 16 items on a

five point semantic differential scale with five indicating the most positive response and

one indicating the most negative. Negatively phrased statements on the scales were

reverse coded and items for each scale were averaged to produce a score for each

variable.

Respondent Profile

A total of 106 student respondents from Oklahoma State University mass

communication classes participated in the study. Of the respondents, 73 percent were

female and 27 percent were male. The respondents were 82 percent white, eight percent

Native American, four percent African American, and two percent international students.

41
Hispanic, Asian American, and Other accounted for an additional one percent each. The

reported average age was 20.5 years. Roughly half of the participants reported eating

breakfast regularly with an additional 38 percent reporting that they sometimes ate

breakfast. Roughly one-third (34%) of the respondents reported exercising two to three

times per week and more than one-third (36%) reported exercising up to four or five

times per week.

When questioned regarding their familiarity with the celebrity athlete model, one

being the least and five being the most familiar, Phelps measured at 4.33 while Manning

measured lower at 3.62. Almost all (97%) of the respondents were aware of the Phelps

drug-use report in the media.

Health Perception of the Product (HPP)

The research hypothesis 1 predicted that the cereal in the ads featuring Michael

Phelps and Eli Manning would be perceived as healthier than the ad featuring the generic

model. The analysis found no statistically significant difference in the health perception

among the three treatments (F=1.785; p=.173; Phelps HPP=3.44; Manning HPP=3.46;

non-celebrity model HPP=3.72). Therefore, hypothesis 1 was not supported.

Health Concern (HC)

Hypothesis 2 predicted an inverse relationship between the personal health

concern of the respondents and their health perception of the cereal in the ad for the

treatments containing Phelps and Manning. The two ads were analyzed separately using

a Pearson Correlation Coefficient, which found no statically significant relationship

42
between the respondents’ health concern and the health perception of the food for either

of the ads containing celebrities (Phelps ad, r=-.090; p=.602; Manning ad, r=-.138;

p=.436). Hypothesis 2 was not supported.

Attitude Toward the Ad

Hypothesis 3 predicted that the ads featuring the celebrity athletes would produce

more positive attitudes when compared to the ad featuring the non-celebrity model. An

ANOVA revealed that the ad featuring Manning (AAd=2.96) generated significantly

more positive attitude toward the ad scores (F-3.64; p=.03) than the ad that featured

Phelps (AAd=2.40) and the ad containing the non-celebrity model. Hypothesis 3 was

partially supported.

Purchase Intent

Hypothesis 4 predicted a positive relationship between the attitude toward the ad

and the purchase intent of the product in the ads featuring Phelps or Manning. The

responses to both ads were analyzed separately using a Pearson’s Correlation Coefficient.

The findings revealed a statistically significant positive relationship between the attitude

toward the ad and purchase intent for both the ad containing Phelps (r=.431; p=.009) and

for the ad featuring Manning (r=.411; p=.018). Therefore, the findings support

hypothesis 4.

43
Michael Phelps Drug-use in the Media

Due to the negative media surrounding Michael Phelps’ drug-use that circulated

prior to the data collection, hypothesis 5(a), 5(b) and 5(c) predicted that Phelps’ source

credibility score, attitude toward the ad score and purchase intent score would rate below

that of Eli Manning. The source credibility findings revealed no significant difference

(t=1.1434; p=.156) between Phelps (3.33) and Manning (3.58). The finding for the

attitude toward the ad, hypothesis 5(b), found a significant difference (t=2.473; p=.016)

between the ad featuring Manning (AAd=2.96) and that of the ad featuring Phelps

(AAd=2.4). Finally, in hypothesis 5(c) the purchase intent between the ads featuring

Phelps (PI=2.36) and the ad featuring Manning (PI=2.55) revealed no significant

statistical difference (t=.673; p=.503). The findings of hypothesis 5 reveal that 5(a) was

not supported, 5(b) was supported and 5(c) was not supported.

44
 CHAPTER V

CONCLUSION

Health trends over the past 30 years indicate an increase in the number of

Americans suffering from obesity. The factors causing such an increase include

changes in our diet and a lack of exercise. Support for both arguments can easily be

found in the amount of calories consumed and the hours Americans spend in front of the

television each day (Hardus, van Vuuren, Crawford & Worsley, 2003). Young people

are not immune to the obesity trends as seen by the statistics from the Centers for Disease

Control and Prevention. Parents and regulators have been quick to call for action to stop

the obesity epidemic from continuing to impact the lives of young adults and children.

The Federal Trade Commission’s 2008 report (FTC, 2008) questions food

advertising practices. The FTC’s investigation and analysis highlight the importance and

magnitude of the obesity problem. The report also draws attention to the need to better

understand the impact of food advertising on American’s food choices.

This study sought to determine if a celebrity athlete endorser in food advertising

affects the health perception of a food product. The research used a post-test only with

control ad experimental method to measure the effects of print food advertisements

containing a celebrity athlete.

45
 The questionnaire was completed by a convenient sample of 106 students from

Oklahoma State University-Stillwater campus. The questionnaire was distributed during

February 2009 to the college students who were willing to participate. The respondents

were randomly assigned to one of the three treatments for a cereal. The treatment

versions contained identical layout and copy, however, the endorser was manipulated in

all three treatments. One version contained Olympian Michael Phelps, one contained

Super Bowl MVP Eli Manning, and the third treatment contained an unknown male

student. Each ad treatment was identical except for the image of the endorser and the

respective cutline used to identify the respective professional athletes. The

questionnaires contained scales to measure the health perception of the product (HPP),

attitude toward the ad (AttAd), source credibility for the ads featuring Phelps and

Manning, health concern (HC) and purchase intent. The data from all three

questionnaires were combined into one file containing all of the responses and analyzed

using t-tests, ANOVAs and tests of correlation.

Summary

The primary research goal of this study was to determine if the presence of a

celebrity athlete transferred the attribute of healthiness to the food product being

advertised. It also sought to determine how personal health concern might influence the

perception of the healthiness of the food product. The results of this study found no

effect on the perception of food healthiness when a celebrity athlete was used in

46
advertising. Likewise, there was no evidence of an influence created by the respondent’s

personal level of health concern.

Therefore, the celebrity athletes, Michael Phelps and Eli Manning, used in the

study did not convince the respondents that the food product was healthier because of

their presence alone. However, given the fairly high rating of health perception of the

product to all three of the treatments (Phelps = 3.44, Manning = 3.46, Unknown Male =

3.72), there is the possibility that the treatments conveyed a sense of healthiness to the

respondents. The contributing factors to this perception could have been produced by the

design of the ad or even the pre-existing perception respondents had toward the general

healthiness of cereal.

There was a relationship between purchase intent and attitude toward the ad,

which suggests that those who like the ad more are more likely to buy the product. When

comparing Phelps’ and Manning in terms of source credibility and purchase intent of the

product, no significant differences were found; however, respondent’s rated the ad

featuring Manning significantly higher than the Phelps or generic model ads.

Discussion

The finding of this study are not consistent with the findings of the previous

research by Adams and Geuens (2007) who found that healthy slogans used with a

generally considered healthy product increased the health perception of the product. The

present study, which replicated parts of Adams and Geuens (2007) study, did not find an

effect on health perception of the product when a healthy endorser was matched to a

healthy product. Beyond the difference in the independent variable, there were many

47
differences between the two studies, including the age of the respondents. Adams and

Geuens (2007) studied school children and the present research used college students.

The present findings were also inconsistent with the research by Andrews,

Netemeyer and Burton (1998) which indicated that the less nutritional knowledge of a

respondent, the more likely the respondent would be to respond favorably to health

slogans. Additionally, Andrews et al. found an overgeneralization of healthiness to

characteristics of the product. This overgeneralization occurred when respondents

transfer the health claims or qualities to other areas of the product that are not present.

Respondents in the present study did not overgeneralize the healthy qualities of the

celebrity athlete to the food product. This may be because the celebrity athletes chosen

were not perceived as healthy. For example, Phelps’ McDonald’s junk-food habit was

widely publicized and may have altered consumers’ perceptions of his healthiness.

The findings of this study were inconsistent with the previous ELM research

(Petty, Cacioppo & Goldman, 1981) regarding personal involvement in persuasion

effectiveness. The low health concern in the present study did not influence the

respondents’ reliance on peripheral cues, such as the models in the ads. This was also

inconsistent with Adams and Geuens (2007) study which found health concern to be a

mediating factor in the perception of the product.

The study revealed information regarding the celebrity athletes selected for the

advertising treatments. The ad featuring Eli Manning received more positive attitude

toward the ad scores than the control ad. This finding is consistent with the findings by

Bush, Martin and Bush (2004), which indicate that celebrities can influence consumer

perceptions. The Manning ad also outscored the Phelps ad, which may be explained by

48
the recent negative media attention surrounding Phelps. The results of the Amos, et al.

(2008) study found that negative publicity of a celebrity endorser has a negative impact

on the attitude toward the endorser and toward the product.

Attitude toward the ad scores for the ads featuring Phelps and Manning were

positively related to the purchase intent of the product. This finding is similar to the prior

research (Mitchell & Olson, 1981; Mitchell, 1986) relating to attitude toward the ad. The

research from previous studies, and the present, suggest that the more a person likes the

advertising for a product, the more willing the respondent is to say that they will purchase

the product.

The present research found little difference among the three treatments. This is

inconsistent findings of Lockwood and Kunda (1997) who found that the success of a

celebrity endorser, or superstar, could either encourage or discourage respondents. In the

case of Phelps and Manning, their presence in the ads did not create significant

encouragement or discouragement in the results.

Implications

The results of the present research, although not generalized to entire population,

would suggest that food advertising featuring celebrity athlete endorsers is not likely to

create a perception of healthiness for the food product. The findings are important to

food marketers who have the goal of generating high sales volumes, and to the regulators

who are seeking a balance between the advertising of food products and the negative

effects of high-calorie food consumption.

49
 Additionally, the timing of the present research suggests the acceptance or

ambivalence of drug-use among young sports stars. This information is important for the

companies who connect their products with Michael Phelps, or for other marketers whose

brands might be affected by the negative media attention occasionally received by

celebrity endorsers.

Limitations

Several limitations should be considered when analyzing the results of the study.

The limitations include such elements as the research population, sample size,

advertisement type, experimental conditions and research instruments.

Population. The convenient sampling method used for the research experiment

cannot be generalized to the entire population of college students. Previous research

supports the influence of celebrities on young people’s impressions and behaviors. The

findings of the present study are limited to students of one university in Stillwater,

Oklahoma and do not represent the overall population of college students.

Respondents. The research respondents were mass communication students,

which should be considered a limitation to the research. Mass communication students

may have a more sophisticated understanding of advertising and, therefore, may have

been more likely to be skeptical of the advertising message.

50
 Sample Size. The size of the treatment groups should be considered as a

limitation because of the impact such sample sizes could have on the sub-analysis. The

treatment groups were relatively similar in size between Phelps (N=36), Manning (N=34)

and the control ad (N=36).

Advertisement Type. The results from the research are limited by food product

selected for the advertising. Cereal was selected because it is a common food for most

people regardless of age or preferences. However, because cereal is such a well known

product it could be argued that people have preconceived impressions of cereals based on

the graphical layout and design of the advertising. Therefore, the results of the

experiment cannot be generalized to all types of foods.

Experimental Conditions. The research participants were asked to evaluate their

opinions of the advertising treatments through the use of an online survey site. The

nature of the research created an artificial environment whereby respondents viewed the

ad and immediately answered questions about their perception of the health quality of the

product, their attitude toward the ad and other questions. Because the process of print

advertising is subtler through newspapers and magazines, and developed over several

viewings, it is difficult to determine if the same results would have been produced if

participants had viewed the ads in a more natural environment. Therefore, the

experimental conditions should be considered a limitation of the research project.

51
 Treatments. The research experiment was initially designed to test the influence

that celebrity athlete Michael Phelps created coming off of his recent Olympic success.

However, just weeks before the experiment was scheduled to launch, the scandal

involving his drug use hit the media. As the results of the research question regarding

Phelps media exposure indicate, almost everyone was aware of his negative media

attention. Phelps’ image through the media was altered as he quickly acknowledged and

attempted to quiet the negative attention. Regardless, the research respondents’

perception of Phelps should be considered a limitation.

Type of Celebrity. The research examined two male athletes who were both in

their 20’s and cannot be generalized to all celebrity athletes. This should be considered a

limitation to the research because the results do not consider the impact that other

celebrity endorsers might have.

Future Research

Children. Young children are likely to be influenced by the peripheral cues in

advertising more than adults (Roe, Levy & Derby, 1999). This is because young children

lack an understanding of advertising and are not as skeptical as adults who have a better

understanding of advertising goals. Future research should examine the effects that

young children have when being exposed to food advertising using celebrity athletes.

Celebrity Athletes. It is important for future research to explore the relationship

that young people have with different celebrity athletes. The athletes in the present

experiment were selected for their opposing images to the public. It is possible that other

athletes from different sports could have a great effect on the health perception of the

52
product. Future research should study how other types of athletes and female celebrity

athletes influence the healthiness perception of food products to young people.

Advertising Type. The print advertising messages that were tested in the present

research do not have the same effectiveness as print advertising may have with other

generations. Young people today are heavy users of the Internet and other new media

channels. Because of this, young people today might not respond as well to advertising

messages that are not similar to the advertising messages they receive daily. Future

research should incorporate new media into the advertising types to better mirror the type

of messages that young people are used to seeing in their environment.

Conclusion

As the importance of reversing the obesity trend continues to become more of a

priority for the government and marketers, it is important to determine what effects food

advertising messages have to people who regularly view these ads. The FTC’s (2008)

review on food adverting shows the large amounts of money that companies use for

advertising food products to young people. However, the report does not make specific

recommendations for further research, on the use of celebrity athletes and their effect on

consumer perceptions of food products.

The results of the present study indicate that college students in this study were

not influenced by the presence of a celebrity athlete in food advertising. This is

important for regulators and researchers who will further explore the research report

developed by the FTC in 2008 in an effort to slow the obesity rates. However, because

the respondents did rate the ad containing Eli Manning significantly more appealing than

53
the ad containing the control ad and Michael Phelps, it is important for regulators to

explore how positive attitudes in advertising can influence purchase behavior over time.

54
 REFERENCES

Adams, L., & Geuens, M. (2007, March). Healthy or unhealthy slogans: that's the
question... Journal of Health Communication, 12(2), 173-185. Retrieved
November 30, 2008, from MEDLINE database.

Almond, S. (2000, October). A11. The influence of the media on eating disorders.
Journal of Human Nutrition & Dietetics, 13(5), 367-368. Retrieved November 27,
2008, from Academic Search Elite database.

Altman, D., Avery, J., Barnes, B., Bendinger, B. & Ganahl, D. (2004). Advertising and
The Business Of Brands. Copy Workshop; 2 edition.

Amos, C., Holmes, G., & Strutton, D. (2008, May). Exploring the relationship between
celebrity endorser effects and advertising effectiveness. International Journal of
Advertising, 27(2), 209-234. Retrieved November 30, 2008, from Business Source
Elite database.

Andrews, J., Burton, S. & Netemeyer, R. (2000). Are some comparative nutrition claims
misleading? The role of nutrition knowledge, ad claim type and disclosure
conditions. Journal of Advertising, (3), 29-42.

Andrews, J., Netemeyer, R. & Burton, S. (October 1998). Consumer generalization of

nutrient content claims in advertising. Journal of Marketing, 62-75.

Baltas, G. (2001). The effects of nutrition information on consumer choice. Journal of

Advertising Research, 57-63.

Bandura, A. (1986). Social Learning Theory. Englewood Cliffs, NJ: Prentice Hall.

Batada, A., & Wootan, M. (2007, July). Nickelodeon markets nutrition-poor foods to
children. American Journal Of Preventive Medicine, 33(1), 48-50. Retrieved
November 30, 2008, from MEDLINE database.

Batada, A., Seitz, D., Wootan, M. Story, M. (2008, April). Nine out of 10 food
advertisements shown during Saturday morning children's television
programming are for foods high in fat, sodium, or added sugars, or low in
nutrients. Journal Of The American Dietetic Association, 108(4):673-678.

Bush, A., Martin, C., & Bush, V. (2004, March). Sports Celebrity Influence on the
Behavioral Intentions of Generation Y. Journal of Advertising Research, 44(1),
108-118. Retrieved November 30, 2008, from Business Source Elite database

CDC (2006, April). Obesity still a major problem. National Center for Health Statistics.

55
CHOI, S., & RIFON, N. (2007, April). Who Is the Celebrity in Advertising?
Understanding Dimensions of Celebrity Images. Journal of Popular Culture,
40(2), 304-324. Retrieved November 30, 2008, doi:10.1111/j.1540-
5931.2007.00380.

Colditz, G. (1999, November). Economic costs of obesity and inactivity. Medicine And
Science In Sports And Exercise, 31(11 Suppl), S663-7. Retrieved December 1,
2008, from MEDLINE database.

Dietz, W. (1998, March 2). Health consequences of obesity in youth: Childhood...

Pediatrics, 101(3), 518. Retrieved December 1, 2008, from MasterFILE Premier
database.

Ebbeling, C., Pawlak, D., & Ludwig, D. (2002, August 10). Childhood obesity: public-
health crisis, common sense cure. Lancet, 360(9331), 473. Retrieved December 1,
2008, from Academic Search Elite database.

ESPN. With Phelps' quest for gold almost over, NBC worries ratings will drop. Retrieved
December 1, 2008, from
http://sports.espn.go.com/oly/summer08/swimming/news/story?id=3538804

Farrell, K., Karels, G., Monfort, K. & McClatchey, C. (2000). Celebrity performance and
endorsement value: The case of Tiger Woods. Managerial Finance, 26, 1-15.

Frewer, L., Howard, C., Hedderley, D., & Shepherd, R. (1997, December). The
elaboration likelihood model and communication about food risks. Risk Analysis:
An Official Publication Of The Society For Risk Analysis, 17(6), 759-770.
Retrieved November 30, 2008, from MEDLINE database.

FTC (2008, July). Marketing food to children and adolescents: A review of industry
expenditures, activities, and self-regulation. A Report to Congress.

Fullerton, J. & Kendrick, A. (2006). Advertising's war on terrorism: The story of the U.S.
State Department's Shared Values Initiatives. Spokane: Marquette Books.

Gresham, L., & Shimp, T. (1985). Attitude toward the advertisement and brand attitudes:
A classical conditioning perspective. Journal of Advertising, 14(1), 10-17.
Retrieved November 30, 2008, from PsycINFO database.

Hardus, P., van Vuuren, C., Crawford, D., & Worsley, A. (2003, December). Public
perceptions of the causes and prevention of obesity among primary school
children. International Journal of Obesity & Related Metabolic Disorders,
27(12), 1465-1471. Retrieved December 1, 2008, doi:10.1038/sj.ijo.0802463

56
Heaner, M. America, 100 Percent Fat. MSN Health & Fitness. Retrieved December 1,
2008, from http://health.msn.com/fitness/articlepage.aspx?cp-
documentid=100168553

Henley, J. (2008, August 15). The need for greed. The Guardian (London). Retrieved
from http://www.lexisnexis.com/us/lnacademic/search/homesubmitForm

Hovland, C., Janis, I. & Kelley, H. (1953). Communications and persuasion:

Psychological studies in opinion change. New Haven, CT: Yale University

Janoff, B. (2008, January 28). Super Bowl XLII Puts Another Manning In the Marketing
Mix. Brandweek, 49(4), 10-10. Retrieved February 22, 2009, from Business Source
Elite database.

Jones, L., Sinclair, R., & Courneya, K. (2003, January). The effects of source credibility
and message framing on exercise intentions, behaviors and attitudes: An
integration of the elaboration likelihood model and prospect theory. Journal of
Applied Social Psychology, 33(1), 179-196. Retrieved November 30, 2008,
doi:10.1111/j.1559-1816.2003.tb02078.

Kautiainen, S., Koivusilta, L., Lintonen, T., Virtanen, S., & Rimpelä, A. (2005, August).
Use of information and communication technology and prevalence of overweight
and obesity among adolescents. International Journal of Obesity, 29(8), 925-933.
Retrieved November 4, 2008, doi:10.1038/sj.ijo.0802994

Kautiainen, S., Koivusilta, L., Lintonen, T., Virtanen, S., & Rimpelä, A. (2005, August).
Use of information and communication technology and prevalence of overweight
and obesity among adolescents. International Journal of Obesity, 29(8), 925-933.
Retrieved November 30, 2008, doi:10.1038/sj.ijo.0802994

Kim, Y., & Na, J. (2007, July). Effects of celebrity athlete endorsement on attitude
towards the product: The role of credibility, attractiveness and the concept of
congruence. International Journal of Sports Marketing and Sponsorship, 8(4),
310-320. Retrieved November 30, 2008, from PsycINFO database.

King, L., & Hill, A. (2008, July 26). Magazine adverts for healthy and less healthy foods:
effects on recall but not hunger or food choice by pre-adolescent children.
Appetite, 51(1), 194-197. Retrieved November 30, 2008, from MEDLINE
database.

Klassen, M., Wauer, S., & Cassel, S. (1990, December). Increases in health and weight
loss claims in food advertising in the eighties. Journal of Advertising Research,
30(6), 32-37. Retrieved November 30, 2008, from PsycINFO database.

57
Kreuter, M., Bull, F., Clark, E., & Oswald, D. (1999, September). Understanding how
people process health information: A comparison of tailored and nontailored
weight-loss materials. Health Psychology, 18(5), 487-494. Retrieved November
30, 2008, doi:10.1037/0278-6133.18.5.487

Larimore, W. & Flynt, S. (2005). Super Sized Kids: How to Rescue Your Child From the
Obesity Threat New York: Center Street.

Lockwood, P., & Kunda, Z. (1997, July). Superstars and me: Predicting the impact of
role models on the self. Journal of Personality and Social Psychology, 73(1), 91-
103. Retrieved November 30, 2008, doi:10.1037/0022-3514.73.1.91

Lowery, S. & DeFleur, M. (1995). Milestones in mass communication research: Media

effects (3rd ed.). White Plains, NY: Longman.

Mikhailovich, K., & Morrison, P. (2007, December). Discussing childhood overweight

and obesity with parents: a health communication dilemma. Journal Of Child
Health Care: For Professionals Working With Children In The Hospital And
Community, 11(4), 311-322. Retrieved November 30, 2008, from MEDLINE
database.

Mitchell, A. (1986), "The Effect of Verbal and Visual Components of Advertisements on

Brand Attitudes and Attitude Toward the Advertisement," Journal of Consumer
Research, 13 (1), June, 12-24.

Mitchell, A., Olson, J. (1981), Are Product Attribute Beliefs the Only Mediator of
Advertising Effects on Brand Attitude? Journal of Marketing Research, 19
(August), 318-32.

Nashawaty, C. (2009, February). Michael Phelps took a hit. His future as a celebrity
pitchman probably won't. Entertainment Weekly, (1034), 13. Retrieved February
22, 2009, from Research Library database. (Document ID: 1639474271).

Nestle, M. (2007). Food Politics: How the Food Industry Influences Nutrition, and
Health, Revised and Expanded Edition (California Studies in Food and Culture).
Berkeley: University of California Press.

Ohanian, R. (1991, February). The impact of celebrity spokespersons' perceived image on

consumers' intention to purchase. Journal of Advertising Research, 31(1), 46-54.
Retrieved December 1, 2008, from Business Source Elite database.

O'Keefe, D. J. (2002). Persuasion Theory & Research (2nd ed.). Thousand Oaks, CA:
Sage Publications.

58
Olshansky, S., Passaro, D., Hershow, R., Layden, J., Carnes, B., Brody, J., et al. (2005,
March 17). A Potential Decline in Life Expectancy in the United States in the 21st
Century. New England Journal of Medicine, 352(11), 1138-1145. Retrieved
December 1, 2008, doi:10.1056/NEJMsr043743

Olson, J., & Zanna, M. (1993). Attitudes and attitude change. Annual Review of
Psychology, 44, 117-154. Retrieved November 30, 2008,
doi:10.1146/annurev.ps.44.020193.001001

Petty, R., & Cacioppo, J. (1979, October). Issue involvement can increase or decrease
persuasion by enhancing message-relevant cognitive responses. Journal of
Personality and Social Psychology, 37(10), 1915-1926. Retrieved April 8, 2009,
doi:10.1037/0022-3514.37.10.1915

Petty, R., & Cacioppo, J. (1984, January). The effects of involvement on responses to
argument quantity and quality: Central and peripheral routes to persuasion.
Journal of Personality and Social Psychology, 46(1), 69-81. Retrieved November
30, 2008, doi:10.1037/0022-3514.46.1.69

Petty, R., Cacioppo, J., & Goldman, R. (1981, November). Personal involvement as a
determinant of argument-based persuasion. Journal of Personality and Social
Psychology, 41(5), 847-855. Retrieved November 30, 2008, doi:10.1037/0022-
3514.41.5.847

Petty, R., Cacioppo, J., & Schumann, D. (1983, September). Central and peripheral routes
to advertising effectiveness: The moderating role of involvement. Journal of
Consumer Research, 10(2), 135-146. Retrieved November 30, 2008,
doi:10.1086/208954

Robinson, T. (2001, August). Television viewing and childhood obesity. Pediatric

Clinics Of North America, 48(4), 1017-1025. Retrieved December 1, 2008, from
MEDLINE database.

Roe, B., Levy, A., & Derby, B. (1999, Spring99). The Impact of Health Claims on
Consumer Search and Product Evaluation Outcomes: Results from FDA
Experimental Data. Journal of Public Policy & Marketing, 18(1), 89-105.
Retrieved April 8, 2009, from Business Source Elite database.

Rogers, E. M. (1963). Diffusion of Innovations, 5th Edition. New York City: Free Press.

Rossiter, J. & Percy, L. (1997). Advertising Communications & Promotion Management

(2nd ed.). New York: McGraw-Hill.

59
Seiders, K., & Petty, R. (2007, September). Taming the obesity beast: Children,
marketing, and public policy considerations. Journal of Public Policy &
Marketing, 26(2), 236-242. Retrieved November 30, 2008,
doi:10.1509/jppm.26.2.236

Severin, W. & Tankard, J. (2001). Communication Theories: Origins, Methods, and Uses
in the Mass Media (5th ed.). New York: Addison Wesley Longman.

Walker, R. (2008, October). Subconscious Warm-Up. New York Times Magazine,22.

Retrieved December 1, 2008, from New York Times database. (Document
ID: 1570761901).

Walsh, B. (2008, June 28). It's Not Just Genetics. Time, pp. 70-80.

Warren, C., Strauss, J., Taska, J., & Sullivan, S. (2005, April). Inspiring or dispiriting?
The effect of diet commercials on snack food consumption in high school and
college-aged women. The International Journal Of Eating Disorders, 37(3), 266-
270. Retrieved November 30, 2008, from MEDLINE database.

Warren, R., Wicks, R., Wicks, J., Fosu, I., & Chung, D. (2008, June). Food and Beverage
Advertising on U.S. Television: A Comparison of Child-Targeted Versus General
Audience Commercials. Journal of Broadcasting & Electronic Media, 52(2), 231-
246. Retrieved November 30, 2008, doi:10.1080/08838150801992037

60
 VITA

Allen Barrow III (Trey)

Candidate for the Degree of Master of Science

Thesis: THE EFFECT OF CELEBRITY ATHLETE MODELS IN FOOD

ADVERTISING ON THE PERCEIVED HEALTHINESS OF FOOD
PRODUCTS.

Major Field: Mass Communication/Media Management

Biographical:

Education:
University of Oklahoma
Bachelor of Arts – Journalism/Advertising Major
June 2003

Oklahoma State University

Master of Science – Mass Communication/Media Management
Expected Completion – May 2009

Experience:
Hilti, Inc. – Tulsa, OK
E-Business Content Developer / Strategic Marketing (2004-Present)
Name: Allen Barrow III (Trey) Date of Degree: May, 2009

Institution: Oklahoma State University Location: Tulsa, Oklahoma

Title of Study: THE EFFECT OF CELEBRITY ATHLETE MODELS IN FOOD

ADVERTISING ON THE PERCEIVED HEALTHINESS OF FOOD
PRODUCTS

Pages in Study: 97 Candidate for the Degree of Master of Science

Major Field: Mass Communication/Media Management

Scope and Method of Study:

Recognizing the seriousness of the obesity problem, the government, consumer groups
and parents have put pressure on the food industry to adjust their advertising practices
(Seiders & Petty, 2007). The use of certain celebrity athletes in advertisements is
particularly concerning as they may be seen as “healthy” role models. Therefore,
empirical research is important to understand the potential impact that the presence of the
celebrity athlete model poses to food choices. The purpose of this study is to determine
how the use of a celebrity athlete model in food advertising affects the perception of
healthiness of the advertised food product among college students. The methodology for
this study is a simple experimental design using an experimental (treatment) ad and a
control ad. Participants were randomly assigned one of three different print ads for a
breakfast cereal. One treatment features celebrity endorser Michael Phelps, another
treatment features celebrity endorser Eli Manning, and the final treatment features a non-
celebrity male model.

Findings and Conclusions:

There were at total of 106 students sampled from Oklahoma State University journalism
classes for the questionnaire. The results of the present study indicate that college
students in the research sample were not influenced by the presence of a celebrity athlete
in food advertising. However, the respondents did rate the ad containing Eli Manning
significantly more appealing than the ad containing the control ad and Michael Phelps.
This partially confirms what is largely accepted in advertising practices, in that celebrity
endorsers do create a more favorable impression of products and brands when they are
used in the ads. The fact that Phelps was not rated higher than the control ad in the
Attitude toward the Ad scale could largely be due to his recent negative media attention
following his drug use scandal.

ADVISER’S APPROVAL: Jami Fullerton, PhD