<%BANNER%>

Sports-Celebrity Endorser

MISSING IMAGE

Material Information

Title:
Sports-Celebrity Endorser The Emotional Response
Physical Description:
1 online resource (86 p.)
Language:
english
Creator:
Starr, Walter J
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Mass Communication, Journalism and Communications
Committee Chair:
Morris, Jon D
Committee Members:
Wanta, Wayne M
Sutherland, John C
Ko, Yong Jae

Subjects

Subjects / Keywords:
advertising -- celebrity -- emotion -- response -- sports
Journalism and Communications -- Dissertations, Academic -- UF
Genre:
Mass Communication thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
In 2010, companies in the United States spent over $17 billion on sports-celebrity endorsements and sponsorships, and the expenditures globally for sports-celebrity endorsements and sponsorships was expected to exceed $46 billion. Nike was projected to pay $712 million to sports-celebrities, teams and leagues during the same year. The costs and expenditures for sports-celebrities has skyrocketed over the past ten years as practitioners and companies attempt to find the right athlete to make their ads more effective. Never has it been so vital to find the right sports-celebrity because of the investment required. If the endorser is the right fit and continues to be popular, the brand flourishes. If not, the impact to the brand could be very devastating. Therefore, practitioners and companies must find meaningful methods to make the sports-celebrity process as successful as possible. Many researchers have completed studies regarding source credibility, some of which have targeted sports celebrities. The purpose of my study was twofold. First, individuals’ affective responses to six print ads with three different spokesperson conditions were measured using the pleasure, arousal and 10 dominance (PAD) model. Second, the effectiveness of each ad by source type on purchase intent was investigated. The major finding of my research was that the sports-celebrity was more effective compared to an unknown spokesperson  in attracting attention to the ad, a vital concern for practitioners. In today’s culture, people avoid ads by cognitive, behavioral and mechanical means. So, if the athlete can bring attention to an ad, the reader, listeners or viewer may consider the arguments, resulting in a higher intent to purchase.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Walter J Starr.
Thesis:
Thesis (Ph.D.)--University of Florida, 2013.
Local:
Adviser: Morris, Jon D.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2015-08-31

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2013
System ID:
UFE0045895:00001

MISSING IMAGE

Material Information

Title:
Sports-Celebrity Endorser The Emotional Response
Physical Description:
1 online resource (86 p.)
Language:
english
Creator:
Starr, Walter J
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Mass Communication, Journalism and Communications
Committee Chair:
Morris, Jon D
Committee Members:
Wanta, Wayne M
Sutherland, John C
Ko, Yong Jae

Subjects

Subjects / Keywords:
advertising -- celebrity -- emotion -- response -- sports
Journalism and Communications -- Dissertations, Academic -- UF
Genre:
Mass Communication thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
In 2010, companies in the United States spent over $17 billion on sports-celebrity endorsements and sponsorships, and the expenditures globally for sports-celebrity endorsements and sponsorships was expected to exceed $46 billion. Nike was projected to pay $712 million to sports-celebrities, teams and leagues during the same year. The costs and expenditures for sports-celebrities has skyrocketed over the past ten years as practitioners and companies attempt to find the right athlete to make their ads more effective. Never has it been so vital to find the right sports-celebrity because of the investment required. If the endorser is the right fit and continues to be popular, the brand flourishes. If not, the impact to the brand could be very devastating. Therefore, practitioners and companies must find meaningful methods to make the sports-celebrity process as successful as possible. Many researchers have completed studies regarding source credibility, some of which have targeted sports celebrities. The purpose of my study was twofold. First, individuals’ affective responses to six print ads with three different spokesperson conditions were measured using the pleasure, arousal and 10 dominance (PAD) model. Second, the effectiveness of each ad by source type on purchase intent was investigated. The major finding of my research was that the sports-celebrity was more effective compared to an unknown spokesperson  in attracting attention to the ad, a vital concern for practitioners. In today’s culture, people avoid ads by cognitive, behavioral and mechanical means. So, if the athlete can bring attention to an ad, the reader, listeners or viewer may consider the arguments, resulting in a higher intent to purchase.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Walter J Starr.
Thesis:
Thesis (Ph.D.)--University of Florida, 2013.
Local:
Adviser: Morris, Jon D.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2015-08-31

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2013
System ID:
UFE0045895:00001


This item has the following downloads:


Full Text

PAGE 1

1 SPORTS CELEBRITY ENDORSER: THE EMOTIONAL RESPONSE By WALTER JOHN STARR A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PH ILOSOPHY UNIVERSITY OF FLORIDA 2013

PAGE 2

2 2013 Walter John Starr

PAGE 3

3 To my loving wife Doreen

PAGE 4

4 ACKNOWLEDGMENTS First and foremost, I express my tremendous gratitude to my loving wife for her guidance and encou ragement du ring the five years I labored pursuing this degree Dr. Jon Morris stimulated my interest about his research, and gave me the impetus to begin the process. He has spent countless volunteer hours with me, providing me with the knowledge and ski lls to complete my research. Dr. John Sutherland knows so much about advertising, and spent hours sharing his knowledge with me. Even with the demands on his time from his family life and departmental work, he would continue to help crunch the numbers for my statisti cal analysis of several project. Dr. Norm Lewis, who works in another department, gave me valuable advice that led to academic success that I would not have achieved without his help. Thanks to his efforts, we were able to have some research pu blished. All the people, who assisted me in the graduate office, deserved my thanks and respect. Jody, Kim and Sarah, I thank you for your assistance and guidance. Members of my committee, Dr. Ko and Dr. Wanta were very kind and helpful to me, and were gr acious to work with me within tight time constraints. I thank them for their insights and help. I also thank Jim Baer, a talented, graphic artist, for designing the six print ads used in my experiment.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 7 LIST OF FIGURES ................................ ................................ ................................ ......................... 8 ABSTRACT ................................ ................................ ................................ ................................ ..... 9 CHAPTER 1 INTR ODUCTION ................................ ................................ ................................ .................. 11 Historical Perspective ................................ ................................ ................................ ............. 11 Behavior of Sports Celebrities Unpredictable ................................ ................................ ........ 11 Need for the Study ................................ ................................ ................................ .................. 13 Chapter Contents ................................ ................................ ................................ .................... 15 2 LITERATURE REVIEW ................................ ................................ ................................ ....... 17 Persuasion ................................ ................................ ................................ ............................... 17 Elaboration Likelihood Model (ELM) ................................ ................................ ............ 18 Affective Emotional Factors ................................ ................................ ........................... 20 Cognitive Versus Affective ................................ ................................ ............................. 22 Source Credibility ................................ ................................ ................................ ................... 25 Source Attractiveness ................................ ................................ ................................ ...... 27 Celebrity Endorsers ................................ ................................ ................................ ......... 29 3 METHOD ................................ ................................ ................................ ............................... 34 Operational Definitions ................................ ................................ ................................ .......... 34 Independent Variables ................................ ................................ ................................ ..... 34 Dependent Variables ................................ ................................ ................................ ....... 34 Pleasure ................................ ................................ ................................ ............................ 34 Arousal ................................ ................................ ................................ ............................ 35 Dominance ................................ ................................ ................................ ....................... 35 The Experimental Stimuli ................................ ................................ ................................ 36 Data Collection ................................ ................................ ................................ ................ 36 Reliability and Validity ................................ ................................ ................................ ... 37 Confounding Variables ................................ ................................ ................................ .... 37 The Experimental Design ................................ ................................ ................................ 38 Sample ................................ ................................ ................................ ............................. 39 Selection of Sports Celebrity ................................ ................................ ........................... 39 Pretest Product Selection ................................ ................................ ................................ 40 Main Experiment ................................ ................................ ................................ .................... 41

PAGE 6

6 Measurement Instruments ................................ ................................ ............................... 41 Missing (Incomplete) Data ................................ ................................ .............................. 42 Statistical Analysis ................................ ................................ ................................ .......... 42 4 RESULTS ................................ ................................ ................................ ............................... 44 Description of the sample ................................ ................................ ................................ ....... 44 Hypotheses Testing ................................ ................................ ................................ ................. 44 5 CONCLUSIONS LIMITATIONS AND FUTURE RESEARCH ................................ ........ 50 Emotional Response to Sports Celebrity Endorser ................................ ................................ 50 The Importance of Arousal ................................ ................................ ................................ ..... 51 Purchase Intent ................................ ................................ ................................ ........................ 52 Future Research ................................ ................................ ................................ ...................... 55 APPENDIX: ONLINE QUESTIONNAIRE ................................ ................................ .................. 63 REFERENCE LIST ................................ ................................ ................................ ....................... 79 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ......... 86

PAGE 7

7 LIST OF TABLES Table page 4 1 Gender by experimental group and product involvement ................................ .................. 47 4 2 Average age by product involvement and celebrity groups ................................ ............... 48 4 3 PAD averages for low involvement product by spokesperson group ................................ 48 4 4 PAD averages for high involvement product by spokesperson group ............................... 48 4 5 Emotional response by product involvement and celebrity group ................................ ..... 49

PAGE 8

8 LIST OF FIGURES Figure page 3 1 Self assessment manikin (AdSAM) ................................ ................................ ................... 43 5 1 Sports celebrity low involvement ................................ ................................ ................... 57 5 2 Unknown spokesperson low involvement ................................ ................................ ...... 58 5 3 No spokesperson low involvement ................................ ................................ ................. 59 5 4 Tebow MAX ................................ ................................ ................................ ................... 60 5 5 Unknown spokesperson MAX ................................ ................................ ........................ 61 5 6 No spokesperson MAX ................................ ................................ ................................ ... 62

PAGE 9

9 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy SPORTS CELEBRITY ENDORSER: THE EMOTIONAL RESPONSE By Walter John Starr August 2013 Chair: Jon Morris Major: Mass Communication In 2010, companies in the United S tates spent over $17 billion on sports celebrity endorsements and sponsorships, and the expenditures globally for sports celebrity endorsements and sponsorships was expected to exceed $46 billion. Nike was projected to pay $712 million to sports celebrit ies, teams and leagues during the same year. The costs and expendi tures for sports celebrities have skyrocketed over the past ten years as practitioners and companies attempt to find the right athlete to make their ads more effective. Never has it been so vital to find the right sports celebrity because of the investment required. If the endorser is the right fit and continues to be popular, the brand flourishes. If not, the impact to the brand could be very devastating. T herefore, practitioners and compani es must find meaningful methods to make the sports celebrity endorsement process as successful as possible. Many researchers have completed studies regarding source credibility, some of which have targeted sports celebrities. The purpose of my study was t ads with three different spokesperson conditions were measured using the pleasure, arousal and dominance (PAD) model. Second, the effectiveness of each ad by source type on purchase intent was in vestigated.

PAGE 10

10 The major finding of my research was that the sports celebrity was more effective compared to an unknown spokesperson in attracting attention to the ad, a vital concern for vioral and mechanical means. So, if the athlete can bring attention to an ad, the reader, listeners or viewer may consider the arguments, resulting in a higher intent to purchase

PAGE 11

11 CHAPTER 1 INTRODUCTION Historical Perspective Sports celebrities have endor sed products in America for a century, dating back to the cigarette packages. In 1909, American Tobacco Company was the first company to insert small baseball cards i n their cigarette packages, one of which fetched $2.8 million in a 2007 auction (Olstad and James, 2009). Since that time, baseball players and athletes such as Pete Rose, Bruce Jenner, Mary Lou Retton, Walter Peyton and other sports celebrities had their images dis played on the front of Wheaties cereal boxes (Sirak 1998). Many remember Joe DiMaggio promoting the new, electric coffee maker, Mr. Coffee manufacturer produced 40,000 units per day (The Pla in Dealer, 2004). From 1986 to 1996, American companies increased their endorsement spending on athletes from $100 million to $1 billion. A large part of this growth is attributed to companies who historically did not use celebrity endorsers (Lane 1996). B ehavior of Sports Celebrities Unpredictable In November, 2009 Tiger Woods, a spectacular professional golfer, who endorsed many to Buicks (McKee 2008), was involved in a minor single car accident. As the story evolved, Tiger Wood faithful, family man was tarnished and his credibility damaged. As the expose of Tiger Woods unfolded, the pundits speculated on how many endorsement deals he would lose, and which companies would contin ue to use him as a spokesman. Although Woods did not win a recognized professional golf tournament for several years after his character was besmirched by his marital affairs, one company in particular continued to employ him as an endorser. Nike

PAGE 12

12 decided to keep Woods under contract even though his behavior turned away an estimated 105,000 golf ball purchasers, for a $1.3 million sales loss. Nike year endorsement deal attracted 4.5 million customers to Nike Golf from other brands (Sir ak 2011; Santoli & Barry 2009). would be a sentinel moment for marketers using sports celebrities as endorsers. In his report on the media fallout with Tiger Woods, Wall away from high profile, multi million dollar celebrity and athlete endorsements has been be fraught with ri Woods is not the first athlete whose image was seriously altered by public revelations, nor will he be the last. There have been many other prominent athletes with major endorse ment deals, whose reputations have been impugned by allegations of serious criminal acts, such as O. J. Simpson charged with a double homicide; and Kobe Bryant, arrested for rape. These sports marvels, admired by millions, can suddenly become monsters resu lting in severe damage to a brand, and chasing loyal customers to competitors (Miciak and Shanklin, 1994). Corporate America has a propensity to use athletes as endorsers, but corporations must invest time and resources to choose their endorsers carefully because their choices occasionally commit serious errors of judgment, resulting in damage to a brand ., (Kridler 2004). Since celebrities are well known, and are subject to human frailties, they attract the paparazzi, looking for that one picture worth t housands of dollars. If not the professional celebrity followers, cell phone cameras are ubiquitous, and anyone can catch a celebrity in an awkward moment. So celebrities are at high

PAGE 13

13 risk of being caught doing something damaging, resulting in a loss of the ir status, and value as an endorser. Some celebrity endorsers fail for several other reasons also, such as product celebrity mismatch, inability of the celebrity to con nect with the target audience, credibility due to an overabundan ce of product endorsements, and injury resulting in a loss of playing time an d status (Miciak & Shanklin 1994). Need for the Study Regardless of the risks associated with well known athletes and personalities, companies in the United States continue to b uy into the premise that this type of ad endorser is more effective, and able to convince consumers to buy their products as evidenced by the fact that about 25% of the commercials in this country employ celebrity endorsers (Amos, Holmes, Strutton 2008). T hese celebrity endorsements often result in a short term sales boost providing management with instant gratification (Brand Strategy 2007). It has been postulated that companies and advertising agencies believe sport and non sport celebrities are powerful enough to help their marketing communications penetrate the clutter of advertisements that confront Americans, who are overwhelmed with advertising messages, including marketing communications on radio, television, the internet, outdoor, on the floors of grocery stores, at the cinema, and just about everywhere people visit during the day (Ha & Liman 1997). It is estimated that Americans are bombarded with at least 1,000 marketing communication attempts each day (Kohler 1997). All of these messages are atte mpting to influence consumer behavior through different forms of persuasion, so they will adopt a certain product, service or idea (Myers Levy & Malaviya 1999). These endorsing superstars are able to grab and hold the attention of viewers, listeners and re aders of the mass media. Advertisers believe their association with a well known and liked

PAGE 14

14 sports celebrity will elevate their company and its products to a higher status by transfering the ath Shanklin 1994). Because of the r ising cost of sport celebrity endorsements, and because only one in five commercials using sports celebrities performs to expectation (Miciak, et al. 1994), the question to hire sports priced, risky athletes really perform, by making ads brand by developing more innovative and creative marketing comm unications? Another alternative might be to discover ways to create a more appealing image for the brand or to change the position of the brand by making its products more affordable. There are indeed various options to increase brand equity without the ex pense or risk of a celebrity endorser. Research has demonstrated that affect is an important aspect of the consumer experience including the reactio n to an ad (Zajonc 1980, Edell & Burke 1989). Logically, then, for an ad to be effective with consumers, it must use an affective component. Since we are examining the source in this study, it would also follow that for sports celebrities to be effective spokespersons, they must evoke an emotional response from consumers. If advertising agencies and corporation s are going to continue to use sports cost, to establish a means of ensuring that these athletes connect with their audience emotionally? The purpose of current study is twofold. First individual ads were measured using the pleasure, arousal, and dominance (PAD) model. The six print ads were divided into three groups of two ads with each group having a different spokesperson condition (sports celebrity, unknown s pokesperson and no spokesperson). In each group, one ad

PAGE 15

15 included a low involvement fictitious product while the other contained a high involvement, fictitious product. The two products were the same in each group, so the manipulation was restricted to th e message source. The headline for each advertisement, and the image of the product were similar in all three conditions. Secondly, the effectiveness of each ad condition on purchase intent was investigated, and other correlations were examined. Therefore following each ad condition, several product questions were asked, and a single item Intent to Purchase Scale was employed Results of this study will give academics and practitioners additional insight into the use of sports celebrities as endorsers for brands, products and services. Since extraneous variables were eliminated or greatly reduced, the study focused on the ability of a sports celebrity to emotional resp onse scale and intent to buy scale, the marketer will be able to determine if a Chapter Contents Chapter 1 revealed that the use of celebrity endorsers has been a long established practice in the Unite d States, resulting in countless brand success stories, of which some memorable ones are mentioned. Whi le the successful examples do not always make headlines, the failures almost always do. The fairly recent case of Tiger Woods comes to the forefront beca use many Americans were shocked at the revelations of his infidelity. To complete the chapter, the need and purpose of the study are discussed as well as the contribution anticipated to benefit the advertising community of academics and practitioners. Chap ter 2 examines the evolution of the persuasion theories from the early work of work was the basis for many researchers that followed, postulating various hypothese s regarding

PAGE 16

16 persuasion, culminating with the Elaboration Likelihood Model ( ELM) of P etty and Cacioppo (1986). Much of the work that followed was thought to be an improvement of the ELM including that of Morris, Woo & Singhl (2005 ), who added an affective perspective to the theory. Since this study focused on the message source, there is a lengthy review of the various source credibility models that emphasize attractiveness, trustworthiness, and other attributes. The chapter concludes with the hypotheses f or my study. Chapter 3 pres ents the method and procedures utilized to investigate the hypotheses. The study is a post test, control group only design employing the manipulation of the source independent variable in three conditions. AdSAM, a non verbal sc ale will be employed to measure the affective response to the ad conditions. The analysis will be performed using an one way and two way ANOVA.

PAGE 17

17 CHAPTER 2 LITERATURE REVIEW Persuasion Persuasion has been the focus of numerous studies conducted by socia l scientists from various fields including advertising, anthropology, communications, psychology, and sociology. Academics and practitioners are fascinated by how people are influenced through communication, and want to learn what factors are necessary to create persuasive messages (Perloff 1993). To clarify the meaning of persuasion, a plausible definition was proposed about 45 years attitude s, beliefs, or behavior of another individual or group of individuals through the & Cody, 1987). No definition is perfect, but this one contains several important concepts including motive, goal, method and obje ct. In the 20 th Century persuasive messages were used by the government to convince the American people to support World Wars I and II. Many social scientists were summoned to Washington to help the government create propaganda. Included in this group were Carl Hovland and Harold Lasswell. These men, along with their colleagues, experienced first hand the effects resulting from persuasion campaigns (propaganda) (Hovland, Janis & Kelley, 1953; Smith, Lasswell, & Casey, 1946). According to Markovi (2008) Smith et al. (1946) developed four successive phases of any communication by posing four

PAGE 18

18 munication take place? Who communicates? What is continued his research at Yale University, and with his colleagues posited a process for attitude change. During the e arly 1950s, many scientists were developing learning theories, so Hovland was influenced by this environment. His systematic process for attitude change included a learning step. The persuasion process involved several stages; a person had to hear or see t he message, understand it, learn it, accept i t, and absorb it This conceptualization was characterized as learning and belief based persuasion theory, and was supported by research (Hovland, Ja nis, and Kelly 1953; Fishbein & Azjen 1975). Elaboration Likel ihood Model (ELM) The underlying theory that has driven persuasion research is the Elaboration Likelihood Model ( ELM) postulated by Petty and Cacioppo (1986). The theory posits that there are two, explicit routes to persuasion: the central route and the pe ripheral route. When people are mentally capable and motivated (the message is personally relevant), a message is given further elaboration, resulting in a closer examination of the arguments. The better the quality of the arguments, the higher the likelih arguments, being well informed, a forewarning of an attempt to persuade, and the lack of distracti on. When less elaboration is needed or desired or the person is incapable of elaboration, a number of cues can facilitate the processing of a message, resulting in a change of attitude. This route is identified as the peripheral route, since the cues or s hortcuts are peripheral to the

PAGE 19

19 mood, rhetorical symbols, quantity versus quality of the expertise on a particular issue. Attitudes are altered by these cues, and do not require any cognitive activation by the receiver. According to the ELM, attitude changes resulting from peripheral cues are considered weaker than those changed by the central process. However, practitioners have used these peripheral cues as an easier way to persuade people to change an attitude towards an advertised object (Perloff, 2008). Other practitioners use a celebrity endorser because the creative pressure to create a great advertising theme is dramatically reduced since the message source becomes the theme (Till, 1998) An attempt to change attitudes is usually accomplished by a persuasive message. Therefore, for a message t o be persuasive, it must changed an opinion about an issue, object or person If it is a persuasive message, there must have been a successful attem pt to influence difficult to claim the attitude change was the result of a message. In essence, the ELM provides two alternative hypotheses to an attitude change, meaning that if it is challenged, it is not possible to falsify either hypothesis (Cook, Moore & Steel 2004). Unlike the central route, resulting in significant exertion of cognitive effort, the heuristic view of persuasion requires very little effort in p rocessing a message. Instead of weighing the strength of arguments, the recipient depends on more easily attainable information including the identity of the source or other non content cues to accept or reject the opinion of the message. So the heuristic approach does not emphasize detailed message processing, but rather focuses on the content cues such as the credibility of the source (Chaiken 1980).

PAGE 20

20 Myers Levy and Malaviya (1999) proposed a third processing strategy espoused by Chaiken, Liberman, and Eagley (1989) and Petty and Cacioppo (1986). Therefore, more than two routes my be used to process messages, significantly altering conventional beliefs about persuasion (Perloff 1993). About the same time Petty and Cacioppo were advancing their model, other researchers psychologists called the information processing theory. Therefore, advertising was believed to be an information source for consumers, assisting them in their purchasin g decisions. However, this conviction came into question as evidence mounted that consumer decisions may be based as much on affective emotional factors as on rational factual inputs (Holbrook et al. 1984). Affective Emotional Factors In the early 1980s, p sychologists regarded feelings last, meaning that emotions were evoked as a response to a stimulus only after considerable cognition. Therefore, appealing or engaging could not be determined until people had adequately thought about the object in question, referring to knowledge of it and deciding whether they liked it. After thorough cognition, an affective response to an object could be elicited. Nothing happened until after cognition (post cognitive), making affect something that comes last. In the inde xes of significant works on cognition at the time, words such as affect, attitude, emotion, feeling and sentiment did not appear. However, using examples of everyday conversation, where humans exchange their opinions, preferences and evaluations, Zajonc (1 980) argued that affect in these intercourses is transmitted by verbal communications, and also by non verbal cues, which actually may communicate significant components of affect. Further, Zajonc believed very few perceptions

PAGE 21

21 lack an affective component, etc. To strengthen the argument Zajonc (1980) also utilized the readings from one of founding father of psychology, Wundt (1907 pp. 243 244) When any physical process rises above the thres hold of consciousness, it is the affective elements which as soon as they are strong enough, first become noticeable. They begin to force themselves energetically into the fixation point of consciousness before anything is perceived of the ideational eleme nts. They are sometimes states of pleasure or unpleasureable character they are predominantly states of strained expectations often there is vividly present he special affective tone of a forgotten idea, although the idea itself still remains in the backgr ound of consciousness. In a similar manner the clear appreciation of ideas by acts of cognition and recognition is always preceded by feelings. to be at least two d ifferent forms of unconscious processes. The first is the behavior that occurs due to the influence of affective factors without any cognitive action. These instances include scrimination, when the response is overlearned, resulting in an automatic processing of information. This involves a cognitive process, but the data are collapsed i nto large molar chunks that could conceal how the original processing was achieved (Zajonc 1980). Zajonc (1984) reaffirmed his belief that affect was independent of cognition, and refuted the arguments postulated by the opposing school of thought led by L azarus. Lazarus believed cognition must occur before emotion, because for humans to experience emotion, they must first comprehend and evaluate a perception. So, the main argument presented by Lazarus focused on t arise out of the blue, they are an integral part of a

PAGE 22

22 Cognitive Versus Affective Several researchers (Sojka & Gies e, 1997; Edwards, 1990; Chaiken, 1978, 1980; Chaiken & Maheswaran, 1994; Woo & Singh, 2005; Maio & Haddock, 2007; Haddock, Maio, Arnold, & Huskinson, 2008) have questioned whether persuasion is cognitive or affective thinking or feeling. Sojka and Giese (1997) conceptualized the relationship betwe en cognition and emotion as a biaxial grouping of four quadrants. In the lower right quadrant are high cognitive and low affective processing, while on the opposite side of the continuum in the upper left quadrant are right quadrant are cognitive and affective processing located in the lower left quadrant are low cognitive and affective processing, resulting in a way of evaluating a stimulus that is unclear to th e researchers. Another approach to the interplay of cognitive and affective responses to an object, issue or person was posited by Edwards (1990): associated with particu lar motivational pressures. For instance, the cognitive component may be dominant for attitudes acquired in service of reality testing or of a need to explain the external world. On the other hand, affective factors may predominate for attitudes arising in response to need gratification or deprivation, threats to the self image, or unconscious motives In essence, the way an attitude is acquired determines the type of processing an individual will be susceptible to, for an attitude change. Therefore, if an attitude was formed by an affective perception of a stimulus, then it can be more easily changed with an affective counter argument. Criticisms of the ELM have been based on the specific way people processed messages and the fact that the processing was mo derated by other variables such as attractive source, the mo od of the receiver, and others not related to this study. According to several researchers

PAGE 23

23 (Bohner & Apostolidpou, 1994; Tiedens & Linton, 2001; Kuykendall & Keating, 1990; Bless, Bohner, Schwartz &Strack, 1990), the mood of the receiver influences how a message is processed. If the message target is in a positive mood, cognitive processing is minimized. On the review the arguments. The implication is that if the arguments are meaningful to the receiver, the impact negative, persuasion may be impacted, causing rejection of the persuasion attempt. Regardless of the antecedent determinants, there was agreement that if people used the central route of ELM to process a message, the processing would be cognitive with a minimal affective component. The idea that the central rout e and the peripheral route were separate, one being cognitive while the other is affective, was challenged by Chaiken and Maheswaran (1994), who described the central route as systematic and the peripheral route as heuristic. So the systematic route for a ttitude change requires careful examination of the message arguments cognitively by a person capable and motivated to spend the time needed for a lengthy process. On the other hand, others, neither capable nor motivated, will not invest time in a cognitive review, but rely on heuristic cues, such as attractive source or length of argument. These simple heuristic cues are most likely learned over time through observations and experience, and are divided into three categories: source, context, and message cue s. People generally believe statements made by expert sources are more likely to be true than statements made by non experts. A consensus cue occurs when it is known many people like or endorse an object, brand, or position, resulting in the object, etc being considered more worthy. In advertising are used to activate this cue, planting the ideas in the minds of consumers that many people like

PAGE 24

24 the product. These peripheral cues can be so powerful that they alone can positively bias cognitive review of the message arguments, if the arguments are ambiguous (Chaiken et al. 1989) Further it was note d by Chaiken and colleagues: almost always consist of clear, unambigu ously strong or weak arguments. The use of such stimuli should reduce the likelihood that heuristic processing will bias systematic processing, and increase the likelihood of attenuation or additivity, depending on the amount and nature of the information that systematic processing yields. Not satisfied with prior research by Mackie (1987), who demonstrated that attitudes can be altered by the consensus cue, even when cognitive processing had occurred, Chaiken conducted research to test the bias hypotheses By manipulating motivational levels and message strength, analysis sh ould generalize to other heuristic cues. Thus, if we replicate the present experiment using a different heuristic cue (e.g., consensus information or message length) we would expect As suggest ed by previous research, the emotional impact of persuasive processing was relegated to the peripheral route, since most researchers agreed that the central route required mental capability and motivation to process the advertisement or message cognitively However, a study revealed that sample adolescent students unable to determine the exact claim of a print message. The teens were unable to discern the difference between the visual imagery used to get their attention and imagery that delivers the relevant brand data (Edens & McCormick, 2000). Morris, et al (2005) interpreted the previous study as support for the concept that the peripheral route becomes the centr al route, adding credence to their belief that it is logical that emotions play a role in cognitive processing. Previous research data support this claim, and so

PAGE 25

25 does physiological evidence established by neuroscientific studies, showing that areas of the brain thought to be the site of cognitive thought are also an area of emotional processing. Areas of the brain identified as responsible for processing different emotions included the amygdala, ventromedial prefrontal, brain stem nuclei, hypothalamus, and basal forebrain, all located in the subcortical region of the brain (meaning below the cerebral cortex), an area of the brain of essential importance to cognitive neuroscientists. Morris et al. (2005) developed two research questions and two hypotheses in an attempt to demonstrate that the emotional aspect in processing ads is as important as the cognitive aspect. pheral route. The personal relevance of the arguments presented. They have the mental ability to process the information, resulting in the belief that this method re quires strong cognitive skills and little or no emotional response. On the emotional cues to process an advertisement. The results that the emotional aspect is m ore involved in the direct, central route of the ELM than measured in the peripheral processing. This finding contradicts what was believed about the dual route processing model. Data did support the ELM regarding purchase intention, even though the purchase intent was stronger via the central route, there was no evidence that these people would in fact purchase the product. Source Credibility A credible source c an produce an attitude change through a psychological process described by Kelman (1961) as internalization. This process causes a person to accept the new

PAGE 26

26 the so urce of the attitude change is long forgotten, the person will retain the changed attitude. attempting to persuade them and in response, consumers have developed met hods to cope with such attempts (Friestad & Wright, 1999). definition postulate marketing academicians, and practitioners have studied source credibility extensiv ely, and the demonstrate no effects in altering the beliefs, attitudes or behaviors of the receivers of a ers have examined this question by identifying the various dimensions of source credibility, and they have agreed on one or more of the dimensions of the construct (Hovland, Janis & Kelley, 1953; Cialdini, 1984; Patzer, 1985; Pornpitakpan, 2004; Clow, Jam es, Sisk & Cole, 2011) However, in some cases, definitions vary for the same named dimension. For instance, Chaiken (1979) describes attractiveness as physical attractiveness, whereas the research by Feingold (1992) supports the concept that people assoc iate physical attractiveness with qualities such as being sociable, dominant, sexually warm, mentally healthy, intelligent, and socially skilled. Regardless of the definition, the attractiveness of the source has received considerable attention from resea rchers of multiple disciplines. The obvious definition of the attractiveness of a source is the physical attractiveness of the communicator, and it is exemplified by the

PAGE 27

27 operationalization of the variable in previous research. For instance, Horai, Naccari, and two other studies the picture of an attractive, smiling student was used as the attractive source (Mills & Harvey, 1972; Norman, 1976). Source A ttractive ness Source attractiveness has historically been thought to comprise of several facets, including familiarity, similarity, and liking as operationalized by Mills and Harvey (1972) and Norman (1976) for use in their research, where the sample was composed of students, and the attractive source was an attractive student. Therefore, it appears that attractiveness may have been conceptualized as an amalgam of the three aforementioned factors plus physical attractiveness. If the concept of attractiveness is mu ltifaceted, then it will be difficult to determine what facet rendered the relevant changes to the dependent variables. Physical attraction is readily apparent, making it a more important component (especially in mass media) than the less recognizable fact ors of familiarity, similarity and liking (Maddux & Rogers, 1980). Several studies provide strong support for the belief that attractive sources influence an attitude change. Mills and Harvey (1972) study subjects were asked to read a message credited to either an expert or an attractive source. Results indicated more agreement with the expert when the information about him was given before versus after the communication. In the case of the attractive source, agreement was not affected by when he was ident ified, leading to the conclusion that supportive arguments are more vital with an expert witness than an attractive one. Therefore, an attractive source can influence opinion simply by association with a position, without the need for strong supportive arg uments. This finding agrees with an earlier study by Kelman (1961). Further support for this postulate was provided by research completed by Norman (1975).

PAGE 28

28 In a study critical of the results posited by Mills and Harvey (1972) and supported by Norman (197 Maddux and Rogers (1980) argued that instead of conceptualizing the variables as expertise versus attractiveness, a more thorough test of the hypotheses can be completed by conce ptualizing the variables as mutually exclusive. Therefore, the independent variables were presence or absence of supporting documents, levels of attractiveness, and degree of expertise; while the dependent variable was agreement with the communication. The attractive source in effects for the attractive source may have bee n related to the extreme levels of attraction that were manipulated. The attractive source was either very attractive or very ugly, leading one and Rogers 1980). Summarizing the role of recall and the acceptance of nor physical attractiveness affected recall and that recall was not significantly correlat ed with In a study regarding attractiveness of the communicator and opinion change, Snyder and Rotbarti (1971) found that an attractive source was associated with more opinion change than an unattractive source for se veral reasons, including the following: an attractive face is distracting and impedes the elaboration of counterarguments, and attractive sources are better liked than unattractive sources, leading to a higher degree of message acceptance. Snyder and Rotb arti also investigated whether an attractive source is perceived as more credible than an unattractive source, finding that source credibility and attractiveness were mutually exclusive.

PAGE 29

29 Celebrity Endorsers For most people, not a day goes by without hea ring or seeing something in the mass media about some American or foreign celebrity, showing how significant these famous people are to pop culture. One of the most cited definitions of celebrity was provided by McCracken defined as any individual who enjoys public recognition and who uses this recognition on behalf of a consumer good by appearing with it in an from movie and TV sta rs to athletes, politicians, business personalities, artists, and military luminaries. endorser is an individual who is known to the public (actor, sports figure, entert ainer, etc.) for his dorsers are likeable and attractive, while the expert has expertise; and typical consumers are used because they enjoy similarity with the audience (Friedman & Friedman, 1979). Several models, hypotheses and theories have been investigated and postulated to explain the effectiveness of celebrity endorsed ads. Generally, researchers agree celebrities, because of their attractiveness and likeable qualities, are able to break through the ad clutter, provide more attention to the endorsed brands, and facilitat e attitude change (Horai et al, 1974; Mills & Harvey, 1972; Chaiken, 1979; Atkin & Block, 1983; Kahle & Homer, 1985; Sherman, 1985; Ohanian, 1991; Miciak & Shanklin, 1994; Erdogan, Baker & Tagg, 2001; Chao, 2005). Models highly researched for celebrit y effectiveness are described by McCracken (1989) as the source credibility model and the source attractiveness model. These models were given

PAGE 30

30 special attention in the previous section becaus e they apply to all sources not just celebrities. As noted earli er, the source credibility model has its foundation in the early work of Hovland, et al. (1951, 1952, 1953) who postulated that the effectiveness of the message depends on the expertness and trustworthiness of the source (McCracken, 1989). Research support s the belief that celebrities are more credible than non celebrities, and they bring their credibility to the products and brands they endorse (Choi & Rifo n, 2007; Atkin & Block, 1983; Oh anian, 1991) The source attractiveness model was developed by severa l researchers including McGuire (1985), who believed that attractiveness encompassed more than just physical attractiveness: including the attributes of familiarity, likability and similarity. Some of the research previously discussed regarding this model used a physically attractive source (Mills & Harvey, 1972; Norman, 1976). Citing earlier studies (Friedman & Friedman, 1979; Kamen, Azhari & Kragh 1975 ), McCracken (1989) said the source models could not explain some of the earlier Fo average consumer) and found that each type of endorser was better able to change opinion for certain types of products. If the source model was correct, the celebrit y endorser should have been successful with all of the products tested. While studying the effects of Johnny Cash as a kind of core around which the substantive campaign using Cash as spokesperson for Amoco Oil negative replies were received from the public about Cash. Kamen, and colleagues believed that, over time, these negative feelings would diminish, as th e campaign continued.

PAGE 31

31 James Garner was a powerful and effective endorser of several products, including Mazda and Polaroid In the Polaroid ads, the on air chemistry between Garner and Marietta Hartley was so powerful that many in the audience believed they were married. According to According to McCracken, Gar Garner i cannot tell us why (McCracken, 1989). These observations led McCracken to believe that the effectiveness of a celebrity source depends partly on the meanings the celebrity brings to the endorsement process, which are then transferred to the brand and ultimately to the consumer, completing the Meaning Transfer Model or Cultural Meaning Transfer (Till & Busler, 2000). and it was first identified by Kahle and Homer (1985), accordi ng to Charbonneau and Garland (2010). Kahle and Homer (1985) discussed several alternative reasons for their findings, attempting to explain why the attractive celebrity was more effective than the other endorsers: None of these processes, however, provi des a compelling explanation for the results of this study, where the sensual aspects were downplayed and the attractiveness manipulation altered recall of arguments and brand. A more probable explanation invokes the matchup hypothesis and the SA (Social A daptation) theory that physical attractiveness is a source of information. If a stunningly attractive person claims to use a beauty product, that product may be assumed to be an

PAGE 32

32 element of the beauty formula. The attractiveness information is conveyed more quickly than other information, even if it is not highly probative. Additional research (Kamins, 1990; Ohanian,1990; Miciak & Shannklin, 1994; Till & Busler, 2000; Charbonneau & Garland, 2005; Biswas, Biswas & Das, 2006) has strengthened the theory by pe rforming experiments that filled gaps in the original work by Kahle and Homer. For an attractive model endorses a product not used to enhance beauty. Finally, Lynch and Schuler (1994) substantiated A celebrity endorser has a certain image that is perceived by the consumers, and this process by endorsed brand follows the hierarchy of effects model; which in its most basic form, demonstrates how consumers grow their knowledge of the brand from a minima l state to a higher level, resulting in ultimately choosing the brand. No matter how the consumer arrives at the purchase decision, occurred between the celebri ty and the brand (Charbonneau & Garland, 2010). In another study, brand, a transfer of affect takes place. But when the spokesperson is incongruent, or irrelevant, the that congruence leads to significantly more positive brand affect than does incongruence.

PAGE 33

33 Based on the extant literature on the effectiveness of celebrity endors ers, their ability to choosing the brand. The following hypotheses are submitted: H1a : A known sports celebrity endorsing a low involvement, fictitious prod uct will be significantly more pleasing (more positive valance) as compared to an endorsement by an unknown spokesperson or no spokesperson at all. H1b : A known sports celebrity will be significantly more exciting endorsing a low involvement, fictitious pr oduct compared to a non sports celebrity or no spokesperson at all. H1c : A known sports celebrity will be significantly more controlling endorsing a low involvement, fictitious product compared to an unknown spokesperson or no spokesperson. H2a : A known sp orts celebrity endorsing a high involvement, fictitious product will be significantly more pleasing (more positive valance) as compared to an endorsement by an unknown spokesperson or no spokesperson at all. H2b : A known sports celebrity will be significan tly more exciting endorsing a high involvement, fictitious product as compared to an unknown spokesperson or no spokesperson at all. H2c : A known sports celebrity will be significantly more controlling endorsing a high involvement, fictitious product com pared to an unknown spokesperson or no spokesperson. H3a : A known sports celebrity endorsing a low involvement, fictitious product will be significantly more pleasing (more positive valance) compared to the same known sports celebrity endorsing a high invo lvement, fictitious product. H3b : A known sports celebrity will be significantly more exciting endorsing a low involvement, fictitious product compared to the same known sports celebrity endorsing a high involvement, fictitious product. H3c : A known sports celebrity will be significantly more controlling endorsing a low involvement, fictitious product compared to the same known sports celebrity endorsing a high involvement, fictitious product. H4a: Purchase intent will be significantly greater for a low in volvement, fictitious product endorsed by a sports celebrity than a low involvement, fictitious product endorsed by an unknown spokesperson or the product without a spokesperson. H4b : Purchase intent ion will be significantly greater for a high involvement, fictitious product endorsed by a sports celebrity than a high involvement, fictitious product endorsed by an unknown spokesperson or the product without a spokesperson.

PAGE 34

34 CHAPTER 3 METHOD Operational Definitions The following variables were used in my st udy. The independent variable is the source of the advertising message, and the three types are enumerated. The dependent variables are discussed, including the three dimensions of emotion response. Independent Variables The message source was manipulate d in six print advertisements. In the first condition (in two of the ads), the message source was a well known sports celebrity, chosen in a pre test survey, endorsing two fictitious products. The second condition features two similar ads with an unknown spokesperson while the third condition has no spokesperson, just the same two fictitious products selected in another pre test survey. Dependent Variables The dependent variables for my study identifying a s et of common dimensions that can be used to distinguish specific emotions from pleasure, arousal and dominance or (PAD), originally identified by Mehrabian and Russ ell (1974) to be reliable and sufficiently able to define all emotional states. For my study, these dimensions were measured using AdSAM (Ad, self assessing manikin) a non verbal tool that assesses the three dimensions of PAD using five graphic displays o f each dimension situated over a continuous, nine point scale (Morris, Woo & Singh 2005). Pleasure As with the other two dimensions, pleasure is a bipolar measure ranging from extreme happiness to extreme unhappiness (Morris et al 2002). According to Mehr abian and Russell

PAGE 35

35 (1974) the dimensions actually exist along a continuum, and in the case of pleasure, the extremes of this line travel from extreme pain or unhappiness to extreme happiness or ecstasy. Mehrabian and Russell (1974) also believe emotion is n ot a sometime, passionate feeling, but that a person face of the AdSAM manikins (Figure 3 1). Arousal Arousal was originally shown on the continuous nine point SAM scale as a range from a manikin with a sleepy face and eyes closed to a manikin with eyes open and excited (Morris 1995). Morris compared SAM scores for pleasure and arousal for each of the 135 adjectives reported from the verbal analysis of Mehrabian and Russell (1974) and found correlations of .93 for pleasure, .93 for arousal, and .66 for dominance. Earlier studies by Mintz and Mills (1971) found that the higher the arousal, the stronger the attitude toward an object. Dominance The two previous dimensions are relatively easy to understand. Pleasure is conceptualized as feeling happy or unhappy, Feelings associated with arousal (such as curious, surpr ised and wondering) can be easily differentiated from words including relaxed, placid, or indifferent. The main characteristic of dominance is power: if people feel powerful, they are experiencing personal dominance; and if they feel week, they are highly submissive (Stamps 2005). Mehrabian & Russell (1974) constructed a PAD scale, using six adjective pairs for each dimension. For dominance, three adjective pairs included controlling controlled, influential influenced, and important awed. A factor analysis for all 18 items revealed three nearly independent dimensions (Mehrabian 1995).

PAGE 36

36 In the SAM scale, dominance is represented by five manikins, whose size increases from very small to quite large. Obviously, the small manikin correlates to a feeling of bein g controlled, and the large manikin indicates that the individual feels in control or controlling. The Experimental Stimuli The stimuli for my study were six print ads created by a professional, graphic artist, who specializes in creating print ads for ma ny commercial businesses and events. All of the ads for each of the two fictitious brands were similar in s ize, color, headline, and so on except for the source of the message and the facial expression of the sports celebrity and the unknown spokesperson In the first two advertisements, the sports celebrity spokesperson appeared, and made an explicit endorsement of the product through the use of quotation marks on the headline copy. In previous studies, researchers showed the endorser and product together in the same ad, implying that the product was endorsed by the celebrity (Atkin & Block 1983, Choi & Rifon 2007). Others such as Miller and Laczniak (2011) defined the endorsement process as celebrities lending their names or images to products. In the sec ond two ads, the spokesperson was an attractive person of the same gender as the sports celebrity. No spokesperson was included in the last two ads, so the viewer of the ad would focus on the products themselves. Data Collection An online questionnaire was the instrument used to collect data resulting from measuring the dependent variables for each manipulation of the printed advertisement. The questionnaire was designed by using Qualtrics Online Survey Software, allowing for randomization of the three groups of two ads for a total of six different ads. In each ad block (group) of questions, the AdSAM continuous, nine point scale was positioned below the SAM graphic displays, for each of the PAD dimensions (Figure 3 1).

PAGE 37

37 In addition to the AdSAM scale, t he questionnaire contained two questions regarding the to buy scale was also included in each of the ad blocks, allowing the researcher to determine the effectiveness of the sports celebrity en dorser compared to an unknown endorser or no spokesperson at all. Reliability and Validity The source credibility scale developed by Ohanian (1990) was used in the pretest portion of the study because it measures the tri component construct of source credi study provided support that the scale was reliable and valid. Many other source credibility scales (Applbaum and Anatolm1972; Berlo, Lement and Myers 1969; Bowers and Phillips 1967; DeSarbo and Harshman 1985; Simpson and Kahler 80 81; Whi tehead 1968; Wynn 1987) according to Ohanian have been used extensively, but not found to be reliable and valid. The reliability and validity of the SAM scale to measure emotional response has been demonstrated in numerous studies. One study found correla tions between scores obtained by correlations were vigorous for pleasure (.94) and arousal (.94), and meaningful for dominance (.66) (Lang 1985). Similar results wer e reported by Morris and Bradley (1994). Confounding Variables The researcher using the experiment method must identify and minimize the impact of confounding variables, so any changes in the dependent variables may be attributed to the independent vari ables (Hart 2007). Six different ads, two per spokesperson type, were randomly shown to survey respondents, creating three different groups. Since each group saw two of the six ads, there should not be an issue with ad wear out. Even though the products ( Power Plus and MAX) are fictitious, there may be certain emotional responses to a particular product line. In this case, the brand is fictitious, but

PAGE 38

38 respondents of the pretest conceptualized Power Plus as a sports drink. This may have occurred because the fictitious brand name was positioned after the brand names of other sports drinks, or it may have been because of the name itself. In the online survey, questions after the emotional response scale for each ad asked respondents if they were familiar with the product or ad. If these questions elicited a positive answer, their responses were eliminated from the analysis. Even though it was impossible for these respondents to have seen these ads or be familiar with these fictitious products, there is concer n that the mere fact that they thought they saw the ads or knew the brand could have an effect on their responses (Bornstein 1989; Hansen and Wanke 2009). My study was not intended to be a repeated measure design (Toutenburg and Shalabh 2009). Therefore, t hese responses could not be included because I needed to maintain the credibility of the experimental design chosen for this study. The Experimental Design My study is a post test, control group only experimental design using the AdSam, non verbal scale, and an Intent to purchase scale. (PAD) model, to six print advertisements. Two ads featured a known sports celebrity endorser with a fictitious product that was deter mined to be congruent; another featuredthe same false brand with an unknown spokesperson, and the third condition will have the same fictitious brand by itself in the advertisement. The ad was similar in all three conditions except for the manipulation of spokesperson. My study examined the effects of the various conditions as measured by AdSAM. The known sports celebrity endorser was chosen by a p re test survey. R espondents were asked to list their favorite celebrities including any sport celebrities.

PAGE 39

39 The effectiveness of each ad condition on purchase intent was investigated as well as other possible correlations. O ne of the main goals was to examine the emotional response to the the unknown product that could be perceived as congruent or incongruent with the endorser. By designing an experiment be confident that the variable being manipulated is causing the change seen within the dependent Sample The subjects used as respondents in the pretests (sports celebrity and product selections) and the main experiment (emotional response to ads) were college students, who attended spring and summer 2012 and spring 2013 classes in various disciplines at a major southeastern university. The subjects were volunteers, and some receive d extra class credit for completing the survey The college students who completed the pr etests and main experiment surveys were similar in age and gender. Selection of Sports Celebrity To identify a sports celebrity to be included in this study, participants were asked to compile a list of celebrities, at the beginning of class. No time lim it was imposed, so the students had sufficient time to write down as many celebrities as they could remember. The instructor told the students they should list anyone, they considered a celebrity, including pe0ople from pop cultural, performing arts, spor included many recording artists, even an American Idol contestant. Tim Tebow was the top choice in the sports celebrity category; he received the most votes, outdistancing his closest rival by a seven to one margin. Others top choices included Derek Jetter (New York Yankees),

PAGE 40

40 Dwayne Wade (Miami Heat), LeBron James (Miami Heat) and Tiger Woods (professional golfer). Pretest Product Selection Two products were selected for the experiment usi ng on an online questionnaire. Respondents were asked to rate the selected sports celebrity (Tim Tebow) as an endorser for point semantic differential, source credibility scale (Ohanian 1990). Demog raphic information was also collected. Three fictitious products tested included Power Plus, a sports drink (low involvement); MAX, a smart phone (high involvement), and Toba sportswear (low involvement). The Ohanian scale includes five items for each of the three source credibility dimensions (attractiveness, trustworthiness, and expertise). In later research, Ohanian (1991) concluded that the most influential dimension of source credibility directly correlating to intent to purchase is expertise. Therefo have direct connections with their endorsed products, and who are perceived to be experts by The five item scale for each dimensi on was transformed into a single variable for each dimension (attractiveness, trustworthiness, and exper tise). A comparison of the mean scores for the three products revealed no significant differences in the three dimensions for source credibility between Power Plus and MAX. Toba sportswear was significantly higher in source credibility. Therefore, the source credibility of Tebow (sports celebrity) was essentially the same for Power Plus and MAX. Therefore, these products were chosen for the experiment, a llowing the researcher to focus on the difference between a low and high involvement product since Power Plus, a sports drink, was determined to be low involvement, and MAX, a smart phone is considered to be high involvement.

PAGE 41

41 Main Experiment The main ex periment used an online questionnaire designed using Qualtrics Online Survey Software. The ideal location for an experiment would be in a facility such as a Research Lab, where the environment could be better controlled, and distractions kept to a minimum. However, Qualtrics software facilitates completion of the instrument by many respondents because they can participate at a time and place convenient for them, increasing the response rate, and allowing for a larger sample. The online questionnaire was de signed to use a post test control group only design with three source conditions: two ads with a known sports celebrity; two ads with an unknown spokesperson; and two ads with the products only (Questionnaire and ads in Appendix). The blocks of questions f or each condition were randomized by the Qualtrics software in order to eliminate any ordering effect, and to assure a minimum of 30 respondents per group, the minimum required. Measurement Instruments Affective response toward Source Type was measured u sing a non verbal measure for emotional response; named AdSAM, for Attitude self assessing manikin (Mehabrian and Russel 1977; Morris et al, 200 3 2005). AdSAM with a graphic character arrayed along a continuous 9 1). The graphic character of AdSAM helps overcome burdensome and time consuming issues with verbal self report scales. Also the graphic characters are simple to understand, preventing respondent wear out, and are not restricte d by age, gender, culture or language differences. AdSAM is (Morris et al, 200 3 2005)

PAGE 42

42 Missing (Incomplete) Data The questionnaire was designed to force answ ers on the AdSAM scale, but answers were not forced on the follow up questions regarding familiarity with the ad or product. Since the familiarity questions did not force an answer, it was possible some of the respondents did not answer all the questions. In an analysis of 181 social and behavioral science research studies completed in 1991, missing data, which was detected by explicit discussion of the issue in the article, by reports of varying sample sizes, and by changes in degrees of freedom. Of the 69 studies with missing data, 47 researchers chose to use all the data available, while 2 0 used only completed data. Two of the studies used different methods to estimate the missing data. Overall, Bodner (2006) was concerned with the lack of reporting offered by researchers with missing data, and was concerned that authors did not discuss the uncertainties present when incomplete data are analyzed. So by deleting the incomplete data, my study avoids generating uncertainty regarding analysis of data, where data are missing. Therefore, responses missing data were eliminated. Statistical Analysi s A one way ANOVA and a two way ANOVA were used to analyze results. Emotional responses (PAD) to each condition were considered a group. The means of the groups were compared for significance. Where the ANOVA was significant, a comparison of means by con fidence intervals was used to determine which groups were significantly different. Finally, the intent to purchase respo nse was used to determine if the ads were effective in producing a desire to purchase.

PAGE 43

43 Figure 3 1. Self assessment manikin (AdSAM)

PAGE 44

44 CHAPTER 4 RESULTS Description of the sample Analysis o f the survey data revealed that 295 participants accessed the study questionnaire online during the time it was activated. A review of the responses revealed that 53 respondents did not fully complete the survey, so this group was deleted from the study, in accordance with the procedure (Bodner 2006) described in Chapter 3. Each of the three random ad conditions in the survey contained print advertisements for two fictitious products (on e for a low involvement product and one for a hig involvement product). The emotional response to each ad was measured using the AdSAM scale, followed by two questions asking respondents if they were familiar with the product or advertisement. Of the 295 respondents, 35 (11.9%) said they were familiar with the ad or the product, and their responses were deleted from the analysis. Subtracting the responses from the two deletions, the total sample size for analysis consisted of 207 participants, which incl uded 57 Males (27.5%) and 150 Females (72.5%). Percentages for of males and females for each of the three groups were the same, and no significant sex differences were noted between groups (Table 4 1). An examination of the age distribution by conditions s howed no significant differences among the groups. Further, no significant differences in age were noted between sexes among the three conditional groups (F = .15, df = 5, p = .98, n.s.). (Table 4 2). Hypotheses Testing The overall premise of my study i s that a known sports celebrity will evoke a stronger emotional response to a marketing communication as compared to an unknown spokesperson or no spokesperson at all. Therefore, the hypothesis for each dependent variable predicted that the

PAGE 45

45 sports celebrit y endorsement would provide more influence on the dependent variables, resulting in a higher mean on the Pleasure, Arousal, and Dominance (PAD) AdSAM scale as compared to the other source conditions. To test Hypotheses 1, 2 and 3, a one way ANOVA was per formed, followed by a post hoc test if a significance difference between groups was noted. A two way ANOVA was used to analyze Hypothesis 4. Study respondents were randomly placed in three different groups. Each group reviewed two ads with the same source condition, but two different fictitious products; one product was identified as Power Plus, a sports drink, considered a low involvement product; the other product was called MAX, a smart phone, making it a high involvement product. H1a: A known sports celebrity endorsing a low involvement, fictitious product will be significantly more pleasing (more positive valance) compared to an endorsement by an unknown spokesperson or no spokesperson at all. Analysis of the pleasure dependent variable for the low involvement product revealed that the differences between the groups were not statistically significant. Therefore, there was no support for this hypothesis F = 2.87, df = 2, p = .059, n.s.) (Table 4 3). H1b : A known sports celebrity will be significantl y more exciting endorsing a low involvement, fictitious product compared to a non sports celebrity or no spokesperson at all. Partial support for this hypothesis was evidenced by the mean scores for the arousal dependent variable among the source groups. The post hoc test revealed that the sports celebrity ad (Arousal = 5.00) and the no spokesperson ad (Arousal = 5.04) were not significantly different, but were significantly different than the unknown spokesman ad (Arousal = 4.02 (Table 4 3). H1c : A know n sports celebrity will be significantly more controlling endorsing a low involvement, fictitious product compared to an unknown spokesperson or no spokesperson. As in the case of H1a, no support was found for this hypothesis. No significant differences we re noted among source groups. Actually, the means were similar: sports celebrity (Dominance = 5.70), unknown spokesperson (Dominance = 5.66) and no spokesperson (Dominance = 5.70) (Table 4 3). H2a : A known sports celebrity endorsing a high involvement, f ictitious product will be significantly more pleasing (more positive valance) as compared to an endorsement by an unknown spokesperson or no spokesperson at all. This hypothesis was not supported by the data, as no significant differences were established among the groups. The means by source type were as follows: sports celebrity, 5.71; unknown spokesperson, 5.60, and no spokesperson, 5.81 (Table 4 4).

PAGE 46

46 H2b: A known sports celebrity will be significantly more exciting endorsing a high involvement, fictit ious product as compared to an unknown spokesperson or no spokesperson at all. A significant difference was noted in the means, and a post hoc test showed that the sports celebrity group (5.04) and the no spokesperson group (4.91) were significantly differ ent than the unknown spokesperson group (4.02). So there was partial support for this hypothesis (Table 4 4) H2c A known sports celebrity will be significantly more controlling endorsing a high involvement, fictitious product compared to an unknown spoke sperson or no spokesperson. Analysis of the means reflected no significant differences among the groups: sports celebrity (5.68), unknown spokesperson (5.65) and no spokesperson (5.78). Therefore, this hypothesis was not supported (Table 4 4). When the end orser and the product are congruent, it has been suggested that the effect of ted that the sports celebrity would produce a significantly higher response in ads with the low involvement product compared to the high involvement product. These hypotheses are as follows: H3a : A known sports celebrity endorsing a low involvement, fictit ious product will be significantly more pleasing (more positive valance) as compared to the same known sports celebrity endorsing a high involvement, fictitious product. There was no support for this hypothesis. F = 1.605, df =5, p = .157 (Table 4.5) H3b : A known sports celebrity will be significantly more exciting endorsing a low involvement, fictitious product as compared to the same known sports celebrity endorsing a high involvement, fictitious product. Significant differences were noted within the sou rce groups for both low and high involvement product, but no interaction was revealed between the types of products. F = 4.86, df = 5, p=.000, sig./ Interaction: F=1.373, df=2, p = .255, n.s. (Table 4.5). H3c : A known sports celebrity will be significan tly more controlling endorsing a low involvement, fictitious product compared to the same known sports celebrity endorsing a high involvement, fictitious product. No significant differences were found, resulting in no support for the hypothesis (Table 4 5). Purchase intent for low and high involvement products was predicted to be higher for ads with sports celebrity as compared to the ads with an unknown spokesperson or no spokesperson. These hypotheses are as follows:

PAGE 47

47 H4a : Purchase intent will be si gnificantly greater for a low involvement, fictitious product endorsed by a sports celebrity than a low involvement, fictitious product endorsed by an unknown spokesperson or product without a spokesperson. The sports celebrity ad (2.70) and the no spokesp erson ad (3.11) were significantly different than the unknown spokesperson ad (2.06), providing partial support for this hypothesis (Table 4 6). H4b : Purchase intent will be significantly greater for a high involvement, fictitious product endorsed by a sp orts celebrity than a high involvement, fictitious product endorsed by an unknown spokesperson or the product without a spokesperson. No significant differences were noted; therefore, no support was found for this hypothesis (Table 4 6). Table 4 1. Gende r by experimental group and product involvement Group/Gender High Involvement Low Involvement Sports Celebrity # % Total # % Total Male 20 29.0 69 20 29.0 69 Female 49 79.0 49 79.0 Unknown # % Total # % Male 22 26.2 84 22 26.2 84 Female 62 73.8 62 73.8 No Spokesperson # % Total # % Male 15 27.8 54 15 27.8 54 Female 39 72.2 39 72.2 Total # % Total # % Male 57 27.5 207 57 27.5 207 Female 150 72.5 150 72.5 Total 207 100.0 207 100.0 1 Each group rated a high involvement print ad and a low involvement print ad. Therefore, there were equal number of males and females within each high and low involvement group by condition. Chi square = .301, df = 5, p=.998, n.s.

PAGE 48

48 Table 4 2. Average age by product involvement and celebrity group s Level of Product Involvement 1 Spokesperson Group Total 1 Low (Power Plus) High (MAX) Average Age Average Age Average Age Celebrity 22.45 22.45 22.45 Non Celebrity 23.39 23.39 23.39 No Spokesperson 23.02 23.02 23.02 Average Age 22.98 22.98 22 .98 2 1 Each group rated a high involvement ad and a low involvement ad. 2 Chi square for Gender by Low/High Involvement Product 2 F=.359, df = 5, p = .876, n.s Table 4 3. PAD averages for low involvement product by spokesperson group Group Sports Ce lebrity Unknown No Spoke s person Avg. Avg. Avg. F df p sig. Pleasure 6.20 5.60 5.96 2.87 2 .059 n.s. Arousal 5.00 4.02 5.04 6.06 2 .003 sig. Dominance 5.70 5.66 5.70 .02 2 .984 n.s. Table 4 4. PAD averages for high involvement product by spokesperson group Group Sports Celebrity Unknown No Spoke s person Avg. Avg. Avg. F df p sig. Pleasure 5.71 5.60 5.81 .32 2 .727 n.s. Arousal 5.04 4.02 4.91 6.08 2 .003 sig. Dominance 5.68 5.65 5.78 .082 2 .922 n.s.

PAGE 49

49 Table 4 5. Emo tional response by product involvement and celebrity group Low (Power Plus) High (MAX) Group Sports Celebrity Avg. Unknown Avg. No Spokes person Avg. Sports Celebrity Avg. Unknown Avg. No Spokes person Avg. Pleasure 1 6.20 5.60 5.96 5.71 5.60 5.81 A rousal 2 5.00 4.02 5.04 5.04 4.02 4.91 Dominance 3 5.70 5.66 5.70 5.68 5.65 5.78 1 Pleasure: F = 1.605, df =5, p = .157, n.s. 2 Arousal: F = 4.86, df = 5, p=.000, sig./Interaction: F=1.373, df=2, p = .255, n.s. 3 Dominance: F = .041, df =5, p = .999, n.s. T able 4 6. Purchase intent by group and product Group Sports Celebrity Unknown No Spokesperson F df P Sig. Purchase Intent Low Involvement 2.70 2.06 3.11 8.92 2 .000 Sig. Purchase Intent High Involvement 2.30 2.06 2.56 2.01 2 .137 n.s. Two way ANOVA: F = 5.213, df = 5, p = .000, sig Interaction: F = .061, df = 2, p = .941, n.s.

PAGE 50

50 CHAPTER 5 CONCLUSIONS, LIMITATIONS AND FUTURE RESEARCH My study investigated emotional response to print ads that manipulated the message source, focusing on a sports celebrity e ndorser. In addition to a sports celebrity endorser group, age and gender. The third group had no spokesperson, and only presented the products. The three groups representing the three source conditions, each containing the same two fictitious products, were randomly assigned to online questionnaire respondents. The theoretical framework for my research included the two message source models: the source credibili ty model and the source attractiveness model. Specifically for the sports celebrity, up well with the attributes of the endorsed product. Here we review findings of the study, and provide guidance to academics and practitioners regarding selection of sports celebrity endorsers. The limitations of the study and future research are discussed. Emotional Response to Sports Celebrity Endorser Hypotheses H1a, H1b, H1c, H2a, H2b and H 2c predicted that the emotional response to a sports celebrity endorsed marketing communication would be significantly more pleasing (more positive valance), more exciting, and more controlling than an unknown spokesperson or no spokesperso n at all for both low and high involvement products. The facial expressions of the sports celebrity and the unknown spokesperson were changed fr om a smiling face for the low involvement product to a more serious facial expression for the high involvemen t product. This was intentional to convey the difference in the product considerations. It w as noted that the pleasure scor es for the sports celebrity were lower for the

PAGE 51

51 high involveme nt product (6.20 5.71), and this may be exp lained by the change in fac ial expressions. However, the pleasure scores for the two unknown spokesperson ads were very similar (5.04 5.00). It was strongly believed that the low involvement, fictitious product (Power Plus, a sports type drink), would match up well with the sport s celebrity, significantly more pleasing (more positive valence) than the other source conditions. Hypothesis 1a received no support. The same was true for the Hypothesis H1c regarding control (dominance in the PAD model). However, some meaningful findings related to arousal may be important regarding use of a sports celebrity. The Importance of Arousal An important finding was that the arousal mean for the sports celebrity and the no spokesperson condition was significantly higher than the unknown sp okesperson. This was found for both low and the high involvement product print ads, reflecting the fact that these ads were determined to be more engaging by the respondents. Advertisers employ celebrities for a myriad of reasons including their ability to attract listeners have hundreds of choices, and are bombarded with advertisements of all descriptions, creating an audience that surfs channels, flips pages and zaps ads (Miciak & Shanklin 1994). Ad avoidance has become part of American life, and it is accomplished by various methods including cognitive, behavioral and mechanical means. Magazine and newspaper readers believe ads are uninteresting, unusefu l, and annoying. Radio and TV ads are avoided by surfing and zapping (Speck & Elliott 1997). Ad clutter has been cited as the main reason internet users avoid ads (Cho & Cheon 2004).

PAGE 52

52 Since the sports celebrity ad was found to be significantly more engagin g than the unknown spokesperson ad for both low and high involvement product ads. Advertisers should seriously contemplate their use. Once readers are engaged in a print ad by a sports celebrity, they may elaborate on the arguments of the advertisement. At a minimum, the celebrity will prevent the reader from flipping the page, and get the reader to look at the print ad. Consideration should also be given to well executed ads, since the ad with no spokesperson also performed equally with the sports celebr ity in this study. The ads were created by a recognized professional designer, whose portfolio includes ads for major brands. Although the products were fictitious, the ads were colorful, they showed the product, and they produced an emotional response. Pu rchase Intent From the literature (Kamins 1990, Lynch & Schuler, Koernig & Boyd 2009), it was believed that the purchase intent for the low involvement product endorsed by the sports celebrity would be significantly higher than it would be for the other two source groups (unknown spokesperson and no spokesperson). It was expected that the sports celebrity would Another relevant finding of my study is that the purchase intent for the sports celebrity and the no spokesperson group were significantly higher than the unknown spokesperson group. involvement product. Prior research by Koernig and Boy up who was not an athlete. So my research partially supports the finding by Koernig and Boyd (2009).

PAGE 53

53 Using proprietary, AdSAM software, emotional groupings of adjectives were produced from experimental data. These groupings are based on pleasure, arousal and dominance (PAD) scores for 135 emotion adjectives (Morris & Bradley 1994; Mehrabian & Russell 1977),. The groupings appear to show a difference between the sports celebrity (Tebow), and the unknown spokesperson, although the difference in the PAD means are not significantly different. Forty involvement were included in comfortable, warmed or enthusiastic emotional groups. Adjectives associated with thankful, exuberant, and courageous reflecting a connection to the spor ts celebrity not the product. This finding may show no transfer of the sports involvement brand (Figure 5 1). The unknown spokesperson for the low involvement condition received no responses in comfortable, warmed or enthusi astic range. Seventy five percent of the responses were indifferent or ambivalent while 25% were sullen. Adjectives associated with these emotional groups were as follows: unimpressed, uninterested, cynical, haughty, unexcited, uncaring, bored, and unconce rned (Figure 5 2). The analysis of these adjectives dictates that practitioners must be Although the unknown model was similar to Tebow in gender and age, the unknown p ersona may not prevent readers from flipping the page, and avoiding the ad. The low involvement ad featuring the product with no spokesperson performed well, but not as well as Tebow in the top tier of the positive range of emotional groups, receiving 31% in the comfortable, warmed, enthusiastic range compared to 45% for Tebow. Adjectives corresponding to no spokesperson source are different than Tebow, including the following:

PAGE 54

54 secure, mature, daring, aspiring, wholesome, playful, amazed, and reasonable, perhaps reflecting feelings more directed to the product (Figure 5 3). An important point to consider in the comparison of the sports s tatistical analysis. However, examining the emotional groupings appears to present different emotions between the two source groups (sports celebrity and no spokesperson). Arousal means for the sports celebrity and the no spokesperson groups were signific antly different for the high involvement product compared to the unknown spokesperson, but the emotional groups expressed by AdSam adjectives may be more helpful in discerning a difference between them. The unknown spokesperson (Figure 5 5) had a higher pe rcentage of responses for enthusiastic compared to Tebow (Figure 5 4). The adjective courageous was missing from the Tebow enthusiastic list, but included for the no spokesperson group (Figure 5 6). A noted shift seems to have occurred for the high involve ment product, meaning MAX is not as good a match up for Tebow. Limitations United States corporations spent over $17 billion on sports endorsements and sponsorships in 2010. Globally, more than $46 billion has been committed on sponsorships. Nike reported in their 2010 annual report that they expected to spend $712 million on sports celebrity endorsements (Koo, Ruihley and Dittmore 2012). It is vitally important for companies to be accurate in their selection of sports celebrity endorsers. Many researcher s have contributed their ideas on how to make the process more successful. Hopefully, my study will contribute to the insight of future researchers, but there are limitations to my research including the following:

PAGE 55

55 First, the size of the sample may have b een too small to cause significant differences in the dependent variables of pleasure and dominance. The differences in arousal should have been accompanied by significant differences in pleasure. The original sample size of nearly 300 was reduced to 207 b ecause of missing data, and elimination of respondents, who believed they were familiar with the fictitious brands and ads. Second, the sample was a convenient sample of college students, so it is not generalizable. Further, students received extra class credit for participating in the survey, and some students may have not been fully engaged in assessing their true feelings. celebrity endorser for my research was completed before the 2012 National Football selection, he had a good year playing for Denver during the 2011 season, and signed a major endorsement deal with Jockey in June 2012 (PR Newswire, New York, June 23, 2012). The 2012 season was a disaster for Teb ow, seeing limited playing time for the New York Jets, completing 6 of 8 attempted passes, and running for 102 yard s (ESPN.Boston ). The lack of playing time may St utton 2008) Future Research My study has produced some very meaningful results regarding sports celebrity endorsers, and emotional response using the AdSAM scale. The arousal factor finding will be helpful to practitioners, searching for a way to bring at breaking through the ad clutter. The emotional groups were very helpful in a more dynamic noting that the sports celebrity per formed better with a low involvement, sports related product.

PAGE 56

56 Finally, there may be support for the concept that a bad season can have an impact on a sports celebrity endorser. Future researchers should consider selecting sports celebrities and products that produce very high scores on the source credibility scale developed by Ohanian (1991). If the sports celebrities are considered to be an expert for a specific product, then a replication of the study may produce more significant differences in (PAD) mo del. Sampling that is more diverse and larger than my study should also be part of future research. In my study, the three source conditions included two ads each; one ad contained a low involvement product while the other had a high involvement product. Future researchers may want to include two low involvement products t o use in the study.

PAGE 57

57 Figure 5 1. Sports celebrity low involvement

PAGE 58

58 Figure 5 2. Unknown spokesperson low involvement

PAGE 59

59 Figure 5 3. No spokesperson low involvement

PAGE 60

60 Figure 5 4. Sports celebrity MAX

PAGE 61

61 Figure 5 5. Unknown spokesperson M AX

PAGE 62

62 Figure 5 6. No spokesperson MAX

PAGE 63

63 APPENDIX ONLINE QUESTIONNAIRE Q1 Thank you very much for agreeing to participate in this study, conducted by a University of Florida researcher. The purpose of the study is to examine your response to a celebr ity endorsing different products. Your participation is completely voluntary and if you choose not to complete the survey, it will not affect you in any way. You do not have to answer any questions you do not wish to answer. There are no right or wrong ans wers, only your valuable opinion. If you do complete the survey, the information you provide will be kept confidential to the extent provided by law. There are no direct benefits or compensation to you for participating in this study. For information about your rights as a research participant contact the IRB office at 352 392 0433. The results of the survey will be analyzed at the University of Florida in Gainesville, Florida, and may be included in a doctoral dissertation. The survey should take approxima tely 10 to 15 minutes to complete. Please try to answer all of the questions to the best of your ability. Participation in this survey should not create any physical, psychological or economic risks. If you have questions about the study, please contact Dr Jon Morris at jmorris@jou.ufl.edu or Walter J. Starr at wjstarr@ufl.edu I have read the information above and understand my options A gree Disagree Q2 What is your gender? Male Female Q3 What is your age?

PAGE 64

64 Instructions This is AdSAM, a si mple but scientifically validated way for you to indicate your emotional response to several questions throughout the survey. and your feelings. The three row s of pictures represent feelings that range in nature from: As you can see, AdSAM is arrayed along three rows, each representing different dimensions of emotion. The ends of each row are the extremes and there is a whole range of reactions in between th e extremes. There are two print advertisements for you to review, for each ad please: Indicate your immediate emotional reaction using all three rows Mark one circle on each row for a total of three marks per question. Either select a circle directly bel ow a Manikin, or in between two Manikins, depending on your reaction. Don't over think it! Simply indicate your immediate reaction. Please do not spend a lot of time thinking about each question. We will show you a few advertisements. Just indicate how ea ch advertisement makes you feel. Before we begin reviewing advertisements, we would like to use the Manikins to answer the following question:

PAGE 65

65 Q4 How do you normally feel? Q5 Q6 Now you will review ads using the Manikins, Q7 Review this ad.

PAGE 66

66 ... how does it make you feel?

PAGE 67

67 Q8 Q9 Q10 Are you familiar with this product? Yes No Q11 Have you seen this ad before? Yes No Q12 How likely are you to purchase the product featured in the ad in the next month? Very Unlikely Unlikely Somewhat Unlikely Undecided Somewhat Likely Likely Very Likely Q13 Review this ad

PAGE 68

68 ... how does it make you fee l?

PAGE 69

69 Q14 Q15 Q16 Are you familiar with this product? Yes No Q17 Have you seen this ad before? Yes No Q18 How likely are you to purchase the product featured in the ad in the next month? Very Unlikely Unlikely Somewhat Unlikely Unde cided Somewhat Likely Likely Very Likely Now you will review ads using the Manikins,

PAGE 70

70 Q19 Review this ad

PAGE 71

71 ...how does it make you feel? Q 20 Q21 Q22 Are you familiar with this product? Yes No Q23 Have you seen this ad before? Yes No Q24 How likely are you to purchase the product featured in the ad in the next month? Very Unlikely Unlikely Somewhat Unlikely Undecided Somewhat Likely Likely Very Likely Q25

PAGE 72

72 Review this ad

PAGE 73

73 ... how does it make you feel ? Q57 Q59 Q28 Are you familiar with this product? Yes No Q29 Have you seen this ad before? Yes No Q30 How likely are you to purchase the product featured in the ad in the next month? Very Unlikely Unlikely Somewhat Unlikely Undecid ed Somewhat Likely Likely Very Likely Q31 Review this ad

PAGE 74

74 ... how does it make you feel?

PAGE 75

75 Q61 Q63 Q34 Are you fa miliar with this product? Yes No Q35 Have you seen this ad before? Yes No Q36 How likely are you to purcha se the product featured in the ad in the next month? Very Unlikely Unlikely Somewhat Unlikely Undecided Somewhat Likely Likely Very Likely Q37 Review this ad

PAGE 76

76 ... how does it make you feel?

PAGE 77

77 Q75 Q77 Q40 Are you familiar with this product? Yes No Q41 Have you seen this ad before? Yes No Q43 How likely are you to purchase the product featured in the ad in the next month? Very Unlikely Unlikely Somewhat Unlikely Undecided Somewhat Likely Likely Very Likely Q50

PAGE 78

78 Thank you for taking the survey. If you are taking the survey for extra class credit, please type your name, your UFID, class name, number, and section number below where indicated. An example of class name and number would be ADV 3502. Q51 Thank you for completing the survey. Your input will provided valuable data that will contribute to the success of the project. If you are completing this survey for class credit, please include your full name, your UFID, and class name such as ADV 3502 with sectio n number. Name, UFID, Class name and section

PAGE 79

79 REFERENCE LIST Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior. Englewood Cliffs, N.J: Prentice Hall. Amos C ., Holmes G ., & Strutton, D. (2008). Exploring the Relatio nship Between Celebrity Endorser Effects and Advertising Effectiveness. International Journal of Advertising 27 (2), 209 Applbaum, Ronald L, Anatol Karl W E. (1973). Dimensions of source credibility: A test for reproducibility. Speech Monographs, 40 (3), 231 237. Atkin, C., & Block M. (1983). Effectiveness of for reproducibility Speech Monographs, 40 (3), 231 237 Celebrity Endorsers. Journal of Advertising Research 23, (1), 57 61 Berlo, D., Lemert, J. & Mertz, R. (1970). Dimensions for Evaluating the Acceptability of Message Sources. The Public Opinion Quarterly 33, (4)), 563 576 Bettinghaus, E. P., & Cody, M. J. (1987). Persuasive communication New York: Holt, Rinehart, and Winston. Biswas, D. Biswas A., & Das N. (2006). The Diffe rential E ffe cts of Celebrity and Expert on Consumer Risk P erceptions. Journal of Advertising, 35, (2), 17 31. Bless, H., Bohner, G., Schwarz, N., & Strack, F. (1990). Mood and Persuasion: A Cognitive Response Analysis. Personality and Social Psychology Bulletin 16 (2), 331 345. Bohner, G. & Apostolidou, W. (1994). Mood and Persuasion: Independent Effects of Affect Before and After Message Processing. The Journal of Social Psychology 134, (5), 707 709. Celebrity Endorsement: Come to your senses. (2007, July). Brand Strate gy,42. Chaiken S (1980) Heuristic versus systematic information processing and the use of source versus message cues in persuasion. Journal of Personality and Social Psychology 39, (5), 752 766. Chaiken, S. & Maheswaran D. (1994). Heuristic Processing Can Bias Systematic Processing: Effects of Source Credibility, Argument Ambiguity, and Task Importance on Attitude Judgment. Journal of Personality and Social Psychology, 66, (3) 460 473. Choi, S. M. & Rifon N J. (2007). Who Is the Celebrit y in Advertising? Understanding Dimensions of Celebrity Images. Journal of Popular Culture, 40, (2), 304 324. Cialdini, R. B. (1984). Influence: How and why people agree to things New York: Morrow

PAGE 80

80 Clow, K.E., James, K.E., Kranenburg, K.E. & Berry, C.T. (2006). The Relationship of the Visual Element of an Advertisement to Service Quality Expectation and Source Credibility. Journal of Services Marketing. 20, (6), 404 411. Clow, K. E James, K. E ., Sisk, S. E & Cole, H. S (2011). Source Credibility Visual Strategy and the Model in Print Advertisements Journal of Marketing Development and Competitiveness 5 (3), 24 31. D ealer, T. P. (2004 11 14). The H istory of Mr. Coffee. T he Plain D ealer (Cleveland, Ohio: 1961), A.15. DeSarbo, W. S. & Harshman R. A. (1985). Celebrity Brand Congruence Analysis. Current Issues and Research in Advertising. H. Leigh and C.R. Martin, Jr., eds., Ann Arbor, MI: Division of Research, Graduate School of Business Administration, The University of Michigan, 17 52. Journal of Consumer Research 3, (4), 223 232. Edell J A. & Burke, M. C. (1987) The Power of Feelings in Understanding Advertising Effects. J ournal of Consumer Research (3) 421 433 Erdogan, B. (1999). Celebrity Endorsement: A Literature Review. Journal of Marketing Management 15, (4), 291 314. Erdogan, B., Baker, M. J., & Tagg, S. (2001). Selecting Celebrity Endorsers: Perspective. Journal of Advertising Research 41 (93), 39 48. Friedman, H. H. & Friedman, L. (1979). Endorser Effectiveness by Product Type. Journal of Advertising Resea rch, 19, (3) 63 71. Friestad, M. & Wright, P. (1999). Everyday Persuasion knowledge Psychology & Marketing 16 (2), 185 194. Journal of Business Communica tion 25, (3), 96 97. Guthrie, K. L. (2009). Emotional response to typography : The role of typographic variations in emotional response to advertising Gainesville, Fla.: University of Florida. Ha, L. & Litman, B. R. (1997) Does Advertising Clutter Have Diminishing and Negative Returns? Journal of Advertising 26, (1), 31 42 understandings of Journal of Advertising 25, (2), 19.

PAGE 81

81 Heiser, R. S. Sierra J. J., & Torres I. M. (2008). Creativity via cartoon spokespeople in print ads. Journal of Advertising, 37 4), 75 84. Holbrook, M. B. & Batra, R. (1987) Assessing the Role of Emotions as Mediators of Consumer Res ponses to Advertising. Journal of Consumer Research 14, (3), 404 420 & Marketing 1, (2), 45 64. Horai, J., Naccari, N. & Fatoullah, E. (1974). The Effects of Ex pertise and Physical Attractiveness Upon Opinion Agreement and Liking. Sociometry 37 (4 ), 601 606 Hovland, C. & Weiss, W. (1952) The Influence of Source Credibility on Communication Effectiveness. The Public Opinion Quarterly 15, (4), 6 35 650. Complexity Journal of Research in Personality 43(5), 880 889, Jain, S., P & Posavac, S. S., (20 0 Repurchase Attribute Verifiability, Source Credibility, Journal of Consumer Psychology 11, (3), 169 180 Propaganda and Persuasion 4 th edn. Thousand Oaks and London: Sage Kahle, L. R. & Homer, P. M. (1985). Physical Attractiveness of the Celebrity Endorser: A Social Adaptation Perspective. Journal of Consumer Research 11, (4), 954 961 Published by: The University of Chicago Press Kamen, J. M., Azhari, A. C., & Kragh, J. R. (1975). What a Spokesman Does for a Sponsor. Journal of Advertising Research 15, (2), 17 24. Kamins, M. A. (1989). Celebrity and Noncelebrity Advertising in a Two Sided Context Journal of Advertising Research 29, (3), 17 34. Kelman, H. C. (1961). Processes of Opinion Change. The Public Opinion Quarterly 25 (1), 57 78 Kim, J. & Morris, J. D. P ower of A ffective response and C ogn itive S tru cture in Product trial A ttitude F Journal of Advertising, 36 ( 1), 95 106. Reeves Journal 84, (6), 20. Koernig, S., & Boyd, T. h up Effect and Athlete Endorsers for Sport and Non Sport Brands Sport Marketing Quarterly 18, (1), 25 37.

PAGE 82

82 Koo, G. Y., Ruihley, B. J, & Dittmore, S. W. (2012). Impact of Perceived On Field Performance on Sport Celebrity Source Credibility. Sport Mark eting Quarterly, 21, (3), 147 158. Kuykendall, D., & Keating, J. P. (1990). Mood and Persuasion: Evidence for the Differential Influence of Positive and Negative States. Psyc hology & Marketing, 7, (1), 1 9. Lane, Randall. (1996, December). Nice guys finish first. Forbes, 158, (14), 236 242. The American Political Science Review 21, (3), 627 631. Lear, K E., Runyan, R. C., & Whitaker, W. H... Products International Journal of Retail & Distribution Management 37, (4), 308 321 Lynch, J. & Schuler, D (1994). The Matchup Effect of Spokesperson and Product Congruency: A Schema Theory Psychology & Marketing 11, (5) 417 430. Maathuis, O., Rodenburg, J. & Sikkel, D. (2004). Credibility, Emotion or Reason? Corporate Reputation Review 6, (4 ), 333 345. Mackie, D. M. (1987). Systematic and nonsystematic processing of majority and minority persuasive communications. Journal of Personality and Social Psychology 53, (1), 41 52. Maddux, J & Rogers R. (1980). Effects of source expertness, physical attractiveness and supporting arguments on persuasion: A case of brains over beauty. Journal of Personality and Social Psychology 39, (2), 235 244. Magnini, V., Honeycutt, E., & Cross, A (2008). Understanding the use of celebrity endorsers for hospitality firms. Journal of Vacation Marketing, 14, (1), 57 69. Markiewicz, D. A. (1996, 8 22). Irvin case il lustrates risk of athletes pitching products Fort Worth Star Telegram 1. McCracken, G. (1989). Who is the Celebrity Endorser? Cultural Foundations of the Endorsement Process. Journal of Consumer Research 16, (3), 310 321. Anthropological Quarterly 74, 4), 163 169 McKee, S. (2008, November 17 ). The Trouble with Celebrity Endorsements http://www.businessweek.com/stories/2008 11 14/the trouble with celebrity endorsementsbusinessweek business news stock market and financial advice Mehrabian, A., & J Russell (1977 ). Evidence for a T hree Factor Theory of Emotions. J ournal of Research in Personality 11, (3 ) 273 294

PAGE 83

83 Meyers Levy, J. & Malaviya, P. (1994). Heuristic Processing Can Bias Systematic Processing: Effects of Source Credibility, Argument Ambiguity, and Task Importance on Atti tude Judgment. Journal of Personality and Social Psychology, 66, (3), 460 473. Miciak, A. R, & Shanklin, W. L. (1994). Choosing Celebrity E ndorsers. Marketing Management 3, (3), 50. ionated Statements and Responses to Persuasive Communications. Speech Monographs 36, (1), 17 34. Mills, J. & Harvey J. (1972). Opinion change as a Function of When Information About the Communicator is Received and Whether He is Attractive or Expert J ournal of Personality and Social Psychology 21, (1), 52 55. Mintz, P. M. & Mills, J., (1971). Effects of Arousal and Information About Its Source Upon Attitude Change. Journal of Experimental Social Psychology 7, (6 ) 561 570. Morris, J. D. (199 5). Observations: SAM: The Self Assessment Manikin. Journal of Advertising Research 35 (6), 63 68 Morris, J. D., Chongmoo, W., Geason, J. A., & Jooyoung, K. (2002). The Power of Affect: P redicting Intention. Journal of Advertising Research 42 (3), 7 17. Morris, J. D., Woo, C. & Singh A. (2005). Elaboration Likelihood Model: A Missing Intrinsic Emotional Implication. Journal of Targeting, Measurement and Analysis for Marketing 12, (1), 79 98. Norman, R. (1976) When What Is Said Is Important: A Comparison of Expert and Attractive Sources. Journal of Experimental Social Psychology 12, (3, 294 300. Ohanian, R. (1990). Construction and Validation of a Scale to Measure Celebrity Endorsers' Perceived Expertise, Trustworthiness and Attractiveness. Journal of Advertising 19, (3), 39 52. Intention to Purchase. Journal of Advertising Research 31, (1), 46 54. Olstad, S., & James, R. (2009). A Brief History Of: Cigarette Advertising. Time 173 (25), 14 Patzer G L. (1983) Source Credibility as a Function of Communicator Physical Attractiveness Journal of Business Research 11, (2), 229 241. Perloff, R. M. (1993). The Dynamics of Persuasion New Jersey: Lawrence Erlbaum Associates Petty, R. E. (1981). Cognitive responses in persuasion. Hillsdale, N.J.: L. Erlbaum Associates.

PAGE 84

84 Petty, R.E. & Cacioppo, J. T. (1986) Communication and Persuasion: Central and Peripheral Routes to Attitude Change, New York: Spr inger Verlag. Poels, K., & Dewitte, S. (2008). Getting a Line on Print Ads. Journal of Advertising. 37 (4), 63 74. Pornpitakpan, C. (2004) The Persuasiveness of Source Credibility: A Critical Review of Five ial Psychology, 34, (2), 243 275. Priester, J. P. & Petty, R. E. (2003). The Influence of Spokesperson Trustworthiness on Message Elaboration Attitude Strength, and Advertising Effectiveness. Journal of Consumer Psychology. 13, (4), 408 421. Rathbun, E. A.. (2000). Clutter, C lutter E Broadcasting & Cable 130 (36), 58. Russell, J. A.&, Mehrabian, A. (1977) Evidence for a T hree F actor T heory of E motions. Journal of Research in Personality 11, (3), 273 294 San toli, M., & Bary, A (2009, December). 2010 Musings; Will Tiger's Loss Be Gain ? Barron's, 89, (51), 15 16 Schefter, A. & Werder, E. (June 11, 20 13 ). Patriots Sign Tim Tebow Tim Tebow signs with New England Patriots ESPN Boston http://espn.go.com/boston/nfl/story/_/id/9362801/tim tebow signs new england patriot s Sherman, S. P. (1985 08 19). When You Wish Upon a Star. Fortune, 112, (4), 66. Sirak, R. (2011 05 23). Tiger's Troubles. Golf World (Pinehurst, N.C.), 64, (33), 47. Smith R. (1973). Source Credibility Context Effects. Speech Monographs 40, (4), 303 Snyder, M. & Rothbart, M. (1971). Communicator Attractiveness and Opinion Change. Canadian Journal of Behavioural Science/Revue canadienne des sciences du comportement 3, (4), 377 387. Speed, R., & Thompson, P. Academy of Marketing Science Journal 28, (2), 226 238. Sports Illustrated. (2011). The 2011 Fortunate 50. Available at http://sports illustrat ed. cnn .com/ specials / fortunate50 2011 / index html Stafford, M., Spears, N. E., & Chung kue, H. (2003). Celebrity Images in Magazine Advertisements: An Application of the Visual Rhetoric Model. Journal of Current Issues & Research In Advertising 25 (2), 13 20.

PAGE 85

85 Sujan, M. (1985). Consumer Knowledge: Effects on Evaluation Strategies Mediating Consumer Judgments. Journal of Consumer Research 12, (1), 31 46. Till, B. D. & Busler M., (1998) Matching Products with Endorsers: Attractiveness Versus Exper tise", Journal of Consumer Marketing 15, (6), 576 586. Whitehead, J. L. (1968). Factors of Source Credibility. Quarterly Journal of Speech 54, (1), 59. Wundt, W. (1907) Outlines of psychology. Leipzig: Wilhelm Englemann. Zajonc, R. B. (1980). Feelin g and Thinking: Preferences Need No Inferences. American Psychologist 35, (2), 151 175. Zajonc, R. B. & Markus, H. (1982). Affective and Cognitive Factors in Preferences. Journal of Consumer Research 9, (2), 123 131.

PAGE 86

86 BIOGRA PHICAL SKET CH Walter John Starr was born in St. Paul, Minnesota in 1945 His Bachelor of Science degree in biology was conferred in 1967. During his service in the United States Air Force, he earned After 32 yea rs working in media time sales and management, h e was accepted into the doctoral program at the College of Journalism and Communication, University of Florida, and began his studies in August 2008. During his time at the University of Florida, he conducted several research projects on advertising under the guidance of Dr. Jon D. Morris. He received his Ph.D. from the University of Florida in the summer of 2013.