EFFECTIVENESS OF DISGUSTING IMAGE S AS A PERSUASION TOOL IN PUBLIC SERVICE ANNOUNCEMENTS FOR CONTROVERSIAL TWO-SIDED PUBLIC ISSUES By REBECCA A. FOWLER A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ADVERTISING UNIVERSITY OF FLORIDA 2006
Copyright 2006 by Rebecca A. Fowler
This document is dedicated to my family.
iv ACKNOWLEDGMENTS I would like to thank Dr. Jo rge Villegas for all of his time, encouragement and words of wisdom throughout this entire pro cess. I could not have asked for a better advisor. In addition, I would like to tha nk both of my thesis committee members, Dr. John Morris and Dr. Cynthia Mort on, for all of their support. I would also like to thank Jody Hedge for all of her guidance, Dr. John Sutherland for granting me the opportunity to pursue my masterÂ’s degree in advertising here at the University of Florida, and all the members of my cohort for their friendship and support. Finally, I would like to thank my family and my friends for believing in me.
v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES............................................................................................................vii LIST OF FIGURES.........................................................................................................viii CHAPTER 1 INTRODUCTION........................................................................................................1 2 REVIEW OF LITERATURE.......................................................................................8 The Effectiveness of Past Public Service Announcement Appeals..............................8 Informational Appeals...........................................................................................9 Social-Modeling Appeals....................................................................................10 Positive Appeals..................................................................................................11 Fear Appeals........................................................................................................12 Empathy Appeals.................................................................................................12 Shock Advertisements and the Violation of Social Norms........................................15 Disgust........................................................................................................................ 20 3 METHOD...................................................................................................................27 Subjects and Setting....................................................................................................27 Measures.....................................................................................................................27 Stimuli........................................................................................................................ .30 Procedure....................................................................................................................33 4 RESULTS...................................................................................................................36 Characteristics of the Sample.....................................................................................36 Main Experiment.................................................................................................36 Posttest.................................................................................................................37 Emotional Response to the Advertisement.................................................................39 Research Question One...............................................................................................40 Pro-Life................................................................................................................40 Pro-Vegetarian.....................................................................................................44 Research Question Two..............................................................................................48
vi Pro-Life................................................................................................................49 Pro-Vegetarian.....................................................................................................51 Assessing Anger.........................................................................................................53 Pro-Life................................................................................................................53 Pro-Vegetarian.....................................................................................................54 5 DISCUSSION AND CONCLUSION........................................................................57 Limitations..................................................................................................................61 Future Research..........................................................................................................61 APPENDIX A SAMPLE OF THE PRE-TEST..................................................................................63 B NEUTRAL PRO-LIFE ADVERTISEMENT.............................................................64 C DISGUSTING PRO-LIFE ADVERTISEMENT.......................................................65 D NEUTRAL PRO-VEGETA RIAN ADVERTISEMENT...........................................66 E DISGUSTING PRO-VEGETA RIAN ADVERTISEMENT......................................67 F SAMPLE OF THE MAIN QUESTIONNAIRE.........................................................68 G POSTTEST.................................................................................................................69 LIST OF REFERENCES...................................................................................................70 BIOGRAPHICAL SKETCH.............................................................................................74
vii LIST OF TABLES Table page 1 Pretest results for four images chosen...........................................................................32 2 Pleasure, arousal and dominance results for pro-life advertisements............................41 3 Open-ended results for Â“pushyÂ” for pro-life advertisements.........................................43 4 Posttest results for pro-life advertisements....................................................................43 5 Pleasure, arousal and do minance results for pro-vegetarian advertisements.................45 6 Open-ended results for Â“defensiveÂ” for pro-vegetarian advertisements........................47 7 Posttest results for pro-vegetarian advertisements.........................................................47 8 Levels of disgust for pro-life adver tisements by open-ended and Izard scale...............50 9 Levels of disgust for pro-vegetarian a dvertisements by open-ended and Izard scale....52 10 Anger for pro-life advertisements by Izard scale.........................................................54 11 Anger for pro-vegetarian advertisements by Izard scale.............................................54
viii LIST OF FIGURES Figure page 1 Preliminary shock appeal base d on a model by Dahl in 2003.......................................16 2 Advertising Self-Assessment Manikin (AdSAM) measure developed by Dr. John Morris. Retrieved with permission from www.adsam.com, August 2005...............29 3 Advertisement Self-Assessment Manikin perceptual map: pleasure and arousal.........38
ix Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Advertising EFFECTIVENESS OF DISGUSTING IMAGE S AS A PERSUASION TOOL IN PUBLIC SERVICE ANNOUNCEMENTS FOR CONTROVERSIAL TWO-SIDED PUBLIC ISSUES By Rebecca A. Fowler May 2006 Chair: Jorge Villegas Major Department: Advertising This research examines the emotion of di sgust and the role it plays in persuasion within public service announcements for twosided controversial public issues. In particular, we look at how much of an im pact a disgusting image has on the emotional reactions of viewers. A review of the literature on shock advertising reveals that shocking tactics have proved effective at capturing the attention of viewers due to the fact that they violate social norms by presenting images or words th at are unexpected. The review also reveals that disgusting images are a specific type of shock appeal that causes the viewers to turn away from what they see. It is unknown how effective di sgust appeals are, but past studies have suggested that they would be effective at persuading viewers toward an advertisement sponsor's message when associated with the opposition's position within a two-sided public issue. Furthe r, the literature reveals that the theoretical versus the lay definition of the term disgust are very different . While the theoretical definition results in
x a feeling of physical illness, the lay defi nition corresponds more with the theoretical meaning of anger or at least a strong combina tion of anger and disgust. Therefore, it is important to note what meaning of disgust is being used in a study about this particular emotion. This study attempts to prove the hypothesis that disgusting images will be effective at persuasion toward the sponsor's message when associated with the opposition's position within a two-sided public issue. This study also looks at the theoretical and lay definition of disgust and how they contri bute to the effectiveness of persuasion. The study determines that disgusting images can be effective, especially when it comes to the initial attention-grabbing factor . However, it also determines that the lay meaning of disgust, which includes anger, may be counterproductive and cause viewers to become defensive or deny the credibility of the message sponsor rather than the opposition as was intended. Further research on this multifaceted emotion is suggested, because both the theoretical and lay definitions of the term hold implications for advertising theory and practice.
1 CHAPTER 1 INTRODUCTION In a world where our senses are cluttered with information on a daily basis, it takes more than a few words on a page to capture the attention of an audience. Whether selling a product or an idea, advertisements are ever ywhere. Even the most creative messages can often be overlooked, which makes standing out from the crowd a daunting task. In an attempt to get the public to take notice, several organizations have turned to shock appeals, which intentionally alarm or offend the audience by showing them images or text that they would not normally expect to see (Dahl et al. 2003). These appeals, which range from disgusting to sexually explicit images or words, spark controversy and capture attention so that even those people w ho disagree with the message itself will not soon forget it. We have seen these shock a ppeals in paid advertisements for consumer products, such as Benetton, as well as public service announcements (PSAs), like those for HIV and AIDS prevention, however the eff ects of these appeals are rarely measured. In the commercial arena, Benetton was the first company to delete its own product entirely from its advertising in favor of some thing different. It was in the late 1980s that the company decided instead of selling th e product of trendy cl othing, it would sell a philosophy that urged others to become more conscious of social issues. The company soon began to break social norms and pus h the moral envelope with their risky advertisements featuring images of dying AI DS patients and slain mafia members, among others. The advertisements won several aw ards for making consumers aware of social issues. However, the ads also led to majo r public outrage by organizations who believed
2 that the unconventional images shown merely represented a marketing tactic used by Benetton, and that the clothing company was exploiting some delicate issues. Furthermore, these organizations believed th at because it was a trendy clothing company that was introducing these images to the public, instead of a reputable non-profit organization, the issues that the images st ood for somehow lost their significance (Dahl et al. 2003; Kalish 1990, Tinic 1997). Tinic (1997) described these ads as Â“public service announcements with a sales pitchÂ” in her pa per dissecting the Benetton campaign (p. 11). Regardless of the ultimate goal of the shoc king and controversial campaign, Benetton has since seen profit increases and market gr owth, as well as increased international recognition. Public Service Announcements of all kinds have increased their attention-grabbing tactics as well, no doubt in an attempt to re ach the virtually un-phased younger audience. HIV/AIDS prevention ads, in particular, use shock advertising to awaken their audience to the reality of a potentially deadly virus. Their tactics range from angry words to fearful images, and anything else they can us e to break through th e clutter. Several researchers have found shocking tactics to be effective in capturing at tention, which is the first stage in information processing. (Dahl et al. 2003, Dillard et al. 1996, Rigby et al. 1989). Disgust appeals are a specific form of shock advertising or shock communication that is also currently being pract iced in the United States, espe cially to persuade people to choose sides on controversial public issues. Ac tivist organizations such as People for the Ethical Treatment of Animals (P ETA), as well as pro-life grou ps all over the country are
3 two examples of organizations that use this approach to discredit the oppositionÂ’s point of view and bring audiences over to their side of an issue. PETA is the largest animal rights orga nization in the world and has been going strong for more than 25 years. The organiza tion firmly stands behind the absolute belief that animals are not ours to wear, eat, expe riment on or use for entertainment purposes. Most of PETA's advertising and publicity effo rts speak out against fast food restaurants, scientific laboratories, the fu r trade, and the entertainment industry in particular. Their campaigns feature photos and videos of animal s being tortured and butchered in order to suit the needs of humans. Since its inception in 1981, PETA has used an abrasive, in-your-face media campaign to change the daily operations of some major United States corporations. Some of the milestones have included ending harsh animal experimentation at Columbia University, convincing cosmetics companies li ke Avon and Revlon to cease testing their products on animals, pressuring fast food restau rants like McDonalds and Burger King to only buy meats from factory farms that take better care of the animals, and getting clothing stores like the Gap and Forever 21 to switch from selling real fur to fake fur ("PETA History" 2005). All of this work toward a better life for animals has not gone unnoticed and PETA now boasts a membership of over 800,000 people, including many celebrities. However, there are also many outspoken people who do not support the extreme views and forceful advertising tactics implemented by the group. Therefore, PETA remains a controversial organization whose strong efforts and disgus ting campaigns have encouraged, as well as turned off, several people at the same time.
4 The landmark Roe vs. Wade United States Supreme Court case of 1973 overturned all state laws banning abortion, due to the f act that these laws violated a womanÂ’s constitutional right to privacy. Ever since this decision was made, abortion has become a highly controversial issue that has split the country into two categories: pro-life and prochoice. Each side uses what tactics they can to garner support fo r their own side, with pro-life groups in particular, using aggre ssive, shocking, and disgusting images in many of their campaigns. One example of these shock tactics can be seen in a recent anti-abortion campaign created by the Center for Bio-Ethical Reform, where disgusting images of aborted fetuses were blown up and pasted in full color to the si des of large trucks. The trucks were then driven through towns all over the country in order to shoc k people on the streets into seeing what an abortion rea lly looks like. The campai gn was called Â“The Reproductive Â‘ChoiceÂ’ CampaignÂ” and it was created as a way to get around common media gatekeepers, such as newspaper and magazi ne editors, who would not allow for these graphic images to appear in their publications. The reason for the use of such disgusting and disturbing images is explained on the homepage of the organizationÂ’s website. It reads: Â“Pictures make it impossible for anyone with a shred of intellectual honesty to maintain the pretense that Â‘it's not a babyÂ’ a nd Â‘abortion is not an act of violence.Â’Â… We want America to be very bothered by this ma ssive and hidden injustice, even if we are hated in the process" (Cunningham 2005). The effectiveness of the campaign so far remains unclear. According to several newspapers that have covered the story, th e images were not effective in swaying viewpoints. As a matter of fact, they claim that the trucks only sparked anger and hate
5 toward the organization. One newspaper ar ticle quoted a local woman as saying that making abortion graphic is not going to change pe ople's minds In fact, it's just going to make them mad (Syed 2003). According to the website, the campaign has met with both positive and negative results in the form of emails from people who saw the trucks. One woman changed her stance on the issue of abortion after seeing the graphic images . She wrote, Â“Your photos changed me from pro-choice to pro-life. Mo re has got to be done to educate the public on the true horrors of abortion. Please c ontinue to spread your word any way you can. This is truly a cause that cannot be ignored. I think it s hould be mandatory for every woman who is to under go this surgery to see photos like these so they to would understand exactly what is taking placeÂ” (Cunningham 2005). Overall the campaign has not yet been seen by many as a positive way to get a message across. However, that does not necessarily mean it has been ineffective. According to the website a few people did chan ge their minds due to the campaign and its shocking images. The organization is not rea dy to give up. The campaign continues to run currently and planes carrying banners with graphic images have been added as well. These planes have been circling over college campuses acr oss the US, including the University of Florida, as recently as November 2005. It is evident that organizations, such as PETA and the Center for Bioethical Reform, use disgusting images in their campaigns in order to persuade. Further, studies have found that the emotion of disgust can be effective in pro ducing the reaction of turning away from the stimuli. This can be negative for ads trying to persuade people to
6 participate in something, like eating certain f oods (Shimp & Stuart 2004). On the other hand, disgust can work to a groupÂ’s advantage when they are trying to persuade people not to do something by using the emoti on to discredit the opposition (Nabi 1998). However, in order to determine the effectiveness of disgust, it is critical to understand its definition in the context of a st udy. Disgust is a multifaceted emotion and it can produce different results when used in the theoretical sense, than when the lay meaning is used. The lay meaning ofte n includes anger, which can hinder its effectiveness on viewers (Nabi 2002). Ther efore, knowing how disgusting stimuli is interpreted by audience members is very important. When thinking of pro-life PSAs that show disgusting images of aborted fetuses and pro-vegetarian PSAs that show the gruesome way that animals are sl aughtered for food, it makes sense that the disgusting images may turn an audience away from the scenes themselves, but they may also turn them away from participating in the behaviors that the opposition is suggesting (such as having an abortion or eating meat) as well. The following study presented subjects with one of four shocking advertisements, two with disgusting images a nd two with neutral images. Then the subjectsÂ’ emotional reactions were measured in three different ways. First, through a Self-Assessment Manikin, then through a semantic differentia l scale, and finally, through an open-ended question. One week later, the subjectsÂ’ emo tions were measured again, as well as their attitude and potential behavior changes, to see if they were different than the initial responses.
7 The main goal was to find out how the emotion of disgust works as a tool for persuasion, by trying to answer the following two research questions: Research Question 1Are disgusting imag es an effective persuasion tool for public service announcements that argue agai nst one side of a controversial public issue? Research Question 2Do the disgusting im ages produce a reaction in the subjects that is more in line with the theo retical or lay meaning of disgust?
8 CHAPTER 2 REVIEW OF LITERATURE This section gives a genera l overview of the past eff ectiveness of common Public Service Announcement (PSA) appeals and in troduces the growing trend of shock advertising, along with how its images tend to capture the viewersÂ’ attention through the violation of social norms. Finally, it narrows its focus to disgust, which is also a currently practiced, but rarely analyzed form of shock appeal that is often used to discredit the opposition in two-sided public issues. The Effectiveness of Past Pub lic Service Announcement Appeals PSAs, whether through print, television or radio are created to persuade the audience to ultimately change a behavior, but ther e is much more to the process than that. In order to be persuaded to change a behavior , an overall attitude must first be changed. An attitude can be defined as a "general and enduring positive or negative feeling about some person, object or issue" (Vidal 1998). Si nce an overall attitude can be made up of several beliefs and is more enduring, it is mu ch more difficult to change than a single belief. An attitude also differs from a behavi or. In fact, a behavior is often the result of an attitude change. Therefore, the only wa y to get someone to effectively change his behavior is by first changing his attitude (Fishbein & Raven 1962). Changing attitudes through PSAs is extremely difficult, because unli ke commercial advertisements that often support widely held beliefs, the goal is to convince the audience to change its mind and subsequently its behavior toward a relevant issue.
9 During a review of the literature, Sl ater (1999) determined that health communication messages have often been di vided into distinct emotional appeal categories, such as informational, soci al-modeling, positive and fear (DeJong & Atkins 1995, Freimuth et al. 1990, Rimal 1997). In hi s analysis of 189 te levised drinking and driving PSAs he focused on all of these categor ies, as well as the additional category of empathy appeals, which were not typically id entified and studied for PSAs in the past. His goal was to find out how ofte n each of the appeals was used. Just as Slater had anticipated, informati onal appeals were the most commonly used type, accounting for almost half of the ads (48.1 %). This was consistent with the findings of Freimuth et al. (1990) in the AIDS context, where informational appeals accounted for a little more than half of the ads . However, Slater was surprised to find that positive appeals, as well as empathy appe als, were both used more often that he had initially assumed (19.6 % and 13.2 %, respective ly). From his research, he concluded that informational appeals generally dominate over st rong theory-based emotional appeals, which may be why so many PSAs are not effective. He suggested that future research focus separately on the effectiveness of each of the different appeal strategies. Several researchers since th en have done just that. Informational Appeals The most widely practiced appeal for PSAs is the informational appeal, which uses basic facts and information to deliver a me ssage that is meant to increase audience awareness and knowledge of a subject. In se veral studies of PSAs, strict informational appeals were often the standa rd that other appeals were measured against (Bagozzi & Moore 1994, Dahl et al. 2003, Lee & Davie 1997). As he analyzed many of these studies, Slater (1999) found that informationa l appeals seem to work best for creating
10 awareness for an unfamiliar issue. However, there is also a problem with informational appeals. The problem with this tactic wh en it comes to public service announcements is that the viewer can acknowle dge the existence of a risk in the abstract without acknowledging that he could pers onally be at risk (Coleman 1993). In other words, an emotional connection with the audience is often missing from strict informational appeals, and simply stated facts are not always enough to induce a needed behavior change in the person who is bei ng exposed to the advertisement. Social-Modeling Appeals Another common appeal is the social-modeling appeal, which combines educational information with an emotional tone that is usually positive. It is based on the social-modeling theory devised by Bandura ( 1986), which emphasizes the importance of observing and modeling the behaviors, attitudes, and emotional reactions of others. This technique models a desired behavior and provides inform ation on how the viewer can overcome obstacles in order to attain this be havior. It can also do the reverse and show an inappropriate behavior and people not wantin g to participate. Slater found that this type of appeal is more effective than a stri ctly informational app eal, but there have not been enough studies on it to determin e just how effective it is for PSAs. One study, however, uncovered a little bit a bout the effectiveness of the socialmodeling appeal in anti-smoking PSAs. In their copytest study, Pechmann and Reibling (1997) found that a Â“Refusal Skills Role Mode lÂ” message was very effective in reducing adolescentsÂ’ intention to smoke. The Â“Refusal Skills Role ModelÂ” message showed attractive, intelligent role models engaging in the desi red behavior of not smoking. The study used 167 PSAs from campaigns across th e country and had each of them viewed by approximately 50 seventh and tenth graders. The results showed that the subjects found
11 only a few message appeals to be effec tive for anti-smoking PSAs and social-modeling was one of them. Campaign costs were also assessed in this study, which led to the final conclusion that a social-modeling appeal is one of the most cost effective and successful ways to persuade adolescents not to smoke. Positive Appeals Positive appeals are emotionally based appeals that send a message which is meant to evoke positive feelings within the viewer . Some research suggests that positive feelings may increase the likelihood that the viewer will comply, but a study done by Fishbein et al. (2002) found that in the case of anti-drug PSAs the opposite was true. This study examined the perceived effec tiveness of 30 anti-drug PSAs by 3,608 students in the fifth through twelfth grades. What they found was that the more positive an emotional response the viewer had to a message , the less likely he or she was to take it seriously or view it as effective, and that humorous PSAs were the least effective way to persuade. The study concluded that when it comes to anti-drug PSAs the negative consequences of using drugs prove to be much more effective. Another study conducted in Canada found that positive appeals were the worst choice when creating effective AIDS preven tion PSAs. One hundred and seventy-eight college adults with the average age of 22 wh ere chosen for this study, which tested three different message strategies: rational (informational), negative emotional and positive emotional. The results showed that the rati onal strategy actually worked best for this topic, because the straightforw ard facts elicited strong self-c oncern within the viewers. This is consistent with SlaterÂ’s (1999) findi ngs that informational appeals work best for issues that the audience may not already know a great deal about. The positive emotional appeal was criticized by the participants who felt they were being talked down to by the
12 announcer, suggesting that a positive appeal does not necessarily lead to a positive impact on AIDS preventative behavior among young adults (Marchand & Filiatrault 2001). Fear Appeals Fear appeals are another common strategy for PSAs. This technique provides a message that shows the negative consequences of engaging in an inappropriate behavior. It is an emotional appeal that scares the view er away from participating in something that could be harmful to him. There have been so many studies done on fear appeals that the results seem to be inconsistent and confusi ng (Austin et al. 1999). For instance, while a moderate fear appeal may lead to compliance, those that are too st rong or too weak often fail. Some researchers have found that when using low to moderate levels of this emotion, there is a direct relationship between fear and attitude change in the audience (Boster & Mongeau 1984, Rotfeld 1988). However, other researchers have found that if the fear appeal is too mild even in the s lightest way, people may not be motivated by it, and if it is too strong people may tune out th e message or deny that it is true (DeJong & Winsten, 1990). This research seems to point to moderate fear appeals as the best choice, however, some researchers cau tion against the use of them at all in public service announcements (DeJong & Winsten 1990, Slater 1999). Empathy Appeals Finally, empathy appeals are increasingly used in PSAs, but not often tested for their effectiveness. It was Slater (1999) who first recognized them as a separate group of PSAs emerging in the media. Accordi ng to Slater, this type of appeal falls between evoking fear and positive arousal. The fear is shown through a negative consequence happening to someone else, and the positive arousa l to help in some way is stirred up in the viewer.
13 According to Bagozzi and Moore (1994), PSA s are aimed at two distinct groups of people. Some are aimed at individuals who ar e in need of help or who may be vulnerable to certain problems, such as alcoholics or te enagers, respectively. The others are aimed at individuals who may be in a position to help the above groups of people by donating time or money. PSAs aimed at the latter gr oup are the ones that often use empathy appeals. In the first study ever done on the role of negative emotions and empathy in the decision to help someone within the contex t of PSAs, Bagozzi and Moore surveyed 143 undergraduate students to determine whether negative emotions lead to empathy, which in turn lead to the decision to help in an ti child abuse ads. The emotional ads, which were compared to strict rational appeals, were effective in producing negative emotions, empathy and the decision to help within the subjects. Bagozzi and Moore also determined from this study that Â“high im pact ads that evoke strong emotions and stimulate empathy could require fewer exposures, yet be successf ul in influencing attitude formation and decision making" (p.67). In other words, th ey could be very cost effective. In conclusion, the ads that produced a strong negati ve emotion did in fact cause the viewers to empathize and strengthened their decision to help. Following the efforts of Bagozzi and Moore, Chabet and Chabet (2003) engaged in a study that measured the effec tiveness of empathy appeals presented in a drama versus a lecture format in low and high self-relevance situations. Low versus high self-relevance conditions make up a basic tene t of the Elaboration Likeliho od Model, which is the most commonly used model in advertising to unde rstand attitude change. The ELM, which was created by Petty and Cacioppo (1986), sugg ests that viewers who are exposed to persuasive messages use a central rout e of processing under high self-relevance
14 conditions and a peripheral route when under low self-relevance conditions. Selfrelevance conditions describe how strong the viewerÂ’s motivation and interest is in processing the message, therefore the central ro ute, where motivation is higher, is thought to produce more enduring effects than the pe ripheral one, where motivation is lower. The researchers showed 130 college student s, aged 18 to 36 years, PSAs focusing on AIDS, which created high self-relevance and malaria, which created low selfrelevance. The study revealed that the format (drama vs. lecture) of the advertisement affects feelings of empathy significantly under low self-relevance situations, but not under high self-relevance ones. This findi ng was consistent with the Elaboration Likelihood Model, since the type of format used was considered a peripheral cue. According to the study, the format (or periph eral cue) that worked best under lowrelevance conditions was the drama fo rmat, as opposed to the lecture one. On the whole, the five types of PSAs di scussed above have been continuously studied and challenged by researchers (DeJong 1990, Grube 1993, Slater 1999). Many times this is because they just cannot seem to capture the target audienceÂ’s attention and stay in their memory. After all, messages can only be effective if people see them and pay attention to them. In addition to trying to change the viewer sÂ’ attitudes, anothe r major problem with PSAs is capturing and holding their attention. Often times the target group finds the messages to be boring, especially in relation to larger budget commerc ial advertisements. PSAs showing the harmful effects of underage drinking, for example, are commonly found to be uninteresting and useless by the targ et audience, particularly in comparison to commercial beer advertisements (Grube 1993) . If a PSA is not interesting enough to
15 capture the audienceÂ’s attention, it will also not be able to remain in their memory. Memory is important for PSAs, because the view er is not likely to make decisions during exposure, but later down the ro ad when needed (Keller 1987). There is a relationship between capturing the attention of the audi ence, storing ideas in their memory, and attitude and behavior change . So, if the PSA can immedi ately capture the audienceÂ’s attention and remain in their memory, attitude change may be possible. This is where shock appeals come in. Shock Advertisements and the Violation of Social Norms Shock advertisements are defined as a ppeals that intentionally, rather than unconsciously, alarm and offend the audience (Gustafson & Yssel 1994). The audience becomes offended through the process of norm violation, which often consists of breaking moral boundaries, laws or customs. In life, norm violation can appear when certain shared expect ations that groups develop through the process of social learni ng are infringed upon. These expectations grow out of ideas that most humans beings have come to know as acceptable and normal behaviors. These appropriat e behaviors are sometimes reinforced by laws and other times by commonly held moral standards. In advertising, content is also expected to follow rules of Â“decency, good ta ste, aesthetic propriety, and personal moral standardsÂ” (Dahl et al. 2003, p. 269). When advertising breaches these rules, the audienceÂ’s attention is captured and although outrage may ensue, the message has been successful in breaking through the co mmon media clutter. Although the amount of studies on the effectiveness of shock appeals is limited, there is a belief that shocking stimuli do in fact attract attention, in turn facilitating other elements of information processing. Dahl et al. (2003) created a preliminary model of
16 consumer reactions to shock appeals based on a past cognitive processing model created by McGuire in 1978 (Figure 1). This model suggests that after attenti on, shocking stimuli should Â“facilitate message comprehension and elaboration, en hance message retention and influence behaviorÂ” (p.269). Figure 1. Preliminary shock appeal based on a model by Dahl in 2003 The central surprise aspect of the model is the most important step in the process, because past findings have revealed that indi viduals engage in higher levels of thought when they are exposed to an unexpected advertisement or image compared to the expected ones (Pyszcynski & Greenburg, 1981). In addition, studies have shown that this element of surprise also aides in memory storage (Heckler & Ch ilders 1992, MacInnis et al. 1991). Shock appeals seem to have grown out of f ear appeals. The two are very similar, and a fear appeal can indeed be shocking as well. An early study conducted by Rigby et al. (1989) looked further into the shocking aspect of some fear appeals five months after one of the first HIV/AIDS awareness campaigns was launched in Australia. This campaign was commonly referred to as the Â“G rim ReaperÂ” campaign and it consisted of televised spots. During the study, a survey wa s given to random residents of Australia testing their memory of the ads and curre nt knowledge, attitudes and beliefs about
17 HIV/AIDS. It was then compared to a 1986 survey of South Australia that was conducted prior to the campaign. The resu lts initially showed that 93.5% of the respondents recalled seeing the campaign five months prior and 63.7% approved of its use of shocking tactics. However, the resu lts also showed that personal and social concern, as well as knowledge a bout AIDS did not increase. In other words, most people remembered the campaign itself, but were una ble to recall or become affected by the actual factual informa tion contained within. In two fear-related studies, Dillard et al. (1996) found th at there were many more affective outcomes of fear-based AIDS PSAs othe r than just being scar ed. The first study asked 180 university students to monitor th eir own emotions while being exposed to 31 AIDS prevention PSAs that used fear appeals. The results revealed that while 61% of the PSAs produced a significant incr ease in self-reported fear by the subjects, 97% produced changes in other affective responses as well, with surprise, anger and sadness increasing, and happiness and puzzlement decreasing. In the second study, 167 st udents were then exposed to the 12 most fear inducing PSAs fr om the first study a nd asked to write down their emotions. The researchers found that both fear and surprise we re potent factors for ensuring that the message was seen as effec tive by the viewers. These studies showed that both fear and shock were successful in getting messages acro ss in AIDS prevention PSAs. A qualitative study conducted by Engineer et al. (2003) surveyed young adults age 18 to 24 years in 16 focus groups in England and Whales. The purpose of the research was to determine young peopleÂ’s experiences of Â“crime, disorder and risk-taking in the night-time economy, and explore ways in which drinking patterns, attitudes to drinking
18 alcohol and the effects of bi nge drinking were related to these experiences" (p.70). A portion of the study discovered that many young adults associ ated PSAs with lectures, like those they would get from parents or t eachers; and this sort of approach would probably be ineffective for them. The partic ipants offered suggestions for the type of messages that would be likely to have more of an impact on them than what they have been used to seeing in the pa st. Most of the groups agreed that PSAs Â“with Â‘shock valueÂ’ were the best way of getting message s about risk acrossÂ” (p. 71). Specific comments reinforced the idea that anti-drinking messages should be Â“as hard-hitting and gory as possibl e, so that they would stic k in peopleÂ’s mindsÂ” (p. 71). One male explained: Â“Like a campaign with a body in a hospital w ith his face opened up from a pint of glassÂ…that would make you th ink, Â‘Oh, I wouldnÂ’t like to be hit with a glass.Â’ Nobody cares about numbers and thatÂ” (p. 72). A female participant mentioned that disgust, in particular, would be very effective in swaying young adults away from binge dr inking. She cited the example of using pictures to compare an alcohol icÂ’s liver with a h ealthy one. Â“With the disgusting liver, just put a big one up, catch peoples attention,Â” she said. Â“You donÂ’t look at it much if itÂ’s not disgustingÂ” (p. 72). This study also examined ideas for controlling binge drinking among young adults and in doing so it concluded that advertisi ng campaigns were less effective than policy interventions. However, if the messa ges were strong enough, communications that highlight unpleasant or shocking conseque nces might stick in their minds best. Two studies conducted by Dahl et al. (2003) measured th e effectiveness of shock appeals in advertisements for HIV/AIDS prevention by comparing them to the more
19 commonly used informational and fear appeal s. The first elaborate study placed 105 university participants aged 18 to 27 in a false waiting room and had them fill out a decoy pricing survey. While the subjects waited in the room for several mi nutes for the survey to begin, the real experiment t ook place. On the walls of the room were five posters: four decoy ones and a last one th at talked about HIV prev ention. There were three experimental conditions: one using an inform ational HIV prevention poster, one using a fear based appeal, and one using a shocki ng poster about HIV prevention. The shocking poster used a headline that re ad, Â“DonÂ’t be a fu**ing idiotÂ” and revealed a blurry image of a nude couple in an intimate embrace. Afte rwards, the studentsÂ’ r ecall of the posters was measured, as well as which of the posters attracted their attention the most. The study revealed that the shock appeal outperf ormed the other two appeals when it came to attention, recall and recognition. The study also reinforced the theory that shocking ads are effective because they violate social norm s, therefore heighteni ng the audienceÂ’s level of awareness. The study concluded that it is in fact possible for a shock appeal to have positive effects on the audience beyond the initial attention grabbing factor. The next study went a step further and measured the behavior of subjects after being exposed to shocking advertising content. This time 140 students participated in an almost identical study. However, after the subjects completed the paper survey, they were told that the company had been cleaning out its clubhous e and that they could take any leftover pamphlets and brochures they f ound from a table set up in the room. This was how the researchers measured behavior. The table consisted of brochures with information on several issues, including al cohol, drugs and HIV prevention. The researchers found that nearly half of the subj ects who were exposed to the shock and fear
20 conditions picked up HIV/AIDS related materi als, compared to a pproximately twenty percent in the information condition. The resear chers concluded that due to this behavior shocking advertising content, as well as f ear-based appeals were able to Â“motivate subjects to acknowledge the risk of AIDS acquisition and obtain information regarding safe sex behaviorÂ” (p. 267). The overall findings suggest that sh ocking content can be an effective communication strategy for HIV/AIDS preven tion messages. The researchers warned, however, that careful target audience selection, a good fit be tween the message topic and shock tactic, and a savvy creative execution also contribute to the success of the advertisement. Dahl et al. (2003) also separated shock a ppeals into seven distinct types to be examined for further study. Religious taboos ma ke inappropriate use of other cultures or religious beliefs or rituals, like a Benett on ad of a priest and a nun kissing. Sexual references include images of nudity and sugge stions of sexual acts. Profanity/obscenity appeals consist of swear words and obscene gest ures. Vulgarity refers to advertisements with crude or distasteful acts being performe d, such as nose picking or drinking from the toilet. Impropriety appeals violate social conventions for dress or manners, for example humorous nudity. Morally offensive appeals sh ow images of gratuitous violence or sex, exploitation of children, and other things th at seem morally inappropriate. Finally, disgusting images consist of references to or images of blood, body parts, fecal matter, and death and decay. Disgust As limited as the literature is on shock advertising, there are even less studies on the effectiveness of disgust appeals in partic ular. Fear is the mo st commonly researched
21 negative discrete emotion and the others, such as guilt, sadness, anger and disgust are often ignored. There are, however, several sc holarly papers on the definition of disgust a rather complex emotion. It is likely that disgust wa s one of the first emotions that emerged in the human race, due to the early biol ogical reaction of approach and avoidance (Izard 1977). Disgust is a negative discrete emotion that is theoretically defined as the result of taking in or being too close to some thing indigestible, either liter ally or metaphor ically speaking (Lazarus 1991). The emotion of disgust is sa id to cause feelings of unpleasantness and even nausea. Many researchers have broadene d their meanings to include a sociomoral element as well (Izard 1977, Rozin et al.1999). However, in everyday life disgust, anger and contempt are often experienced in the same situation. Disgust can even go as far as to encompass feelings of annoyance and irri tation and because of this general society sometimes lives by a very different idea of disgust in comparison to the theoretical definition. In a study by Nabi (2002), she questioned separate samples of college students about times when they felt angry, disguste d, revolted, or grossed out. Upon analyzing their essays in response to these terms, she found that in everyday life the term disgust encompasses more than just nausea, but also many elements of anger as well. She also found that the lay understanding of the word di sgust corresponds more closely with the theoretical meaning of anger or at least a mixture of anger and disgust, than it does with disgust alone. As a matter of fact, the subj ects in the study found disgust to be something that combines feelings of repulsion with feelings of irritation or annoyance.
22 The final conclusion was that Â“events r ecalled based on the trigger words disgust and disgusted tended to reflect primarily anger-related themes rather than classic disgustrelated themes whereas the co lloquial term grossed out evoke d the types of scenarios we expect to be associated with the theoretical meaning of dis gustÂ” (p. 701). This difference between disgust as a theoretical concept and the general everyday us age of the word is important for researchers to know before conducting a study on the subject. Since the meaning of disgust can encompass something as strong as feeling physically ill after witnessing a gruesome murder to feeling ir ritated by someone who eats food with their mouth opened, understanding what the audience be lieves to be the meaning of or feelings associated with the term dis gust is a critical component in creating persuasive messages that actually work. In another study, Nabi (1998) hypothesized that the emotion of disgust may be effective in two-sided refutational messages. For her study, she ha d 134 college students view two-sided refutational video messages ar guing in favor of animal experimentation in order to determine the effect of disgus t on attitude change when it is the dominant emotion elicited by the message. Nabi assume d that if disgust is the dominant emotion driving a message, then Â“its action tendency of turning away from or avoiding the disgust-eliciting material will encourage a re spondent to similarly turn away from, or reject, the position associated with that ne gative affectÂ” (p. 32). In other words, associating the emotion of disgust with the oppositionÂ’s position should cause the message receiver to reject the opponentÂ’s view point and in turn strengthen persuasion toward the view of the organi zation sponsoring the message.
23 Nabi first found that disgust can, in fact , be the single dominant emotion elicited by a persuasive message, and sec ond, that the level of disgus t present can influence the persuasive outcome of the message. Nabi also found that disgust had a negative linear relationship with attitude change, which went against her original hypothesis. In this particular study, a PSA was shown that presented two sides of the issue of using animals for experimentation. The PSA favored the use of animals to help cure di seases, but in contrast it showed disgusting images of animals being poked and prodded upon for the opposing argument. This caused respondents to have a negative feeli ng about animal testing even though the PSA was meant to persuade them that animal testing is safe and needed. Nabi believed that the negative relationshi p was due to the placement of the disgust in this case, and inferred that had the PSAs been arguing the opposite position that animal testing is wrong, the disgusting images would have resulted in a very positive linear relationship. Or if the PSAs had argued th e same position, but shown disgusting images of children dying from lack of medication, that also would have been effective. Nabi felt that more research needed to be done, but anticipated that new studies would find that Â“in two-sided issues, a disgust response will e nhance persuasion if associated with the oppositionÂ’s position within the two-sided persuasive messageÂ” (p. 33). Shimp and Stuart (2004) looked at disgust in the food arena and the role that it plays in the effectiveness of the advertisement toward purchase intentions. Their hypothesis was that showing raw cuts of meat in an attempt to convey the freshness of the food in an advertisement would prove to be ve ry ineffective in persuading the consumer to purchase the food. This hypothesis was suppor ted by the Pavlovian theory that if an
24 unconditioned stimulus (the uncooked meat) and a conditioned stimulus (the fast-food restaurant) are placed together and the unc onditioned stimulus (the uncooked meat) is seen unfavorably, then the conditioned s timulus (the fast-food restaurant) will be perceived in a negative light as well. Two nearly identical experiments were conducted. The only difference was that one experiment used raw beef and the other used raw chicken. Both experiments asked approximately 130 students aged 18 to 24 years to look at four 30-second advertisements, in which one of them was for an imaginar y fast-food restaurant named KirbyÂ’s. The KirbyÂ’s ad had two versions: one that showed a cooked and ready-to-eat meat sandwich and the other that added a quick glimpse of the meat before it was cooked. The real purpose of the study was hidden, so subjects were told instead that they were testing out a new method for tracking emotional reactions to advertisements. The subjects were asked to answer several questions regarding their fee lings, attitudes, a nd possible purchase intentions for each of the advertisements. The results of both studies confirmed that Â“f eelings of disgust lessened intentions to eat at KirbyÂ’s and mediated the effect of ad version on purchase intentionsÂ” (p.48). In other words, subjects were much more likely to be interested in eating the sandwich if the inherent animalness of the product was dis guised. If not, it was instead seen as disgusting and unappetizing. A follow-up study done by Shimp and Stuart (2004) looked more closely at the everyday meanings of disgust that Nabi previous ly addressed in her paper. In this study, 25 college students were asked to each pick out two examples of advertisements that they found to be disgusting by their own definiti on, and explain why. The results were then
25 classified into six categories according to their frequency of occurrence; however, the first three categories accounted for most of the responses from students. The most common category referred to wa s labeled as gross depictions, which by NabiÂ’s (2002) original findings are defined mo re by the theoretical meaning of disgust. An example of gross depictions that was gi ven by several subject s was a MikeÂ’s Hard Lemonade commercial, where a young man is s hown with a second smaller head attached to his shoulder. Students described it as Â“m utant-like,Â” Â“freakish,Â” and Â“grotesqueÂ” (p. 50). The next most common category referred to by subjects was labeled as indecent, sexually oriented, sexist , and sexually objectifying and encompassed anything deemed sexually inappropriate. An exam ple given by several subjects was the series of late night commercials for Â“Girls Gone Wild.Â” Subject s felt these commercials for videos of nude college students were disgusting in the way that they exploited women. The third most common category referred to was labeled as gross animal and human waste scenes, which consisted of gross de pictions of animal or human waste. An example of this category was a commercial for Budweiser showing a woman entering a portable toilet during an outdoor event. While she is inside, a male accidentally knocks the toilet over an embankment and the woman is heard screaming as the compartment she is in tumbles along the ground. One subject reac ted by saying, Â“I can only think of all the human waste in the portable to ilet and the smell. She is yelling, so some of it has to be getting in her mouthÂ” (p. 51). One more category to mention, which wa s actually the fourth most common, was disgusting political advertising, where negative political ads were also seen as disgusting.
26 One subject mentioned campaigns that show aborted fetuses as a way of demonstrating that the candidateÂ’s opponent wa s pro-choice. Another subject claimed that Â“all negative political advertising is disgusti ng because it is unfair" (p. 51). In conclusion, Shimp and Stuart found that disgust took on several different meanings to the subjects who were sampled. These results were Â“in support of NabiÂ’s (2002) demonstrationÂ…that disgust is a multif arious notion.Â” The results also showed Â“that the theoretical concept of disgust as an imal-based and predominantly food-related is too limiting and that many aspects of advertis ing are considered disgusting by laypeopleÂ” (p. 51). Based on some of the literature that wa s found on the emotion of disgust, the following two hypotheses were tested in this study: Hypothesis 1An image that correlates w ith the theoretical meaning of disgust will enhance persuasion when associated with the oppositionÂ’s position within a two-sided controversial public issue. Hypothesis 2An image that correlates mo re with the lay meaning of the word, which includes anger, may cripple persua sion and more likely enhance a transfer of negative feelings toward the or ganization that is showing the ad.
27 CHAPTER 3 METHOD Subjects and Setting A 2x2 between-subjects experimental design that manipulated the nature of the pictorial stimuli (disgusti ng/neutral images) and cont roversial issue (abortion/ vegetarianism) was used in orde r to test the effects of the emotion of disgust on audience emotional reaction, memory, attitude and beha vior within the realm of public service announcements (PSA) that argue against one side of a controversial two-sided public issue. The respondentsÂ’ initial reactions a nd feelings toward the advertisement were measured immediately after they were expos ed to the stimulus. Attitude, memory and behavior were then assessed one week later in a posttest. The issues of pro-life and vegetarianism were chosen due to their hi story of using shocking tactics in their campaigns in an attempt to persuade viewer s to change their viewpoints on these topics. In this study, 83 undergraduate students par ticipated for course credit. Subjects ranged in age from 18 to 23 years. The main experiment consisted of an on-line questionnaire, where each subject who logge d in was randomly shown one of four different public service announcements. The four experimental conditions included a disgusting pro-life ad, a disgus ting pro-vegetarian ad, a neutral pro-life ad and a neutral pro-vegetarian ad. Measures The methods used to gauge the subjectsÂ’ s hort and long-term reactions to the public service announcements were IzardÂ’s Differe ntial Emotions Scale (DES II) and the
28 advertising Self-Assessment Ma nikin (AdSAM), as well as open-ended questions. Both the DES II and AdSAM measurements ha ve been commonly used to understand emotions, and have produced valid and reli able results for several researchers (e.g., Morris (1996) use of the PAD, Stout, et al.Â’s (1998) application of the DES). However, they work in very different ways. AdSAM is a measure used to understand the emotional impact of an environmental stimulus in a visual, as opposed to verb al, way. AdSAM is based on the Russell and Mehrabian PAD model (1974), which assesse s feelings of pleasure, arousal and dominance within the viewer. In the PAD model, Â“pleasure is the hedonic assessment of stimuli, while arousal is the activity elicited by the stimuli and domina nce is the feelings created by the stimuli related to how many be havioral options a person interacting with the medium hasÂ” (Mehrabian, 1998). In othe r words, dominance is a measure of the amount of control that the subject feels. In the AdSAM measure, a character is shown along three different scales. For each scale, there are nine options to choose from , however, there are not nine characters. Some options fall between characters. Pleasure measures how positively or negatively the subject is feeling while being exposed to the stimuli. Arousal measures the level of intensity or strength of feelings that the subject has. Finally, dominance measures the degree of empowerment that the respondent f eels. Pleasure ranges on the scale from a happy face to a sad one. Arousal ranges from a picture that shows the ultimate level of stimulation to the ultimate level of calmness. And dominance ranges from a very small character representing lack of empowerment to a very larg e one representi ng the feeling of being completely in control (Figure 2).
29 When working with AdSAM, participants are strongly encouraged to rate their feelings about the item they are view ing, rather than the item itself. Figure 2. Advertising Self-Assessment Ma nikin (AdSAM) measure developed by Dr. John Morris. Retrieved with permission from www.adsam.com, August 2005. In his work, Izard (1991) c onsiders that there are ten basic universal emotions: interest, joy, surprise, sadness, anger, disgust, contempt, fear, shame and guilt. Each of these emotions is commonly set upon a five-poi nt Likert scale. Th e Izard scale is based upon the differential emotions theory, which em phasizes discrete emotions as distinct processes that influence cogniti on and action differently. The th eory also elaborates that these ten fundamental emotions make up the ma in motivational system for human beings.
30 Further, each fundamental emotion has pa rticular motivational functions and unique experiential properties. Also, each emotion leads to a different inner experience and has different effects on cognition and action. The two methods work together well, b ecause while the PAD model reveals the subjects overall feelings about the stimuli in a way that words could not, the DES II looks at the discrete categories of emotions that are experienced by the subjects as they view the stimuli. Stimuli To prepare the stimulus materials, web re search was first done to find disgusting images relevant to pro-life and pro-vegetarian campaigns, as well as neutral images for each topic. These images were judged to be disgusting or neutral by the researcher and in correspondence with IzardÂ’s writings on the emo tion. A total of ten images were selected from the internet and then shown to 20 graduate students through a PowerPoint presentation, in order to determine how the s ubjects felt about them. The subjects filled out a questionnaire that containe d nine emotions from IzardÂ’s scales: interest, happiness, surprise, sadness, anger, disgust, fearfulne ss, comfort, and guilt. The reason for these nine emotions was to determine which images were found to be the most disgusting, as well as to rule out other strong emotional reactions to the images, and be sure that disgust was the single dominant emotion present. Each of the nine emotions was placed on a five-point Likert scale, where one (1) meant subjects did not experien ce the emotion at all and five (5) meant they felt the emotion very strongly. They were asked to fill out one column for each of the ten images that they viewed. They were also asked to write down which of the nine emotions they felt was the dominant one experienced for each image.
31 The open-ended question asking what the subjects felt the dominant emotion was helped as a validation check to make sure that the emotion that each subject felt strongest about on the scale matched with what he felt was the dominant emotion conveyed by the image. An analysis of the means for each of the emotions felt while viewing each of the images was performed to determine the result s. Please refer to Appendix A for a sample of the pretest. The results turned out just as had been expected. The images that were most disgusting were indeed seen that way by the pa rticipants who viewed them. Further, the more neutral images were found to score very low for levels of disgust, anger, fear, sadness, surprise and guilt. A possible explanation for why the neutral images scored so low, rather than somewhere in the middle of th e scale, is that when contrasted with the other graphic and disgusting images shown s ubjects felt very emotionally unaffected by the neutral images in comparison. The data revealed that the most disgus ting image for pro-life was the image of a broken aborted fetus on an operating table w ith a mean of 4.67, as opposed to 4.61 and 3.61 for the disgusting images of the fetus in a trashcan and a fetus hand reaching out of a stomach, respectively. The former image, though it had the highest disgust level, was not chosen for the main experiment due to the fact that disgust was not actually the dominant emotion felt by the pa rticipants; sadness was with a mean of 4.78. Instead, the image of the fetus in the trashcan was chosen because it still had a high level of disgust and disgust was the dominant emotion fe lt by the students who viewed it. The data also showed that the most dis gusting image for pro-vegetarianism was the image of a cow being slaughtered with a mean of 4.17, as opposed to 3.89 and 3.06 for
32 the disgusting pig and duck im ages, respectively. All thr ee scored very low for the emotions of happiness and comfort, and high in the other discrete negative emotions like sadness and anger, just as IzardÂ’s scale predicts. For the disgusting cow and duck images, disgust was the dominant emotion be ing experienced, however, for the image of the pig sadness was felt most. Therefore, the disgusting image of the cow being slaughtered was chosen for the main experiment. The two neutral photos chosen were those that produced the least amount of disgust in the viewers. These images were of a cute baby pig, with a mean of 1.11, for provegetarianism and a fetus in the wom b, with a mean of 1.31, for pro-life. Table 1. Pretest results for four images chosen Emotion Dead Cow (Disgust) Fetus in Womb (Neutral) Fetus in Trash Can (Disgust) Cute Pig (Neutral) Interested 1.94 3.39 2.22 3.00 Happy 1.06 2.67 1.11 3.22 Surprised 3.44 2.67 4.39 1.78 Sad 3.44 1.33 4.33 1.11 Angry 2.78 1.28 3.56 1.11 Disgusted 4.17 1.31 4.61 1.11 Fearful 2.94 1.44 4.06 1.11 Comfortable 1.28 2.67 1.33 3.78 Guilty 2.22 1.22 2.72 1.11 N=20 After the pretest determined the single be st image to use for each of the four categories (pro-life disgust, pro-vegetarian disgust, pro-life neut ral, pro-vegetarian neutral), print ads were created to go with each image. The ads were original, but borrowed headlines and themes from pro-life and pro-vegetarian campaigns that have been used in the past. Although the images were different, the copy in the ads did not differ. The text used in the advertisements was meant to shock, no matter what image was used. The text of both pro-life ads read: Â“Life is a gift. Abortion is murder. Become Pro-Life.Â” The text of both pro-vegetarian ads read: Â“J ust a reminder. This is how your
33 food looked before it hit your plate. It could breathe, feel and had a heart. Just like youÂ…Go Vegetarian.Â” For each of the two i ssues, everything about the ads was the same, except for the image shown. Please s ee Appendix B for the neutral pro-life ad, Appendix C for the disgusting pro-life ad, A ppendix D for the neutral pro-vegetarian ad, and Appendix E for the disgusting pro-vegeta rian ad (Warning: They are graphic). Procedure The main experiment, which took place on-lin e, used AdSAM and DES II, as well as an open-ended question to de termine the participan tsÂ’ feelings about certain issues, as well as their reactions to the ads they viewed. Please see A ppendix F for a sample of the questionnaire. The researcher went to three large cla ssrooms and informed the subjects of the questionnaire. An informed consent sheet was handed out, which explained that the questionnaire may contain graphic images a nd that participation was not mandatory and at their own discretion. The s ubjects were told they would receive extra credit for the class even if they logged on, but did not co mplete the questionnaire due to the graphic nature of the images. They were given a dead line of which they had to complete the online questionnaire. The website that took them to the ques tionnaire was written in bold letters at the bottom of the informed consent sheet. When subjects went to the we bsite that was given in the hand-out, they had to first read the informed consent online again and then click on a link to show that they agreed to participate in the stud y. This link then took them to the main experiment. The first section of the questionnaire us ed AdSAM to determine how the subjects felt about three controversial public issues, which were the death penalty, vegetarianism and pro-life. The question about the death pe nalty was added so th at the subjects would
34 be less likely to anticipate what the experiment might be abou t before they participated. The answers to this question were not impor tant, but the responses to the other two questions were very important. These respons es were used to determine how the students initially felt about the issues, so that during th e posttest one week la ter, we could find out if their views had changed at all due to the advertisement that they viewed. For the next part of the questionnaire , the subjects viewed one of the four advertisements. They were then asked to answer an AdSAM question to determine how they felt about the ad that they viewed, as well as to write in their own words how they felt about the ad. This was done to comp are what the AdSAM revealed about their feelings to how they described the way they were feeling. The open-ended question was later made quantifiable by taking some of the most common key words and placing them in a spreadsheet where each subject who mentioned some fo rm of each word was given a one (1) and those subjects who did not men tion it were given a zero (0). The common key words that were made quantifiable were Â“guilty,Â” Â“defensive,Â” Â“disgusted,Â” Â“sickened,Â” Â“pushy,Â” Â“confused,Â” Â“compassiona te,Â” Â“disturbed,Â” and Â“angry.Â” Then means were run for these key words to see how often they came up from subjects and whether there was a significant difference be tween the neutral and disgusting ads when it came to any of these key words. The third part of the questi onnaire asked the subjects to rate how strongly they felt eight distinct emotions from IzardÂ’s book on a scale from one to five, where one (1) meant they did not feel the emotion at all a nd five (5) meant they felt the emotion very strongly. The eight emotions chosen were: disgust, anger, disdain, distaste, sadness,
35 surprise, sickness, and interest. The reason fo r these eight was that they all in some way, whether large or small, make up IzardÂ’s e xperience of the emotion of disgust. The final part of the ques tionnaire asked basic demogra phic questions, such as age and gender, in order to get an idea of who the participants were. After the questionnaire was completed and the subjects submitted their answers, they were immediately taken to a screen that gave a brief synopsis of the main purpose of the study, as well as additional resources, such as local counseling services, in case they continued to be troubled by the images they had seen. The summary of the study did not contain any information about a posttest or measuring memory, attitude or behavior changes among participants. One week later, a posttest was done to fo llow up with the partic ipants. Please see Appendix G for a sample of the posttest. The researcher went back to the three classes just once and asked for the subjects who t ook the questionnaire on-line to fill out a brief follow up paper questionnaire in the clas sroom. The posttest only contained six questions. The first question was an open-e nded question that asked what the subjects remembered about the advertisement they had viewed one week ago. The second question was also open-ended and asked them what their stance was on the issue that was addressed in the ad. The next three questi ons asked how often the subjects had thought about the issue in the past week, how much th eir attitude had changed regarding the issue in the past week, and how much their behavi or had changed regarding the issue in the past week. Each question was placed above a five -point Likert scale, similar to the one in the main experiment, where one (1) meant not at all and (5) meant all the time. Finally, subjects were asked an open-ended questi on to elaborate on how their behavior had changed in the past week, if it indeed had.
36 CHAPTER 4 RESULTS Characteristics of the Sample Main Experiment For the main experiment, the sample turned out to be overwhelm ingly female. Of the 83 subjects who participated in the ma in study, only 13% of them were male. The average age was 21 years. The sample was split up almost evenly between those who saw the pro-life and the pro-vegetarian adver tisements, as well as between those who saw the neutral and disgusting images within each group. For example, 40 subjects viewed the pro-life ads, with 20 seei ng the neutral image and 20 seeing the disgusting image. Further, 43 subjects viewed the pro-vegetarian ads, with 20 seeing the neutral image and 23 seeing the disgusting image. According to the advertising Self-Assessm ent Manikin (AdSAM) results, subjects did not display any extreme feeli ngs or emotions about the issu es before viewing the ads. Since the AdSAM scale went from one (1) to nine (9), a score of 4.5 fell right in the middle of the scale. When it came to the issue of pro-life, the mean for all the participants was 5.49 for the feeling of pleasure, which showed that th e subjects in general fe lt slightly unpleasant about the issue. For arousal, the mean was 3.48, which showed that subjects were aroused, but not an extreme amount. For dominance, the mean was 4.79, which meant subjects felt slightly dominant about the issu e. The AdSAM Perceptual Map for arousal
37 and pleasure showed that overall the subjec tsÂ’ feelings about the issue of pro-life encompassed emotions, such as Â“cynical,Â” Â“anxious,Â” Â“suspicious,Â” and Â“startled.Â” When it came to the issue of pro-vegetarian ism, the mean for all the participants was 4.05 for the feeling of pleasure, which s howed that the subject s generally had good feelings about the issue. For arousal, the mean was 6.68, which showed that the subjects for the most part were only minimally arous ed by the thought of vegetarianism. For dominance, the mean was 6.75, which means that the subjects felt very much in control of the issue of vegetarianism. The AdSA M Perceptual Map for arousal and pleasure showed that the subjectsÂ’ feelings about the issue of pro-vegetarianism encompassed emotions, such as Â“modest,Â” Â“nonchalant,Â” and Â“serene.Â” As the AdSAM results reveal, the subject s felt very differently about the two issues. The issue of pro-vegeta rianism did not have as strong of an initial effect on them as did the issue of pro-life. Since the issues used were so different, it was almost as if two separate experiments followed (Figure 3). Posttest The sample for the posttest was not nearly as large as the original 83 subjects who participated in the main questionnaire. As a matter of fact, less than half of them completed the follow up questionnaire in the classroom. A total of 35 subjects participated in the posttest, of which four of them were male and the rest were female. The average age was again approximately 21 years. The responses were split up fairly evenly between the two issues, with 15 subjects from the pro-life group and 20 subjects from the pro-vegetarianism group participating. Within each of the two groups, nine subjects fr om the neutral pro-life group, six subjects
38 from the disgust pro-life group, six subjects from the neutral provegetarian group, and 14 subjects from the disgust provegetarian group took the posttest. Figure 3. Advertisement Self-Assessment Mani kin perceptual map: pleasure and arousal
39 Emotional Response to the Advertisement The participantsÂ’ emotional response to the advertisements was measured in three separate ways through AdSAM, an open-ended question that asked how they felt, and by Izard's Differential Emotions Scale (DES II). AdSAM measured their emotions using thr ee scales that showed several different characters and nine levels underneath them to choose from. The first scale rated pleasure, where one (1) meant that the subject felt very happy and nine (9) m eant the subject felt very sad. The second scale rated arousal, where one (1) meant the subject felt highly stimulated and nine (9) meant the subject felt very calm. The third scale rated dominance, where one (1) meant the subject felt a strong lack of control and nine (9) meant he or she felt a strong level of empowerment. The open-ended question simply asked the s ubjects to state how they felt about the advertisement after they viewed it. Some of the keys word s and phrases that resulted from this question had to do with experienci ng the emotion of disgust, as well as the feeling of being physically sick to oneÂ’s stomach. The DES II measure placed each of the eight emotions that encompass disgust on five-point Likert scales. These emotions were : disgust, anger, disdain, distaste, sadness, surprise, sickness, and interest. The scales we re set up so that one (1) meant the subjects did not feel the emotion at all and five (5 ) meant the subjects felt the emotion very strongly. Since the two issues were so different, th e results cannot be ad equately compared in this paper. The results are therefore wr itten separately, almost as if two separate experiments had taken place.
40 Research Question One Are disgusting images an effective pe rsuasion tool fo r public service announcements that argue one side of a controversial public issue? For this study, effectiveness of the ad was determined by the subjectsÂ’ initial reaction to the ad, as well as their memory, attitude and behavior one week later. The subjectsÂ’ immediate reaction to the ad was assessed through the use of AdSAM and an open-ended question, both asking how the ad made them feel. Pro-Life When it came to the AdSAM results for the pro-life ads, the neutral ad was significantly different from the disgusting ad for all three dimensions (Table 2). For pleasure, the mean for the neutral ad was 7.10 and for the disgusting ad it was 8.35. This means that those who viewed the disgusting image felt a significant amount less pleasure than those who viewed the neutral image. The total mean for both ads together was 7.73 out of 9.00. Since one (1) meant the subject was feeling very happy and nine (9) meant the subject was very sad, a total mean of 7.73 shows us that subjects found both of the ads to be much less than pleasurable. This ma kes sense because of the fact that the words in both ads made the same shocking and accusa tory statement that abortion is murder. The significant difference between the mean s for neutral and disgusting images proves that it was only the image, and no other f actor or extraneous variable, which was responsible for the different re actions. The disgusting image really did make people feel much more negative than the neutral image did. For arousal, there was also statistica l significance between the neutral and disgusting ad. The mean score for the neut ral ad was 3.95 and for the disgusting ad was 2.20, where one (1) meant the person was extremely aroused by the ad and nine (9) meant
41 they were not at all bothered or interested. Results showed that both of the ads stimulated the subjects because they fell below the me dian score of 4.50, but the disgusting one aroused them a significant amount more. For dominance, there was also statistical significance betwee n the neutral and disgusting ad. On the scale, one (1) meant th e subject felt a strong lack of control and nine (9) meant the subject felt extremely empowered. The mean for the neutral ad was 5.80 and the mean for the disgusting ad was 3.60, which shows that the students felt more dominant when viewing the neutral ad and le ss in control when viewing the disgusting ad . Table 2. Pleasure, arousal and dominance results for pro-life advertisements N M SD SS df MS F Sig. Pleasure Neutral Disgust Total 20 20 40 7.10 8.35 7.73 1.55 1.27 1.54 Between Groups Within Groups Total 15.63 76.35 91.98 1 38 39 15.63 2.01 7.78 .01 Arousal Neutral Disgust Total 20 20 40 3.95 2.20 3.08 1.73 1.15 1.70 Between Groups Within Groups Total 30.63 82.15 112.78 1 38 39 30.63 2.16 14.17 .00 Dominance Neutral Disgust Total 20 20 40 5.80 3.60 4.70 2.35 2.14 2.48 Between Groups Within Groups Total 48.40 192.00 240.40 1 38 39 48.40 5.05 9.58 .00 N=40; p< .05 These results lead us to believe that bot h the disgusting and neutral images when coupled with shocking words were effective in eliciting a definite emotional reaction from the subjects. The neutral ad resulted in a strong negative reac tion for pleasure, but only a slight level of positive arousal and dominance. This meant that subjects felt saddened by the ad, but still fe lt in control of the situation and not incredibly stimulated by it. On the other hand, the disgusting ad re sulted in a very strong negative emotional response from the subjects for pleasure, a very strong positive level of arousal, and still a fairly strong lack of control from the subjects . This meant that subjects felt saddened by the ad. They also felt highly stimulated by the ad, but at the same time not confident that
42 they could do anything about it because it was out of their control. We can infer from these strong reactions that the disgusting ad, in particular, wa s very effective at capturing the subjectsÂ’ attention by cr eating a strong negative emotional response within them. However, the disgusting ad may not have been enough for them to change their behavior, as they felt that they did not ha ve much control over the issue. For the pro-life ads, the open-ended react ions from the subjects revealed common responses, such as Â“disgusted,Â” Â“physically i ll,Â” and Â“disturbed.Â” However, statistical significance was found for one of the key words when it came to the differences between reactions to the neutral ad as opposed to the disgusting ad. Seven subjects out of 20 in the neutral condition stated that they felt th at the person or orga nization behind the ad was trying to push its viewpoints unfairly upon them. However, not a single participant in the disgusting condition made any such comment (Table 3). The open-ended responses ranged from zero (0) for subjects who did not mention any form of the key word to one (1) for s ubjects who did mention some form of the key word. From these open-ended responses we can infer that while both the neutral and disgusting ad captured the subjectsÂ’ attenti on by producing strong nega tive reactions, the neutral ad also caused subject s to feel as if it was pus hing its views upon them, which may have caused them to become angry, in turn hindering persuasi on. Since the image was neutral, the harsh and accusatory wo rds must have been what produced the significant difference. These words had much more of an impact when coupled with a neutral image than they did when put with a disgusting one, in turn making the entire ad less effective. This is rein forced through the previous AdSA M results, where arousal was significantly higher for the neutral image than the disgusting one.
43 Table 3. Open-ended results for Â“pus hyÂ” for pro-life advertisements N M SD SS df MS F Sig. Pushy Neutral Disgust Total 20 20 40 .35 .00 .18 .49 .00 .39 Between Groups Within Groups Total 1.23 4.55 5.78 1 38 39 1.23 .12 10.23 .00 N=40; p< .05 The posttest results for the pro-life ads reve aled that all of th e subjects who took it recalled the issue and image that was present in the ad they viewed. The subjects were also asked in the posttest to answer three que stions about their fee lings on a five-point Likert scale where one (1) meant not at all a nd five (5) meant all of the time. The three questions asked the subjects how often they t hought about the issue conveyed in the ad over the past week, as well as how much thei r attitude and behavior regarding the issue had changed during that time. There was not a significant difference between the neutral and disgusting group for these three questions (T able 4). However, the results show that more subjects in the neutra l condition remembered the ad, as well as felt they had changed their behavior due to the ad. On the other hand, more subjects in the disgust condition felt that their attitude had changed in the week following the ad. Table 4. Posttest results for pro-life advertisements N M SD SS df MS F Sig. Memory Neutral Disgust Total 9 6 15 2.11 2.00 2.07 .93 1.26 1.03 Between Groups Within Groups Total .04 14.89 14.93 1 13 14 .04 1.15 .04 .85 Attitude Neutral Disgust Total 9 6 15 1.22 1.50 1.33 .83 1.22 .98 Between Groups Within Groups Total .28 13.06 13.33 1 13 14 .28 1.00 .28 .61 Behavior Neutral Disgust Total 9 6 15 1.56 1.51 1.54 .88 1.22 .99 Between Groups Within Groups Total .01 13.72 13.73 1 13 14 .01 1.06 .01 .92 N=15; p< .05 An open-ended question asking for an exam ple of how the subjectsÂ’ behavior had changed elicited only two responses from the 15 participants of the posttest. One response came from a female who saw the neut ral pro-life ad. She wrote that she would now be more likely to tune into political debates to hear wh at a candidateÂ’s views on this
44 issue are. The other response came from a fe male who saw the disgusting pro-life ad. She wrote, Â“I had a conversation with my boyf riend regarding my fee lings on abortion.Â” These responses showed that the ads aff ected these two subjects enough that they changed their behavior. The disgusting image for pro-life was e ffective in eliciting a strong emotional response from the viewers. It captured their at tention, which is the first step in cognitive processing. Also, the overall ad itself did not cause the subjects to feel as if its message was being pushed on them, as the neutral ad did. However, changes in attitude and behavior for the subjects who viewed the disgus ting ad were not signifi cant in this study. Therefore, we can say from this study that the disgusting image for pro-life was only effective in the fist step of cognitive processi ng. However, this important step could lead to attitude and behavior change in some subjects. Pro-Vegetarian When it came to the AdSAM results for the pro-vegetarian ads, the neutral ad was significantly different than the disgusting ad for the dimensions of pleasure and dominance, but not for arousal (Table 5). For pleasure, the mean for the neutral ad was 6.50 and for the disgusting ad it was 8.17. Th is means that subject s found the disgusting ad to be much less pleasurable than the ne utral ad, however, both groups of subjects felt sad due to the fact that both ads had negative words in them. When it came to arousal there was no st atistical significance between the neutral and disgusting ads. The mean for both ads together was 3.72, which means that overall both groups of subjects were slig htly aroused by the ads that they saw, however, there was not a large enough difference between the two for it to be significant.
45 When it came to dominance for the pro-vegetarianism ads, there was a significant difference between the neutral and disgusting ads. On the scale, one (1) meant the subjects felt a strong lack of control and nine (9) meant the subject felt extremely empowered. The mean for the neutral ad wa s 5.95 and for the disgusting ad it was 3.35. This means that those who saw the neutral ad felt a significant amount more empowered than those who viewed the disgusting ad. Table 5. Pleasure, arousal and dominance resu lts for pro-vegetarian advertisements N M SD SS df MS F Sig. Pleasure Neutral Disgust Total 20 23 43 6.50 8.17 7.40 2.07 1.07 1.80 Between Groups Within Groups Total 29.98 106.30 136.28 1 41 42 29.98 2.59 11.56 .00 Arousal Neutral Disgust Total 20 23 43 3.90 3.57 3.72 1.89 2.13 2.00 Between Groups Within Groups Total 1.20 167.45 168.65 1 41 42 1.20 4.08 .29 .59 Dominance Neutral Disgust Total 20 23 43 5.95 3.35 4.56 2.11 1.72 2.30 Between Groups Within Groups Total 72.44 150.17 222.61 1 41 42 72.44 3.66 19.78 .00 N=43; p< .05 These results lead us to believe that both the neutral and disgusting ads when coupled with shocking words were effective in eliciting a defin ite emotional reaction from the subjects. The neutral ad revealed a strong negative reaction for pleasure, a slight bit of arousal, and a st rong feeling of empowerment from the participants. This means that subjects felt saddened by the ad, as well as in control of the situation, but not really compelled to do anything about it. Further the disgusting ad was effective at producing a very strong negative emotional response from the subjects when it came to pleasure, but not so much when it came to arousal and dominance. This means that the subjects felt very saddened by the ad. However, they were not very stimulated by the ad and they also felt like they did not have a lot of control. We can infer from these differences that the disgusting ad was bette r at capturing the s ubjectsÂ’ attention by creating a significantly stronger negative emo tional reaction in them. However, the
46 disgusting ad may not have been enough for th em to change their attitude or behavior toward the issue, as they di d not feel highly aroused by what they saw and they did not feel that they had much cont rol over the issue either. When it came to the open-ended questions for the pro-vegetarianism ads, some common responses were Â“sad,Â” and Â“defensive .Â” Â“DisgustedÂ” was mentioned often for the disgusting ad, but not once for the neutral one . Overall, the neutra l ad did not reveal any extreme feelings from the group. Howe ver, when it came to the disgusting image, several subjects became very defensive about wh at was conveyed in th e ad. As a matter of fact, eight subjects out of 23, roughly 35%, used the open-ended question to convey that there is nothing wrong with eating meat and that no one will be able to convince them otherwise. While there was not a si gnificant difference be tween the levels of defensiveness displayed from subjects in both conditions, it is important to note that more subjects became defensive in the disgust c ondition than the neutra l one, with means of .35 and .20, respectively (Table 6). The open-ended responses ranged from zero (0) for subjects who did not mention any form of the key word to one (1) for subjects who did mention it. From the openended questions we can infer that the disgus ting pro-life ad likely captured the attention of the students better than the neutral ad did, because it produced a significantly stronger negative emotional response. However, the neut ral ad elicited a milder reaction from the subjects, which may actually have been more effective. The disgusting ad, in contrast, caused several subjects to become defensive a nd quick to convey that they would not be changing their minds. In other words, the s ubjects may have put up a wall that did not allow them to be persuaded by what they saw in the disgusting pro-vegetarian ad.
47 Table 6. Open-ended results for Â“defensi veÂ” for pro-vegetarian advertisements N M SD SS df MS F Sig. Defensive Neutral Disgust Total 20 23 43 .20 .35 .28 .41 .49 .46 Between Groups Within Groups Total .23 8.42 8.65 1 41 42 .23 .21 1.14 .29 N=40; p< .05 The posttest results indicated no significan t differences between the neutral and disgusting pro-vegetarian groups (Table 7). However, they did reveal that while the disgusting ad was remembered more, the neutra l ad was more effective at attitude and behavior change within the viewers. This mean s that according to the posttest results, it is possible that while a disgusting ad caused the subjects to th ink about the issue more, it did not make them more likely to change their attitude or behavior. However, the neutral ad did. Table 7. Posttest results for pr o-vegetarian advertisements N M SD SS df MS F Sig. Memory Neutral Disgust Total 6 14 20 1.17 1.79 1.60 .41 .89 .82 Between Groups Within Groups Total 1.61 11.19 12.80 1 18 19 1.61 .62 2.59 .13 Attitude Neutral Disgust Total 6 14 20 1.67 1.43 1.50 1.21 1.09 1.10 Between Groups Within Groups Total .34 22.76 23.00 1 18 19 .24 1.27 .19 .67 Behavior Neutral Disgust Total 6 14 20 1.67 1.36 1.45 1.21 .93 1.00 Between Groups Within Groups Total .40 18.55 19.00 1 18 19 .40 1.03 .39 .54 N=20; p< .05 An open-ended question asking for an exam ple of how the subjectsÂ’ behavior had changed elicited only two responses from th e 20 who took the posttest. Both responses came from females who saw the neutral pro-ve getarianism ad. One wrote that she had eaten less meat in the past week and the other said she had ea ten more chicken in the past week. There were no open-ended behavioral responses for the disgust condition, even though more than twice as many subjects from that group participated in the posttest. The disgusting image for pro-vegetarianism caused a significantly stronger reaction than the neutral image did, therefore we can in fer that it was more effective at capturing
48 the viewersÂ’ attention. However, the high level of defensiveness that was felt by the subjects who viewed the disgusting image may have hindered its overa ll effectiveness. This can be seen in the posttest results, wh ere subjects who viewed the neutral ad proved to have slightly higher levels of attitude and behavior chan ge than those who viewed the disgusting ad. From this study, we can say th at the disgusting ad was only effective at capturing the attention of the subjects. However, the strong defensiveness they felt may have hindered them from reaching the levels of attitude and behavior change in the cognitive processing model. Research Question Two Do the disgusting images produce a reaction in the students that is more in line with the theoretical or lay meaning of disgust? When it came purely to the emotion of disgus t, it was measured in two ways. First through an open-ended question and then through a five-point Likert scale where five (5) represented the strongest fee ling of the emotion and one (1) represented the weakest feeling of the emotion. Each person who mentioned some form of the word Â“disgustÂ” or Â“sickenedÂ” in response to the open-ended question was given one point and those who did not were given zero. Then the results were compared to the specific emotions of disgust and being sickened represented through the Izard scale. Anger was also assessed to determine if the definition of disgust was si milar to that of the lay meaning versus the theoretical meaning. The openended questions revealed a lo t, and in a sense could be considered more valuable, because of the motiv ation that the subjects needed to type out their feelings, as opposed to just checking off a preset answer.
49 Pro-Life For the subjects who answered in the openended question that th ey were disgusted by the pro-life ads in the ne utral and disgusting categories, there was not a significant difference (Table 8). The word Â“disgustÂ” was not mentioned often for either category. However, three times as many subjects took the time to write down that they were disgusted by the disgusting image than did for the neutral image with means of .30 and .10, respectively. The difference in responses to the open-ended question may not have been statistically significant, but it is clear that more subjec ts responded that they were disgusted by the disgusting image than the neutral one. For those subjects who answered that they were Â“sickenedÂ” or made Â“physically illÂ” by the pro-life ads in the neutral and disgusting categor ies, there was a significant difference. While not one single person mentioned being made physically ill by the shocking words and neutral image, there was a mean of .25 for the disgusting ads, which meant a significant amount more subjects who viewed the disgusting pro-life ad were sickened by the image they saw. Twenty-fiv e percent of subjects in the disgust condition felt sick, while none in the neutral condition did. Since there was a significant difference wh en it came to being sickened, we can infer that the disgusting image for pro-life may have been more in line with the theoretical definition of disgust, which pr oduces feelings of being physically ill. When it came to the DES II scale for disgust for the pro-life ads there was a significant difference between the neutral and di sgusting ad. The mean for the neutral ad was a 3.15, which was slightly above average. However, the mean for the disgusting ad was 4.70 out of 5.00, which meant that the subjec ts who viewed it felt close to the highest
50 level of disgust that is possibl e. In other words, they were extremely disgusted by what they saw when it came to the disgusting pro-lif e image of an aborted fetus in a trash bag. When it came to being sickened by the ads according to the Izard scale, there was also a significant difference between the neutral and disgusting images. The mean for the neutral ad was just below average at 2.50. Ho wever, the mean for the disgusting ad was much higher with a mean of 3.95 out of 5.00. According to the DES II results, subjects who viewed the disgusting image were far more disgusted and sickened th an those who did not. Furthe r, it was the disgusting image alone that produced these feelings, sin ce the copy was the same in both ads. Table 8. Levels of disgust for pro-life a dvertisements by open-ended and Izard scale N M SD SS df MS F Sig. Open-Ended Neutral Disgusted Disgust Total 20 20 40 .10 .30 .20 .31 .47 .41 Between Groups Within Groups Total .40 6.00 6.40 1 38 39 .40 .16 2.53 .12 Open-Ended Neutral Sickened Disgust Total 20 20 40 .00 .25 .13 .00 .44 .34 Between Groups Within Groups Total .63 3.75 4.38 1 38 39 .63 .10 6.33 .02 DES II Neutral Disgusted Disgust Total 20 20 40 3.15 4.70 3.93 1.27 .73 1.29 Between Groups Within Groups Total 24.03 40.75 64.78 1 38 39 24.03 1.07 22.40 .00 DES II Neutral Sickened Disgust Total 20 20 40 2.50 3.95 3.23 1.32 1.40 1.53 Between Groups Within Groups Total 21.03 69.95 90.98 1 38 39 21.03 1.84 11.42 .00 N=40; p< .05 The results lead us to believe that when s ubjects were given the ability to rate the level of disgust and sickness th ey were feeling, they had si gnificantly stronger feelings toward the disgusting ad for both emotions. However, when it came to giving subjects the opportunity to write down a nything about how they were feeling; several of them were motivated to write down how sickened th ey were by the disgusti ng ad. In contrast, a significant number of subjects were not motivated to write down that they were disgusted by what they saw. Therefore, it is clear that a large am ount of sickness was felt
51 by subjects who saw the disgusting ad, maki ng the feelings produced by the disgusting image itself, fall more in line with the theoretical meaning of the word. Pro-Vegetarian For the open-ended question, subjects who wrote down that they were Â“disgustedÂ” by the pro-vegetarian ads in the neutral a nd disgusting categories, there was a very significant difference (Table 9). Not one subj ect mentioned being disgusted at all when it came to the neutral ad, even though the copy in the ad was meant to e voke a bit of disgust and shock. However, the mean for the disgusting ad was .52, which meant that more than half of the subjects who viewed the disgusti ng image for the pro-vegetarian ad were motivated enough to write in their own word s how disgusting they found it to be. For the subjects who wrote down that they were Â“sickenedÂ” or made Â“physically illÂ” by what they saw, there was no signi ficant difference between the neutral and disgusting ads. The responses were very lo w for both, in fact, with means of .05 for the neutral ad and .17 for the disgusting ad. Ne ither the neutral nor the disgusting image produced a strong enough feeling of illness for su bjects to be motivated to write about it. Since there was a significant difference when it came to being Â“disgusted,Â” but not when it came to being Â“sickened,Â” we can infer that the disgusting image for provegetarianism did not produci ng feelings of physical illn ess, making it correspond less with the theoretical meaning of disgust. When it came to the DES II scale for the pro-vegetarian ads, there was a significant difference between the neutral and disgusting ads for both the emotion of disgust and the feeling of being physically ill. For disgus t, the neutral group had a mean of 2.55 and the disgusting group had a mean of 4.39. For si ckness, the neutral group had a mean of 2.25 and the disgusting group had a mean of 4.13. The means were very similar for both
52 emotions and in both cases, the disgusting ad s brought out the emotions almost twice as strongly as the neutral ads. The emotion of disgust for the pro-vegetarian ads, however, had a slightly higher overall mean than the feel ing of sickness did and this was due solely to the disgusting image itself. Table 9. Levels of disgust for pro-vegeta rian advertisements by open-ended and Izard scale. N M SD SS df MS F Sig. Open-Ended Neutral Disgusted Disgust Total 20 23 43 .00 .52 .28 .00 .51 .45 Between Groups Within Groups Total 2.91 5.74 8.65 1 41 42 2.91 .14 20.80 .00 Open-Ended Neutral Sickened Disgust Total 20 23 43 .05 .17 .12 .22 .39 .32 Between Groups Within Groups Total .16 4.25 4.42 1 41 42 .16 .10 1.58 .22 DES II Neutral Disgusted Disgust Total 20 23 43 2.55 4.39 3.53 1.40 1.08 1.53 Between Groups Within Groups Total 36.27 62.43 98.70 1 41 42 36.27 1.52 23.82 .00 DES II Neutral Sickened Disgust Total 20 23 43 2.25 4.13 3.26 1.41 1.25 1.62 Between Groups Within Groups Total 37.83 72.36 110.19 1 41 42 37.83 1.77 21.43 .00 N=43; p< .05 The results lead us to believe that when s ubjects were given the ability to rate the level of disgust and sickness th ey were feeling, they had si gnificantly stronger feelings toward the disgusting ad for both emotions. However, when it came to giving subjects the opportunity to write down a nything about how they were feeling, more than half of them were motivated to mention how disgus ted they were by the disgusting ad. In contrast, they were not motivat ed to write down that they we re sickened by the disgusting ad. Therefore, it is clear that a very large amount of disgust was felt by students who viewed the disgusting ad, but not much sickness was felt toward the ad. This makes it seem possible that the feelings produced by the disgusting image itself are less in line with the theoretical meaning of disgust, and consequently may be more in line with the lay meaning of the word. Anger must be assessed, however, before we can make the
53 assumption that the disgusting pro-life image corresponds to the lay meaning of the word, since anger plays such a large part in that meaning. Assessing Anger In his book, Â“The Psychology of Emoti ons,Â” Izard (1991) wrote about the interaction between anger and disgust. He fe lt that disgust by itself may not be a serious cause of aggressive behavior. However, when disgust is combined with anger the emotion can become very dangerous. He wrote, Â“Instead of motivation to simply reject or avoid the object, combined anger and dis gust may create motivation to get rid of the object through attack or dest ructionÂ” (p. 264). Further, Nabi (2002) found from her study that the lay meaning for disgust took on all of the feelings associated with anger or at least a mix of anger and dis gust. Therefore, the emoti on of anger is important in determining whether a disgusting image co rresponded with the lay or theoretical definition of the word. Pro-Life For the pro-life ads, anger was experienced almost as highly as disgust which goes along with IzardÂ’s profile of em otions in the imagined disgus t situation. However, while there was a significant difference between feelin gs of disgust for the neutral versus the disgusting image as noted earlier, there was not a significant difference when it came to the feeling of anger for the pro-life ads (Table 10). This means that the disgusting image used for pro-life was not solely responsible for the feelings of a nger that made up the emotions of the subjects. The words, which stated, Â“Life is a gift. Abortion is murderÂ” were more likely the main reason for the ange r experienced, since the images themselves did not vary significantly in their results. This further supports the belief that the disgusting image for pro-life fell more in line with the theoretical meaning of the word
54 than the lay one. This image did not pr oduce a significant amount more anger in the subjects than the neutral one did. Table 10. Anger for pro-life advertisements by Izard scale N M SD SS df MS F Sig. Pro-Life Neutral Disgust Total 20 20 40 3.30 4.00 3.65 1.22 1.12 1.21 Between Groups Within Groups Total 4.90 52.20 57.10 1 38 39 4.90 1.37 3.57 .07 N=40; p< .05 Pro-Vegetarian For the pro-vegetarian ads, anger was e xperienced almost as highly as disgust, which also goes along with IzardÂ’s profile of emotions in the imagin ed disgust situation However, when it came to significance for th e level of anger for the neutral versus the disgusting pro-vegetarian ads, there was a significant difference (Table 11). This means that the disgusting image used fo r pro-vegetarianism alone produced a strong feeling of anger mixed with disgust. This st rengthens the belief th at the disgusting image for pro-vegetarianism corresponded less with the theoretical defin ition, and now due to the significant amount of anger that was expe rienced by subjects in this condition, we can also infer that it fell more in line with the lay meaning of disgust. Table 11. Anger for pro-vegetarian advertisements by Izard scale N M SD SS df MS F Sig. Pro-Veg. Neutral Disgust Total 20 23 43 2.55 3.84 3.23 1.47 1.15 1.45 Between Groups Within Groups Total 17.42 70.25 87.67 1 41 42 17.42 1.71 10.17 .00 N=43; p< .05 The anger experienced by the viewers of the disgusting pro-life ad cannot be attributed to the disgusting image, because there was not a significant difference between the level of anger in the neutral and disgusti ng pro-life ads. However, since there was a significant difference between the neutral and disgusting pro-vegetarian ads when it came to anger, we can attribute this emotion to the image itself, since that was the only thing
55 that differed in the two conditions. Therefor e, the disgusting image for pro-vegetarianism encompassed the lay meaning of disgust, which may have made the entire ad less effective to its viewers. All of the results that were found to help answer the res earch questions lead us to believe that both of the original hypotheses can be accepted for this study. Hypothesis 1 predicted that an image that correlates with the theoretical meaning of disgust will enhance persuasion when associated with the oppositionÂ’s position within a two-sided controversial public issue. Th is was the case for the disgustin g pro-life image, which was very effective at capturing attention and influencing some attitude and behavior change. The disgusting pro-life image did not contain a significant am ount of anger, which is why it fell more in line with the theoretical de finition of the emotion. Also, the entire disgusting pro-life ad did not at all cause view ers to feel defensive or that the sponsorÂ’s views were being pushed upon them. Overall, the disgusting pro-life ad, which used the theoretical definition, was very effective. Hypothesis 2 predicted that an image that correlates more with the lay meaning of disgust, which includes anger, may cripple persuasion and more likely enhance a transfer of negative feelings toward the organization that is showing the ad. This was true for the disgusting pro-vegetarianism ad. The ad was effective at capturing attention, but not at eliciting any real attitude or behavior change among th e viewers. In addition, the disgusting image itself led to a significant amount of anger, which put it more in line with the lay meaning of the emotion of disgust. This disgust mixed with anger may have been the cause for the defensiveness that was felt by th e subjects in this c ondition. Overall, the
56 disgusting pro-vegetarianism a d, which used the lay definition, was less effective in the long run.
57 CHAPTER 5 DISCUSSION AND CONCLUSION The current study not only looked at the e ffectiveness of shocking advertisements, but shocking ads that contained disgusti ng images, in particular. Through norm violation, shocking ads are meant to capture th e attention of an audience and potentially lead to attitude and behavior change. A ll of the ads that were used were shocking, because even the ones with a neutral image included some alarming statements. As DahlÂ’s preliminary shock model suggested, these ads captured the subjectsÂ’ attention, enhanced message retention and influenced so me of their behaviors. All four of the experimental conditions resulted in emotional reactions from the subjects, with all but the neutral vegetarian image elic iting strong responses. The goal of this study, though, was to determine how disgusting images impact the emotional responses of subjects. Using disgusting images in an ad is one form of shocking advertising. When the audience engages in an experience of disgust, they should feel as if they want to turn away from what it is that they find disgusti ng. However, Nabi (1998) elaborated further by stating that the Â“action tendency of turning away from or avoiding the disgust-eliciting material will encourage a responde nt to similarly turn away from, or reject, the position associated with that negative affectÂ” (p. 25). Based on her studies, she also went on to say that if disgust is associ ated with the oppositionÂ’s posi tion in a refutational message, then the audience will turn aw ay from or reject the opposit ionÂ’s position, and in turn strengthen persuasion toward the sponsorÂ’s message.
58 For this study, the ads containing disgusting images were seen as more disgusting by the subjects who viewed them. Both dis gusting ads also stimulat ed a strong negative emotional response from the subjects, which is in line with NabiÂ’s beliefs. However, the open-ended questions led us to believe that only the disgustin g pro-life ad caused subjects to associate this negative f eeling with the issue at hand. The disgusting pro-vegetarian, on the other hand, caused subjects to associate this feeling more with the sponsor of the message than the issue itself. The open-ended responses revealed that this ad caused some of the subjects to react in a very defensive way and to explicitly state that they woul d never give up eating meat. The disgusting pro-vegetarian ad was not in line with Nabi Â’s belief that a disgusting image used to discredit the opposition will cause the audien ce to turn away from the oppositionÂ’s viewpoint. The reason for this could have to do with anger. The significant amount of anger produced from the disgusting pro-vege tarian image may have even turned the subjects away from the idea of becoming a vegetarian, in turn strengthening and reinforcing their decision to eat meat. Izard wrote that when anger is mixed w ith disgust, it can turn into a hostile experience where the observer feels the need to attack the source of the disgust, who in this case would be the message sponsor. The statistical signif icance found between the neutral and disgusting pro-vegetarian images revealed that the disgusting image itself led to a strong element of anger along with disgus t. Anger was high in all four conditions, but this likely had something to do with the words used in th e ads as well. However, the disgusting pro-vegetarian ad was the only one where the image, in particular, produced a statistically significant amount of anger. This anger may be why the disgusting pro-
59 vegetarian image was less effec tive. It caused students to be come defensive and attribute any negative feelings they had to the sponsor of the message, rather than the opposition. The results show us that th e disgusting ad was more eff ective for pro-life than for pro-vegetarianism. In other words, wh en it comes to abortion, strong words and a disgusting images should both be present in orde r to be effective. If shocking words are used without a shocking image, students beco me angry and develop negative feelings toward the organization that is sending the message. In contrast, the disgusting ad was less effective for the issue of pro-vegetarianis m. Further, when it comes to eating meat, a disgusting image causes students to become very defensive, as well as a strong combination of anger and disgust. It is impor tant to note, however, that these two issues were so different and cannot easily be compar ed. The issue of abortion is more often seen as a moral one. On the other hand, ea ting meat is an everyday occurrence in the lives of millions of Americans. The disgusti ng ad for pro-life was very effective and the disgusting ad for pro-vegetari anism was not as effective for each of the two separate issues, but it is best not to compare them. Also, the anger that took place with the di sgusting pro-vegetarian ad leads us to believe that the image in the ad corresponded mo re with the lay meaning of disgust. In contrast, the image in the di sgusting pro-life ad corresponde d more with the theoretical meaning of disgust. Nabi found that in ev eryday life the term di sgust corresponds more closely with the theoretical meaning of anger or at least a mixture of anger and disgust, than it does with disgust alone . Since the disgusting pro-lif e image was effective and the disgusting pro-vegetarian image was not, th ese results support NabiÂ’s paper on the
60 differences between the theore tical and lay meanings of th e word disgust as well as IzardÂ’s description of what happens when th e emotions of anger and disgust interact. Overall, when it comes to the emotion of disgust, the lay meaning may be less effective than the theoretical one when working with controve rsial issues, although further research should be done and each issue should be separately examined. To sum, with the growing trend of shoc k advertising, there needs to be more research done on its effectiveness as a persuasi ve tool. We have learned from this study that disgust can be an effective tool for persuasion, when used appropriately. The theoretical use of the term produces better results than th e lay meaning, because as we have seen the lay meaning causes feelings of anger. This anger when mixed with disgust can lead to hostility in the audience, and results that are the opposite of what you had hoped for. From this research we learn th at a disgusting image may either repel the audience away from the opposing viewpoint or repel the audience away from the message sponsor. There is a very fine lin e and a researcher needs to understand his audience, as well as be careful about the amount of anger that is present. And understanding the audience is a key factor in getting them to accept a point of view and follow through with a behavior. If nothing el se, this study revealed that the lay meaning of disgust should probably be avoided in certain issues if the intention is to persuade the audience to accept your view and discount some one elseÂ’s. However, more research should be done to determine whether the lay meaning of disgust is always less effective than the theoretical one when trying to persuade.
61 It is important for researchers to know a lot about the issue they are working with and how it makes audience members feel before they determine whether or not to use disgusting images or shocking tactics. Limitations A major limitation in this study was that th e researcher was not able to gather enough information on how the subjects really fe lt about each of the issues before they participated in the study. It is crucial to know all about th e viewersÂ’ feelings regarding controversial issues before trying to persua de them to accept the sponsorÂ’s message. A more in depth look into their feelings just on these issues alone w ould have led to much more insight. Another limitation was acce ss to a larger group of participants for the main experiment, as well as for the posttest. The pos ttest did not reveal si gnificant results, in part, because so few people took it. In the future, more time and preparation would work to remedy this problem. Future Research In future studies, more time should be sp ent looking at the initi al feelings of the participants before they even fill out the ques tionnaire. Controversial public issues are all so different and very sensitive, so it is im portant to know exactly how your audience feels about them. Also, future studies should c ontinue to be used to understand the lay meaning of the emotion of disgust and the imp lications that it has for creating effective persuasive messages. More research should look at the results fo r the combination of disgust and anger and if they are ever eff ective for persuasion when used together. Different levels of disgust s hould be measured to determine whether strong feelings of disgust are more effective than mild feelings of the emotion. It would also be helpful to
62 study the defensiveness that can sometimes be seen in a significant anger and disgust situation and whether it causes audience memb ers to put up walls, discredit the sponsorÂ’s message, and not allow themselves to become affected by what they see.
63 APPENDIX A SAMPLE OF THE PRE-TEST What are your thoughts and/fee lings about this image? How interested in the image do you feel? Not at All 1 2 3 4 5 Very Interested How happy does the image make you feel? Not at All 1 2 3 4 5 Very Happy How surprised were you by the image? Not at All 1 2 3 4 5 Very Surprised How sad does the image make you feel? Not at All 1 2 3 4 5 Very Sad How angry does the image make you feel? Not at All 1 2 3 4 5 Very Angry How disgusted does the image make you feel? Not at All 1 2 3 4 5 Very Disgusted How fearful does the image make you feel? Not at All 1 2 3 4 5 Very Fearful How comfortable does th e image make you feel? Not at All 1 2 3 4 5 Very Comfortable How guilty does the image make you feel? Not at All 1 2 3 4 5 Very Guilty
64 APPENDIX B NEUTRAL PRO-LIFE ADVERTISEMENT Life is a Gift ABORTION IS MURDERÂ… Become pro-life. Adapted from www.abortionismurder.org.
65 APPENDIX C DISGUSTING PRO-LIFE ADVERTISEMENT Life is a Gift ABORTION IS MURDERÂ… Become pro-life. Adapted from www.abortiontv.com.
66 APPENDIX D NEUTRAL PRO-VEGETARIAN ADVERTISEMENT Just A Reminder This is how your dinner looked before it hit your plate. It could breathe, feel and had a heart. JUST LIKE YOUÂ… Go Vegetarian. Adapted from www.nal.usda.gov.
67 APPENDIX E DISGUSTING PRO-VEGETARIAN ADVERTISEMENT Just A Reminder This is how your dinner looked before it hit your plate. It could breathe, feel and had a heart. JUST LIKE YOUÂ… Go Vegetarian. Adapted from www.peta.org.
68 APPENDIX F SAMPLE OF THE MAIN QUESTIONNAIRE BEFORE EXPOSURE TO AD Please indicate your feelings about the death penalty (AdSAM). Please indicate your feelings about vegetarianism (AdSAM). Please indicate your feelings about abortion (AdSAM). AFTER EXPOSURE TO AD Please indicate how the ad makes you feel (AdSAM) and in your own words. How disgusted are you by the advertisement? Not at All 1 2 3 4 5 Very Disgusted How angry does the ad make you feel? Not at All 1 2 3 4 5 Very Angry How disdainful do you find the ad to be? Not at All 1 2 3 4 5 Very Disdainful How distasteful do you find the ad to be? Not at All 1 2 3 4 5 Very Distasteful How sad does the ad make you feel? Not at All 1 2 3 4 5 Very Sad How surprised are you by the ad? Not at All 1 2 3 4 5 Very Surprised Does the ad make you feel sick to your stomach? Not at All 1 2 3 4 5 Very Sick How interesting do you find the ad to be? Not at All 1 2 3 4 5 Very Interesting
69 APPENDIX G POSTTEST Which ad and image did you view on-line? What do you remember about the ad you viewed on-line? Whether good or bad, how often have you thought about the ad you saw since taking the survey? Not At All 1 2 3 4 5 All of the Time How much has your attitude changed rega rding this issue in the past week? Not At All 1 2 3 4 5 Very Much What is your current attitude toward th e issue raised in the advertisement? How much has your behavior changed rega rding this issue in the past week? Not At All 1 2 3 4 5 Very Much If your behavior has changed in the past week, please explain how.
70 LIST OF REFERENCES Austin, E.W., Pinkleton, B., & Fujioka, Y. A ssessing prosocial message effectiveness: Effects of message quality, produc tion quality, and persuasiveness. Journal of Health Communication 4: 195-210, 1999. Bagozzi, R.P., & Moore, D. J. Public service advertisements: Emotions and empathy guide prosocial behavior. Journal of Marketing 58: 56-70, 1994. Bandura, A. Social foundations of thought and action: A social cognitive theory, Englewood Cliffs, New Je rsey: Prentice Hall, 1986. Boster, F.J. & Mongeau, P.A. Fear-arousing pe rsuasive messages. In R.N. Bostrom (Ed.), Communication Yearbook 8: 330-375, 1984. Chabet, J.C. & Chabet, G.C. Drama advertis ements: Moderating eff ects of self-relevance on the relations among empathy, information processing, and attitudes. Psychology Reports 92: 997-1014, 2003. Coleman, C. The influence of mass media a nd interpersonal communication on societal and personal risk judgments. Communication Research 20: 611-628, 1993. Cunningham, Gregg. The reproductive choice campaign. The Center for Bio-Ethical Reform, 2003. Retrieved from www.abortionno.org , September 2005. Dahl, D. Does it pay to shock? Reactions to shocking and non-shocking advertising content among university students. Journal of Advertising Research , 268-280, September 2003. DeJong, W. & Atkin, C.K. A review of nationa l television PSA campaigns for preventing alcohol-impaired driving. Journal of Public Health Policy 16: 59-80, 1995. DeJong, W. & Winsten, J.A. The use of ma ss media in substance abuse prevention. Health Affairs , 30-46, 1990. Dillard, J. P., Plotnick, C. A., Godbold, L. C ., Freimuth, V. S. & Edgar, T. The multiple affective outcomes of AIDS PSAs: Fear appeals do more than scare people. Communication Research 23: 44-72, 1996.
71 Engineer, R., Phillips, A., Thompson, J., & Nicholls, J. Drunk and disorderly: A qualitative study of binge drinking among 18-24 year olds. Home Office Research , 22-31, February, 2003. Fishbein, M. & Raven, B.H. The AB scales : an operational definition of belief and attitude. Human Relations 12: 32-44, 1962. Fishbein, M., Von Haeften, I.; Hall-Jamies on, K., Johnson, B., & Kirkland, R. Avoiding the boomerang: Testing the relative e ffectiveness of antidrug public service announcements before a national campaign. American Journal of Public Health 92: 238-245, 2002. Freimuth, V.S. Reaching those at risk: A content analytic study of AIDS PSAs. Communication Research 17: 775-791, 1990. Grube, J.W. Alcohol portrayals and alcohol advertising on television. Alcohol Health & Research World 17: 61-66, 1993. Gustafson, Bob, & Johan Yssel. Are advertisers practicing safe sex? Marketing News , 23-28, March 1994. Heckler, Susan E., & Childers, Terry L. The ro le of expectancy and relevancy in memory for verbal and visual inform ation: What is incongruity? Journal of Consumer Research 18: 475-92, 1992. Izard, C. Human Emotions . New York: Plenum, 1977. Izard C. The Psychology of Emotions . New York: Plenum, 1991. Keller, K. L. Memory factors in advertising: The effect of advertis ing retrieval cues on brand evaluations. Journal of Consumer Research 14: 316-324, 1987. Kalish, David. Doing well by doing good. Marketing and Media Decisions 25: 22-23, 1990. Lazarus, Richard S. Emotion and Adaptation , New York: Oxford University Press, 1991. Lee, J.S. & Davie, W.R. Audience recall of AIDS PSAs among U.S. and international college students. Journalism and Mass Comm unication Quarterly 74: 7-22, 1997. Loxafamosity Ministries. Facts a bout abortion, 2005. Retrieved from www.abortionismurder.org , July 2005. MacInnis, Deborah J., Moorman, Christine, & Jaworski, Bernard, J. Enhancing and measuring consumersÂ’ motivation, opport unity, and ability to process brand information from ads. Journal of Marketing 55: 32-53, 1991.
72 Marchand, J & Filiatrault, P. AIDS prevention advertising: Different message strategies for different communication objectives. International Journal of Nonprofit and Voluntary Sector Marketing 7: 271-187, 2002. Nabi, Robin L. The effect of disgust-elic iting visuals on attitudes towards animal experimentation. Communication Quarterly 46: 472-480, 1998. Nabi, Robin L. The theoretical versus the lay meaning of disgust: Implications for emotion research. Cognition and Emotion 16: 695-703, 2002. No Fear Media, Inc. Aborti on pictures, 2005. Retrieved from www.abortiontv.com , July 2005. Pechmann, C & Reibling, E. Do anti-smoking ads combat underage smoking? A review of past practices and research. Social Marketing: Th eoretical and Practical Perspectives . Hillsdale, New Jersey: 189-216, 1997. People for the Ethical Treatment of Animals. PETA's history: Compassion in action, 2005. Retrieved from www.peta.org , September 2005. Petty R.E. & Cacioppo, J.T. The elabor ation likelihood model of persuasion. Advances in Experimental Social Psychology. New York: Academic Press. 19: 123-205, 1986. Pyszczynski, Thomas A., & Greenberg, Jeff. Ro le of disconfirmed expectancies in the instigation of attributional processing. Journal of Personality and Social Psychology 40: 31-38, 1981. Rigby, K., Brown, M., Anagnostou, P., Ross, M.W. & Rosser, B. R. S. Shock tactics to counter AIDS: The Australian experience. Psychology & Health 3: 145-159, 1989. Rimal, R. Differential impact of emotional and rati onal PSAs on information processing and learning. Paper presented at the annual c onferences at th e International Communication Associati on, Montreal, Canada, 1997. Rotfeld, H.J. Fear appeals and persuasion: Assumptions and erro rs in advertising research. Current issues and Research in Advertising 11: 21-40, 1988. Rozin, P. Disgust: The body and soul emotion. Handbook of Cognition and Emotion, New York: Wiley, 429-445, 1999. Shimp, T.A. & Stuart, E.W. The role of disgus t as an emotional medi ator of advertising effectiveness. Journal of Advertising 33: 43-53, 2004. Slater, M.D. Drinking and driving PSAs: A c ontent analysis of behavioral influence strategies. Journal of Alcohol and Drug Education 44: 68-81, 1999.
73 Syed, Kirran. Anti-abortion campaign hits Chillicothe. Chillicothe Gazette, 2003. Tinic, Serra A. United colors and untied m eanings: Benetton and the commodification of social issues. Journal of Communication 47: 3-25, 1997. United States Department of Agriculture, Agricultural Research Service. National agricultural library, 2005. Retrieved from www.nal.usda.gov , July 2005.
74 BIOGRAPHICAL SKETCH The author graduated from Rutgers Univer sity in New Brunswi ck, New Jersey, in 2004 with a bachelor's degree in journalis m and mass communication and a minor in women's studies. She moved to Gainesville, Flor ida, to receive a master's degree in advertising from the University of Florida in the spring of 2006. She currently resides in New York City, where she is working on an agency career in advertising with a particular interest in acc ount management.