Masking the Offense? An Ethical View on Humor in Advertising

Kati Frster^a and Cornelia Brantner^b

aDepartment of Communication, University of Vienna; bInstitute of Media and Communication, Dresden University of Technology

ABSTRACT

The use of humor in advertisements has been the subject of numerous studies, most of which have emphasized humor's effectiveness while neglecting the ethical issues that it often poses when used in advertising. In response, the authors look at the extent to which humor masks the ethical concerns inherent in offensive advertising. The authors examine advertisements brought before the Austrian Advertising Council, as well as the case-specific decisions made by councilors. With their analysis, they contribute to advertising ethics by turning attention to an important orga- nization in the practice of critiquing advertisements.

ARTICLE HISTORYReceived 18 June 2015

Accepted 21 October 2015

Without question, humor is a popular communication strategy in advertising (e.g., Alden,Mukherjee, & Hoyer, 2000; Eisend, 2009; Weinberger & Gulas, 1992; Weinberger & Spotts, 1989). For example, Toncar (2001) estimated that between 21% and 48% of television advertisements in the United States are humorous. At the same time, Laroche, Nepomuceno, Huang, and Richard (2011) investigated the use of humor in magazine advertising in the United States, China, and France and found considerable differences. While in France and China only 16% of advertisements involved humor, in the United States the figure was 24%. The extensive use of humor in advertising occurs for good reasons, and several studies have attested to its positive impact on potential customers, namely by increasing the recipients' attention, likability of the product, and purchase intention.

Yet, while these analyses focus often on examining the effectiveness and contributions to individual buying behavior of humorous advertisements, ethical issuesconcerningtheuseofhumor in ads have been widely ignored. Murphy (1998) attributes the general lack of attention on advertising ethics to the "unholy trinity" (p. 318). He argues that advertisers, agencies and the media as the three main parties in any advertising campaign are not willing to accept their responsibility for ethical standards. Nevertheless and exactly for that reason, scholars and professionals alike have highlighted advertisers' responsibility in promoting societal well-being (e.g., Baker & Martinson, 2001; P. Cunningham, 1999; Waller, 2012). Among others, Fry and Polonsky (2004) stated that advertising effects cannot be limited to the relationship between advertisers and consumers, but extend to the sociocultural environment and society at large. Potter (2009/1954) even ascribed advertising the role of an "instrument of social control" (p. 175) and suggested that "advertisingnow compares with such long-standing institutions as the school and the church in the magnitude of its social influence" (p. 167). However, since the beginning of advertising, this role has not always been fulfilled, for many advertisements contain statements and/or visual elements that violate ethics and offend audiences. Beard's (2008a) summary that in U.S. advertising "racist, sexist, and other potentially offensive themes and portrayals were prevalent" in the last years attests to these circumstances (p. 1). Advertising ethics literature has classified offensive advertising into matter (offensive products and services) and manner (offensive themes; Barnes & Dotson, 1990; Prendergast, Cheung, & West, 2008). The latter includes controversial themes, such as sexual

2016 Taylor & Francis

practices, racial or religious prejudice, and antisocial behavior (Beard, 2008a; Waller, 1999). Beard (2008b, p. 17) found in his content analysis of complaints brought before the New Zealand Advertising Standards Authority that "images/words unsuitable for children" was the most common reason for complaints about potentially offensive ads, followed by "bad language," "portrayed or strongly implied violence," "sexual themes and nudity," "gender-related demeaning or degrading situations," and "use of stereotypes/prejudice." Moreover, he found that audiences are more likely to be offended by manner (offensive themes) than matter (offensive products and services).

In this respect, the masking effect of humor is of particular interest (e.g., Breitmeier & Ogmen, 2000; Hsieh, Hsu, & Fang, 2010; Shabbir & Thwaites, 2007), since humor is apparently able to cover other stimuli. More precisely, masking appears "when the processing of a stimulus is interrupted by the subsequent immediate presentation of a second, different stimulus. The second stimulus acts retroactively to obscure the former one" (Moore, 1988, p. 302). Masking can thus precipitate situations in which the recipients of advertisements do not recognize any originally neutral or negative stimuli (e.g., offensive themes) or perhaps even evaluate ethically questionable ads asbeing positive as long as they are funny. In this sense, humor can act as a masking stimulus that reduces the visibility and thus the perception of ethical offenses in advertisements.

With the present study, we fill a gap in the literature addressing humor in advertising by focusing on the masking effect of such humor from the perspective of ethics. In connecting research regarding advertising ethics with the effects of humor in advertising, we aim to bridge academic and profes- sional interests. We round out our discussion of theoretical considerations with an empirical study of advertisements that Austrians found offensive and thus brought before the Austrian Advertising Council, as well as the decisions that the councilors made regarding these ads. The study focused on the question of whether advertising councilors perceive (potentially) ethics-violating advertisements as less unethical if they contain humor. To the best of our knowledge, the investigation was the first to use data from actual, topical decisions by advertising standards authorities about real advertise- ments that citizens claimed were offensive before the council.

Theoretical framework

When examining humor in advertising from the viewpoint of ethics, the goal is not to advocate banning humor from advertisements, but to better understand the potential threats of humor actingas a stimulus that masks ethical offenses. The theoretical framework applied in this study is thus primarily concerned with advertising ethics.

Within the scope of applied ethics, P. Cunningham (1999) has defined advertising ethics as "what is right and good in the conduct of the advertising function. It is concerned with questions of what ought to be done, not just what legally must be done" (p. 500). Drumwright (2007) furthered this distinction by stating, "Laws are ultimately a reflection of ethical judgments. . .A fundamental mistake, however, is to assume that because something is legal, it is ethical, or if something is unethical, it will be made illegal. . . Advertising law is a subset of the domain of advertising ethics" (p. 399). In investigating advertising ethics,

P. Cunningham (1999) has suggested viewing the ethical criticism of advertising at three different levels: the macro level of society at large, the meso-level of the advertising industry as a whole, and the micro level of the individual practitioner, campaign, or as Drumwright (2007) added, individual customers or specific advertising practices. By their nature, yet also due to diverse research interests, credible perspec- tives from these three levels require distinct theoretical foundations and methodologies. For example,while the micro perspective focuses chiefly on the short-term effects of specific stimuli by drawing on psychology and experimental paradigms (Drumwright, 2007), the macro perspective concentrates on the aggregate effects of advertising, including advertising to vulnerable segments (e.g., Bonifield & Cole, 2007) and advertising's role in consumption and collective welfare (e.g., O'Guinn, 2007).

A common criticism of advertising from the macro level is that advertising protects the existing social order and promotes inequality, particularly in terms of race and gender (O'Guinn, 2007). Drumwright (2007) has categorized social criticism of advertising on the macro level according to its

three primary critiques: advertising encourages excessive materialism, advertising engenders and/or reinforces problematic stereotypes, and advertising cultivates false values and thus problematic behavior. O'Guinn (2007) has stated that even though the majority of studies critical of advertising have specifically focused on issues of beauty and objectification, the lack of (careful) longitudinal research makes it impossible to confirm that repeated exposure to gendered representations yields unrealistic expectations of beauty or attaches social anxiety to gender role enactments. However, O'Guinn (2007) also argued that, according to socialization theory and social reality effects, we have to acknowledge that representations become part of assumed norms of constructed assumptionsabout the world and the attractiveness and happiness of other people. This is also supported by correlational studies that have linked sociodemographics to viewing preferences and users' under- standing and interpretation of sexual portrayals in the media. As an example, Walsh-Childers and Brown (1993) found that adolescents with an intense media usage are more likely to accept stereotypes of sex roles on television as realistic than less frequent viewers. Moreover, objectification theory (Fredrickson & Roberts, 1997) posits that sexual objectification experiences lead to an internalization of cultural standards of attractiveness provided by the media. Such experiences, in turn, foster self-objectification and persistent body surveillance, which can lead to body shame and reduced psychological well-being (Moradi, 2010). Several studies (e.g., de Vries & Peter, 2013; Harper & Tiggemann, 2008) have demonstrated the negative effects of advertising images on women's self-objectification and body image. Also, Grabe, Ward, and Hyde (2008) confirmed intheir meta-analysis of 77 studies that exposure to media images depicting thin-ideal bodies is related to body image concerns for women.

Based on interviews with industry leaders, Drumwright and Murphy (2009) further distinguish advertising ethics in a message perspective and a business perspective. While message ethics refer to creating and delivering the commercial message, business ethics has to do with the processes involved in running the advertising agency. A central and highly practical approach to messageethics is the TARES test (Baker & Martinson, 2001; Lee & Cheng, 2010), which enumerates five traits that an advertising message must exhibit: the truthfulness of the message, the authenticity of the persuader, respect for the person being persuaded, the equity of the persuasive appeal, and social responsibility for the common good. For instance, Lee and Ngyen (2013) investigated 380 fast-food ads in Singapore regarding their ethical dimensions using TARES measures. Professionals in the advertising industry thus possess practical instruments for evaluating the messages that they sendand the personal roles that they play in the "advertiser-media-consumer triad" (A. Cunningham, 1999, p. 83).

When scrutinizing the social role of advertising, people have to ask how the advertising industry itself shoulders its responsibility towards the public in practice. For starters, advertising is a form of public communication and thus subject to regulation, both internally and by law. In the UnitedStates, for example, the Federal Trade Commissionan independent agency of the U.S. government

regulates advertising. This is complemented by the Advertising Self-Regulatory Council. Many countries in Europe have established strong self-regulatory bodies to ensure that ads that violatetheir advertising standards are corrected or withdrawn. In Europe, internal or self-regulation is implemented by the European Advertising Standards Alliance, of which currently 38 self-regulatory bodies are members: 27 from 25 European countries and 11 from non-European countries, including India, Australia, Brazil, Peru, and Canada. The self-regulatory bodies that operate in different countries are typically established by the communication industry itself (e.g., media, advertisers, advertising and media agencies) and societal organizations (e.g., universities, lawyers, chambers of commerce) to maintain a high quality of advertising and ensure consumer trust and protection forthe benefit of the industry as a whole.

Advertising councilors are thus an important group in the lived practice ofadvertising critique. On the one hand, they represent the ethical values and perceptions of morality of the respective advertising branch, and on the other, with both individual and group decisions, they shape theethical standards of the communications industry. Examining the decisions that councilors make

concerning individual ads is thus of specific relevance, for the councils act to filter out ethics- violating ads and, in doing so, develop a leverage effect regarding not only individual consumers and professionals, but the societal consequences of advertising in general. Advertisers, consequently, may be motivated to shift from being positivist utilitarian pragmatists, who are liable to support whatever works and proves useful, toward being partners in a social contract, in which "corporations have a responsibility to use their power to benefit, rather than harm, society" (A. Cunningham, 1999, p. 90). These two positions, utilitarian and social contract, represent two different perspectives in which advertising ethics can be discussed. The first, utilitarian, builds on principles by asking what is best for all, or the greatest number of people possible (Mill, 2003). For Mill moral reasoning "was equivalent to calculating consequences for human happiness" (Christians, 2007, p. 118), that is, utilitarianism requires advertisers to maximize happiness for all (A. Cunningham, 1999).Thesecond, social contract theory, posits that persons' moral obligations are dependent upon a contract "among them to form the society in which they live in" (Friend, 2004, para. 1). Rawls (1971) argues in his influential work, A Theory of Justice, that the best society would be formed by principles of justice chosen by rational citizens behind a "veil of ignorance," in which their own social status and goals would not affect their decisions. The mentioned TARES test comes close to Rawls' considera- tions because the veil of ignorance requires professionals "to step conceptually out of their roles as powerful disseminators of persuasive promotional messages and to evaluate the equity of the appeal from the perspective of the weaker parties" (Baker & Martinson, 2001, p. 166).

The masking effect of humor in advertising

Despite its popularity, humor represents one of the least understood advertising mechanisms (Sutherland & Sylvester, 2000), and empirical proof of its effects remains insufficient and even paints a contradictory picture (e.g., Lyttle, 2001). As a consequence, debates over humorous adver- tisements' ability to promote favorable reactions from consumers persist. In these debates, the sole common tenet seems to be that the impact of humor in advertising depends on product character- istics such as involvement (e.g., Chung & Zhao, 2003) on the one hand, and on the other, audiences' peculiarities, including prior knowledge, attitude toward brand, gender, and culture. Humor is a complex topic that has been studied widely in recent decades. Nevertheless, research in the field is characterized by contradictory findings, which often poses more questions than answers.Inresponse, researchers have conducted meta-studies to structure and evaluate findings in the field (Eisend, 2009; Gulas & Weinberger, 2006; Weinberger & Gulas, 1992). Eisend (2009) found, for example, that humor in ads significantly enhances recipients' attitudes toward those ads, attitudes toward the brand, attention, positive affect, and purchase intention. At the same time, he found that humor in advertising significantly reduces source credibility, as well as discovered no evidence of humor's impact on recipient cognitions or preferences for the advertiser. Furthermore, though Weinberger and Gulas (1992) underscored the positive effect of humor in ads on recipients' likingthe advertiser, Eisend (2009) found no support for such a relationship.

Even less is known about the masking effect of humor in advertising. Gass and Seiter (2003) define the masking effect as follows: "humor may divert attention away from the content of a message, thereby interfering with the ability of listeners to carefully scrutinize the message or engage in counterarguing" (p. 292). Among the few studies on the topic, Shabbir and Thwaites (2007) found that humor was used to mask deceptive claims in 75% of ads they investigated. In their study, they presumed that humor is either consciously or unconsciously used as an appeal to conceal and thus mask certain advertising messages. More recently, Hsieh and colleagues (2010) investigated the perception of deceptive commercials and the masking effect of humor, the results of which indicate "that humorous ads produce a masking effect that not only reduces the negative effect of deceptive claims, but also causes deceptions in the ads to appear more reasonable" (p. 21). In both studies, the authors began with the presumption that humor possesses a masking ability that primes recipients to absorb deceptive content, thereby altering the perception of the deceptive claim in the ad "by

mediating its perceived severity through the illusionary association with the positive aspects of humor" (Shabbir & Thwaites, 2007, p. 76; see also Hsieh et al., 2010, p. 2). In a similar manner, several authors have argued that combining humor with aggressive and brutal content leads to a "desensitization effect" (Scharrer et al., 2006) or a "camouflage effect" (Potter & Warren, 1998), resulting in ads portraying violence in tandem with humor being perceived as less aggressive thansimilarly violent ads without humor.

In this regard, related literature offers different theoretical considerations of why and how humor may mask other content:

Humor as distraction: First, researchers have found that using humor is an attention-gaining mechanism. In this sense, what advertisers perceive as a desired effect upon recipientsnamely, the positive effects of increased attention on physiological reactions (Szabo, 2003) and awareness (Cline & Kellaris, 2007; Goldberg & Gorn, 1987; Madden & Weinberger, 1982)needs to be critically viewed in terms of its masking effect of ethical offenses. In fact, research has shown that humor distracts receivers from constructing counterarguments (Osterhouse & Brock, 1970; Romero, Agnew, & Insko, 1996), which reduces the probability of central processing. Put differently, humor redirects attention from ethical violations in advertising. Lyttle (2001) has added that this effect may be even stronger when ironic humor is used, since irony calls for deeper processing and thereby urges an even stronger distraction toward the target stimulus.

Humor as mood-maker: Second, humor is a feature that induces positive mood states (e.g., Kamins, Marks, & Skinner, 1991; Kuiper, McKenzie, & Belanger, 1995; Moran, 1996). Research suggests that positive mood in turn promotes the use of heuristics and/or reliance on peripheral cues (Bless & Schwarz, 1999; Isen, 1989) and that people in a good mood are less likely to disagree with a persuasive message (Freedman, Sears, & Carlsmith, 1978). Furthermore, positive mood is attributed to the source (Sinclair, Mark, & Clore, 1994) and thus enhancesits likability (Sternthal & Craig, 1973).

Humor as relationship-builder: Third, research has indicated that humor shapes relationships between advertisers and recipients by illustrating a shared set of underlying values: a shared sense of humor. This commonality in turn promotes a more positive image of the advertiser (Chung & Zhao, 2003; Gelb & Picket, 1983). While Eisend (2009) found no empirical evidence of the presumption that humor enhances source likeability, Weinberger andGulas(1993) found strong support for this relationship and concluded that "the link between humor and liking is stronger than for any of the other factors" (p. 57). Likability in turn reduces the visibility of any unethical message by increasing the probability that recipients use a peripheral route of processing.

Hypotheses

In light of all of these considerations, we formulated four hypotheses. The first hypothesis empha- sizes the potential masking effect of humor on the decisions of advertising councilors about potentially offensive ads.

H1: Advertising councilors issue milder decisions on humorous than non-humorous advertisements.

Some studies have emphasized message characteristics as moderating variables in the effectiveness of humor in advertising. In other words: It has been argued that the impact of humor in advertising depends on the messages itself. Beard (2008a), for example, examined 300 complaints published online by the Advertising Standards Authority of New Zealand on offensive themes and intentional humor depicted in ads. He found that audiences are more sensitive to offensive themes than to anything else, as well as that "offense has less to do with intentional humor than with the humor's inclusion of inherently offensive themes" (p. 13). Inappropriate humor can therefore become the

subject of ethical violations and even recognized as such. Prendergast, Ho, and Phau (2002) confirmed this suggestion by identifying sexism as a chief reason why ads are perceived to beoffensive. Such a situation would pose sure implications for the masking effect of humor insofar as the combination of humorous and sexist elements in ads reduce the masking effect, thereby initiating an amplifying effect that renders the ad's being perceived as even more offensive. To put it more formally:

H2: Advertising councilors issue more rigorous decisions on overt sexist advertisements thatinclude humor than on overtly sexist depictions without humor.

In addition, in a prior study, Waller (1999) found that women are more likely to be offended by indecent language and nudity, as well as sexist, racist, and/or antisocial behavior. Such a trend would indicate that the gender of councilors affects their decisions, or:

H3: Women advertising councilors issue more rigorous decisions regarding ethics-violating advertisementsthan men councilors.

Given that we have also acknowledged evidence of humor's ability to increase the likeability of ads among men but not women (Gulas & Weinberger, 2006), we have reason to suppose that gender also acts as a moderating variable for the masking effect of humor. We have thus hypothesized:

H4: The masking effect of humor is weaker for women than for men councilors.

Altogether, we hypothesized that the use of humor in ads affects the decisions of advertising councilors concerning whether and to what degree ads are determined to be offensive (maskingeffect; Hypothesis 1), as does the gender of the councilor (gender effect; H3). With Hypotheses 2 and 4, we expanded these predictions by averring the moderating effects of an ad's use of overt sexism and the councilor's gender on the masking effect of humor.

Method

The Austrian Advertising Council provided us with data including ads about which citizens have complained to the Austrian Advertising Council and the respective decisions of council members regarding these ads. In Austria, as in nearly all European countries, a dual system is established for the regulation of advertising. Along with following legal regulations, the Austrian advertising sector implements self-restricting guidelines that its members have voluntarily accepted. The Austrian Advertising Council is the self-regulatory body of the Austrian advertising industry that issues decisions concerning complaints of infringements upon the rules ofconductsetdowninthebody's code of ethics. In addition to the council, an advisory board established in 2011 raises awareness within the advertising industry about non-discriminatory portrayalsof sexes, largely asa means to counteract gender-discriminatory ads. The board reviews complaints about gender- discriminatory advertising and states its opinion to the council.

In 2013, the council received 209 complaints and issued 145 decisions; of the 209 complaints, nearly half were of gender discrimination (48%), followed by unethicality and immorality (16%), misleading and/or deceptive elements (8%), advertising with children or endangering children (7%), and violence (6%) (sterreichischer Werberat, 2014). That same year, the Austrian Advertising Council established a Young Council of 38 members, none of whom may serve as councilors beyond the age of 25 years. The 160 members of the Austrian Advertising Council and Young Advertising Council have backgrounds both in the advertising industry, either as representatives of media or advertising agencies, for example, or as clients, and in societal organizations (e.g., universities, law firms, chambers of commerce). All councilors use an online system for judging the cases that are presented to them. Each member decides whether he or she (a) sees no reason for intervention, (b) sees reason to notify the company to be more sensitive in its future advertising campaigns, or (c) sees reason to order a company to cease and desist the present campaign. In this sense, councilors

evaluate whether an advertisement is an offense against the rules of the Code of Ethics and may provide justification for any of their decisions.

The Austrian Advertising Council was able to provide us with data regarding all cases decided between October 2013 and May 2014. The dataset consists of 3,468 decisions regarding 27 cases that were made within this time period. As such, this study was to the best of our knowledge the first to rely on data from actual decisions made by advertising councilors regarding real advertisements that offended recipients. We thereby addressed a major concern raised by Eisend (2009), who identified the use of student samples and laboratory settingsas two principal biases in extant research onhumor in advertising. In our analyses, we scrutinized both the content of the advertisements and the councilors' decisions.

We performed content analysis upon each of the 27 advertisements in regards to its (a) categoryof ethics violation, (b) (intentional) use of humor, and (c) overt sexist tone and target (Table 1). The category of ethics violation was provided by the Austrian Advertising Council according to its ethics code that includes 20 rules. Both authors assessed and coded whether the potentially rule-breaking advertisement was humorous (1) or not humorous (0). In accordance with extantstudies on the useof intentional humor in advertisements (Beard, 2008a; Weinberger & Spotts, 1989), the codes were not identified from subjective impressions only but also from the statements of the complainants, the responses that the advertisers gave to the advertising council, and the description of the ads that were provided by the council. Percent agreement was 0.96, and the more rigorous Krippendorff's alphawas 0.92; we discussed a sole inconsistently coded case and ultimately coded it as not humorous. Accordingly, we coded advertisements for whether they were explicitly sexist (1) or not (0)that is, if they contained or indicated inappropriate sexual activity or if women's or men's bodies were explicitly sexually objectified or demeaned. Percent agreement was 1.0. We did not code type of humor in ads or whether the humor itself bore any sexist theme.

We additionally analyzed the decisions of councilors (n = 3,468) concerning the 27 cases, all of which we coded according to gender of the councilors (using their first names; 0 = male, 1 = female). In some cases, we could not clearly assign a gender to the councilor's names, and our analyses thus treats only 3,439 decisions upon applying this variable. In all, 1,547 decisions were made by women (45%) and 1,892 by men (55%). Using chi-square tests and ordinal regression analysis, we investi- gated the impact of humor on the decisions (i.e., masking effect) and the role of explicit sexism displayed in the advertisements, as well as the role of gender and of whether councilors were established (0) or young (1). The latter variable (age) was used as an additional control variable inthe ordinal regression model. The information concerning whether or not councilors belong to the established or young board was provided by the council.

To gain more profound insights into the councilors' decisions regarding humorous ads and to enrich the quality of the data, we examined the justifications for their rulings provided as open answers in their written decisions. In doing so, we focused on references to humorous elements inthe ads. Our qualitative analysis of these subjective views of councilors enriched our analysis, for these statements bear additional valuable information for both interpreting results and discerning whether humor influenced their decisions. Using a filter of keywords related to humor,¹weidentified 335 (24.8%) justifications among the 1,350 decisions that referred to humor in the advertisements.

Results

As will be shown below, Hypothesis 1about the general masking effect of humor on decisions about ethically questionable ads-and Hypothesis 4 -which argues that gender moderates the masking effectare not confirmed by this study, whereas our findings support Hypothesis 2that an amplifying effect triggers more rigorous decisions when humor and explicit sexism are presented together and Hypothesis 3that women councilors in general issue stronger decisions on offensive ads.

Our results reveal that 11 of the 27 cases analyzed contain (intentional) humor, while 16 are not considered to be humorous. Moreover, 16 ads discriminate against women and two objectify men. Four of these ads use humorous elements, and all four display women inappropriately. Table 1 shows the advertisements that we examined in our study, including the name of the ad, the use of humor,the complaint category, the display of overt sexism, the media in which the ad appeared, the final decision of the council (majority decision), and the number of total decisions issued in response to the ad.

To avoid misinterpretation: the following results are based on the councilors' individual decisions and not on the actual final majority decisions (see Table 1) about the ads. Our analysis of the councilors' decisions shows that humorous advertisements are less frequently assessed with orders to companies to discontinue ad campaigns (35%) than those without any humor (41%; ² = 33.15; df = 2, p < .001; Cramer's V = 0.10), which indicates the presence of humor's masking effect for the sample investigated. Humorous ads are also more frequently deemed inoffensive (29%; no reason for intervention) than non- humorous ones (20%). It is interesting that women councilors decided more often that an ad campaign had to be stopped (47%) than men councilors (32%; ² = 122.45; df = 2, p < .001; Cramer'sV= 0.19). The masking effect nevertheless appears for both women (² = 24.73; df = 2, p < .001; Cramer's V = 0.13) and men (² = 12.80; df = 2, p < .01; Cramer's V = 0.08). In the assessment of humorous ads, 43% of women decided to stop the respective campaign (versus 50% for nonhumorous ads), whereas 29% of men decided to stop humorous ads compared to 34% for nonhumorous ones.

The results further indicate that the use of explicit sexism impacts councilors' decisions (² = 535.707; df = 2, p < .001; Cramer's V = 0.393). As expected, overtly sexist ads are far likelier to be stopped (49%) than all other potentially offensive ads (17%). Further analysis shows that combining humorous and sexist elements in an ad prompts significantly stronger evaluations; regarding such ads, 61% of decisions favored stopping the ads compared to 46% for sexist ads without humor (n = 2,347 decisions concerning four ads that combine humor and sexism; ² = 35.936, df = 2, p < .001, Cramer's V = 0.124.)

We tested an ordinal regression model involving the primary effects of humor, sexism,and gender, as well as the control for age, to which we added interaction effects between humor and sexism, humor and gender, and sexism and gender (Table 2).

The results show that councilors evaluated ads using humor (p < .001) and sexism (p < .001) more rigorously. While this trend is unsurprising in the case of depicted sexism, we hypothesized that the effect of humor on councilors' decisions would work in the opposite direction. More specifically, women councilors evaluated ads more rigorously (p < .001), though age is irrelevant (p = .280). A significant interaction effect becomes apparent between humor and sexism (p < .001), and though it cannot be concluded whether humor moderates the sexism effect or sexism moderates the humor effect, the negative sign of the interaction term signals that ads including sexism without combining humor garnered milder decisions. Councilors evaluate ads that combine both elements significantly more often as being unethical than ads that do not combine the elements; this finding takes support from the crosstab shown in Table 3. Sexism and humor thus work together in a mutually reinforcing way in ads, which prompts councilors to issue more rigorous decisions concerning those ads.

In sum, our analysis confirms that the correlation between humor and decisions regarding the masking effect is spurious. In fact, the opposite is more accurate; humor in ads achieves an amplifying effect when presented with sexist elements. These findings correspond to Hypothesis 2, whereas the presence of humor's masking effect among advertising councilors remains unsupported (Hypothesis 1). At the same time, the findings also support a general gender effect on councilors' decisions, in which women councilors issued more rigorous decisions on offensive ads (Hypothesis 3). By contrast, since we found no evidence of any moderating role of the councilor's gender on humor's impact, we also rejected Hypothesis 4. In short, the gender of the councilor didnotmoderate the assessment of sexist ads.

In our content analysis of the councilors' open justifications, we found 78 statements referring to 11 humorous advertisements (Table 4). A considerable share of these arguments relate to humorous ads by the furniture retailer Mmax and two yogurt brands: NM FruFru and Mller Joghurt. In

Table 1. Advertisements included in the study.

Humor

Ad

Complaint category

Explicit sexism and target

Final decision of the council

Media

n

Humorous

Mmax

Ethics and morality

No

2 (sens.)

TV

141

Hornbach

Gender discrimination

Yes (w)

2 (sens.)

Print

136

Peithner KG

Health (deception and trivializing the taking of

No

2 (sens.)

Radio

126

medication)

SoulSista

Gender discrimination

Yes (w)

3 (stop)

Poster

120

ToolsonAir

Gender discrimination

Yes (w)

3 (stop)

Print

70

Der Mann

Gender discrimination

No

1 (no int.)

Print

65

KIA

Endangering children and young people

No

2 (sens.)

TV

147

Gutschein Walgauer

Gender discrimination

Yes (w)

2 (sens.)

Print,

141

Internet

NM FruFru

Ethics and morality

No

2 (sens.)

TV

141

Mller Joghurt

Endangering children and young people

No

2 (sens.)

TV

140

Schartner Bombe

Endangering children and young people

No

1 (no int.)

Internet

123

Not humorous

P&G "Danke Mama"

Gender discrimination

No

1 (no int.)

TV

132

Mdchenedition -Ei

Gender discrimination

No

1 (no int.)

TV

106

Mdchen-Flohmarkt (Disco-

Gender discrimination

Yes (w)

3 (stop)

Poster

143

Cabrio)

Jusfest

Gender discrimination

Yes (w)

2 (sens.)

Poster

163

OSR Schdlingsbekmpfung

Gender discrimination

Yes (m)

3 (stop)

Print

142

Klampfer Druck

Gender discrimination

Yes (w)

3 (stop)

Print

139

Platzer Transporte

Ethics and morality

Yes (w)

3 (stop)

Internet, Transport

153

Steirer TV

Gender discrimination

Yes (w)

3 (stop)

Poster

75

Freudenhaus Skiverleih 1

Gender discrimination

Yes (w)

2 (sens.)

Print

135

Freudenhaus Skiverleih 2

Gender discrimination

Yes (w)

2 (sens.)

Print

145

Mucha Herzbuben

Gender discrimination

Yes (m)

2 (sens.)

Print

151

Pfennigeralm

Gender discrimination

Yes (w)

2 (sens.)

Internet

144

Bet at home

Gender discrimination

Yes (w)

3 (stop)

Internet

134

College Invasion

Gender discrimination

Yes (w)

2 (sens.)

Poster

116

Badvergrerung

Gender discrimination

Yes (w)

2 (sens.)

Poster

127

Morning Glory

Gender discrimination

Yes (w)

2 (sens.)

Print

113

Note. Regarding targets of explicitly sexist advertisements, w = women and m = men. The final decision of the council is that of the majority of the councilors: 1 = "No reason for intervention";2

= "Call to be more sensitive in future advertising campaigns; 3 = "Stop immediately."

Table 2. Ordinal regression analysis with individual decision as dependent variable (Estimate).

Variable

Estimate

p

Humor (0 = Not humorous)

-0.605

Significant (p < .001)

Sexism (0 = Not explicitly sexist)

-1.968

Significant (p < .001)

Gender (0 = Male)

-0.807

Significant (p < .001)

Age (0 = Established)

0.089

N.s. (p = .280)

Humor Sexism

-0.757

Significant (p < .001)

Humor Gender

-0.044

N.s. (p = .802)

Sexism Gender

0.082

N.s. (p = .656)

Note. All predictor variables are coded to be dichotomous, and the parameters for the other categories are set to zero since they are redundant. Scale for dependent variable: 1 = "No reason for intervention";2= "Call to be more sensitive in future advertising campaigns";3= "Stop immediately" (n = 3,439; Cox and Snell's pseudo R² = .205, ² = 790.660, df = 7, p < .001; goodness-of-fit, ² = 45.876, df = 23, p < .01). Table and calculation contain the councilors' individual decisions and not the final decisions on ads.

Table 3. Comparison of individual decisions on ads using and not using humor or explicit sexism.

Decision

No explicit sexism/no humor

No explicit sexism/ humor

Explicit sexism/no humor

Explicit sexism/ humor

1 (no reason for intervention)

68.9%

39.3%

14.1%

8.4%

2 (call to be more sensitive in future

27.3%

39.5%

40.0%

30.6%

campaigns)

3 (stop immediately)

3.8%

21.2%

45.9%

61.0%

n

238

883

1,880

467

Note. ² = 663.931, df = 6, p < .001; Cramer's V = 0.309. Table and calculation contain the councilors' individual decisions and not the final decisions on ads.

what follows, we focus on our analysis of the justifications for decisions regarding these three television spots.

Mmax

This television spot presents a young man at a casting show stage who is so nervous that he wets his pants. The spot ends by showing the man drying his pants in his bathroom while he sings happily. The slogan reads, "Looks much better now" ("Sieht doch gleich besser aus"). The spot caused a slightly less-than-average number of open justifications (23%), though every second argument refers to humor explicitly; most underscore the humoristic tone of the spot and cite that as areasonfor their milder decisions: "The humor of the spot prevented me from admonishing it," "Clearly recognizable satire," or "I consider the Mmax advertisement as satire, and I think that there is

Table 4. Open justifications for the decisions on humorous ads.

Advertisement

Complaint category

Media

Number of decisions

Open justifications

Justifications with humor references

Mmax

Ethics and morality

TV

141

32

16

Hornbach

Gender discrimination

Print

136

37

6

Peithner KG

Health

Radio

126

28

0

SoulSista

Gender discrimination

Poster

120

35

1

ToolsonAir

Gender discrimination

Print

70

16

0

Der Mann

Gender discrimination

Print

65

20

5

KIA

Children and young

TV

147

33

2

people

Gutschein

Gender discrimination

Print,

141

30

5

Walgauer

Internet

NM FruFru

Ethics and morality

TV

141

49

30

Mller Joghurt

Children and young

TV

140

33

11

people

Schartner Bombe

Children and young

Internet

123

22

2

people

no reason for an intervention." Here, the masking effect of humor is not merely an automated or unconscious reaction; instead, the humor depicted itself serves as an apology for the decision. Other statements criticize this kind of humor: "This spot is extremely distasteful. Incontinence is not atopic that should be made fun of."

NM FruFru

This spot presents the slogan of the yogurt brand"Mixing because it's fun" ("Mischen wie es Spa macht"), which refers to the product's layer of fruit under plain yogurtby way of a nurse who "mixes" newborns at a hospital by switching their nametags. This spot caused the greatest number of open justifications (49; 35%) and of justifications referring to the intended humor (n = 30). The dominant strain of these arguments is that the theme of mixing up infants is offensive: "To use the nightmare of millions of parents for an advertising gag is cheap and disgusting," "Baby-switching is not funny or appropriate for an advertising spot," "There are topics where the fun stops," and "Although the satire can be seen immediately, unbelievable suffering occurs for parents when their babies are accidently switched." A few voices, though far fewer than in the case of Mmax, emphasize the satirical and thus acceptable character of the spot: "Because of the satirical depictionin the spot, I see no reason for an intervention" and "[It is a] clearly satirical exaggeration. Because of that, there is no reason for an intervention."

Mller

This spot opens with a family eating dinner. The daughter enters the room singing a well-known children's song and displaying her new tattoo that stretches across her belly. The mother takes up the same melody, suggesting that she should show the tattoo to her father, who is expected to slap her. The voiceover says, "What a little fruit!" ("Na, so ein Frchtchen!"), an ambiguous slogan that could be understood as a reference to the advertised fruit yogurt yet that also connotes something to effect of "What a rascal!" The television spot caused an average number of open justifications (n = 33;23%) among councilors, 11 of which explicitly mention humor. The argument of the councilors in this case is balanced; on the one hand, it criticizes the depicted threat of violence as intolerable, yet on the other sees that the humorous tone of the ad mitigates and even reverses the message: "It's noticeable to attentive viewers that this threat is not meant seriously in the way it is presented," "By contrast, the spot resolves the situation by showing the daughter eating a Mller Joghurt and by that

without getting slappedretaining the upper hand."

Discussion

As one of the first studies to investigate advertising councilors' decisions on real, offensive humorous and non-humorous advertisements, this article adds an applied ethics perspective tothediscussionof humor in advertising. In our content analysis of the advertisements that recipients found offensive and brought before the Austrian Advertising Council, we found that most offenses related to gender discrimination (21 of 27), followed by the endangerment of children and young people, ethics and morality in general, and health. Of these ads, 11 were intentionally humorous, while another 16 did not adopt a humorous tone; furthermore, 18 ads contained sexist elements (e.g., explicit sexual objectification). Concerning intentional humor in these ads, only four of the 11 humorous ads depicted sexism, though 14 of the 16 non-humorous ads did as well.

This finding can be understood in two ways. On the one hand, most of the advertisers do not consider that combining humor with sexual themes is a suitable advertising strategy. This sentiment, of course, could be a cultural peculiarity of Austria or even Europe, an aspect of humor inadvertising that requires further investigation, especially in other regions. The question of whether this ethical standard is common among advertisers across national borders as a professional norm is indeed a vital one. Such a circumstance would speak to a professional ethos of shared values affected also by increasingly transnational teams and the international experiences of respective players in the

markets. On the other hand, there are arguably fewer complaints about advertisements containing sexism and humor than about those with sexism only. This situation would indicate that the masking effect of humor deters complaints about advertisements, for audiences may feel less offended by ads that temper sexism with humor than those with no humor whatsoever. This possibly takes support from statements made by councilors arguing that humor reverses the original message by ridiculing it. This strategy, however, might not work for all kinds of offenses, as Beard (2008a) has convincingly shown in his investigation.

Yet, as our analysis of the councilors' decisions revealed, the masking effect we first found was spurious. This finding clearly does not support the assumption that the effect of humor can mask the unethicality of advertisements as assessed by advertising councilors. In this sense, the present study of advertising councilors' decisions does not confirm the masking effect. On the contrary, using humor may even increase the likelihood of a potentially offensive ad's being perceived to be in violation of ethical standards. It even seems that humor does not divert attention away from the offensive content. To add to the few studies of the masking effect of humor in advertising (Hsieh et al., 2010; Shabbir & Thwaites, 2007) that have analyzed the effect upon deceptive claims, our study has revealed that Austrian citizens complain primarily about offensive ads. Whereas discrimination and endangering children and young people are the top drivers of bringing a claim before the board, deception was the reason for only one complaint. Further studies should therefore investigate whether the masking effect of humor influences deceptive advertisements instead of other ethical violations in advertising.

The results further show that women opt significantly more often to stop advertising campaigns than men. This effect of gender is in line with previous findings (e.g., Waller, 1999) that women are more likely than men to feel offended by ethical violations in ads. Our finding thus suggests that, in advertising councils, a balanced ratio of the genders should be standard in order to give equal weight to women's interests and perceptions as that given to the interests of men. In fact, in our study we reveal a slight gender imbalance, for only 45% of the overall decisions by either the established or young council members were made by women. In this regard, however, a backlog persists in the Austrian Advertising Council, since within the established board only 36% of decisions were issued by women. A closer look at the Young Advertising Council, however, promises positive develop- ments regarding the future composition of the established council, for 75% of the decisions by young councilors were made by women.

Contrary to our expectations, the results of our ordinal regression analysis furthermore show that gender has not moderated the impact of humor on the decisions of councilors. In line with Beards's (2008a) results that, more than anything else, humor combined with inherently offensive themes precipitates complaints, our findings clearly show that sexism in humorous advertisements is considered to be inappropriate. This finding means that humor is unable to mask any form ofethical violation and, more importantly, that humor may increase the perception of an advertise- ment's being offensive when it contains explicit sexism. We thus contend that there is an amplifying effect when humor and sexist elements are combined. However, we are cautious in articulating this assumption, for our study contained only four ads in which humor and overt sexism are presented in tandem. We also bear in mind that our results are based on cases brought before an advertising standards board only; this circumstance is certainly a top limitation of our study, since the findings are based on advertisements that were known to offend people, who in turn complained about the offense to the advertising council.

References to humor occur to a different extent in the open justifications of the councilors' decisions. Yet, the statements regarding the three analyzed ad campaigns (Mmax, NM Frufru, and Mller) commonly articulate that the humorous content brought about milder decisionsregardingthe advertisements. Gulas and Weinberger (2006) similarly contended that "the fact that the ads are intended to be humorous is commonly used as a defense for their offensiveness" (p. 187). Accordingly, the councilors have argued that the humorous tone discouraged them from opting to order the company to cease and desist with the campaign. Other voices underscored how irony reversed or resolved the ethical violation depicted. The question here is whether an average audience

with a shorter attention span and less sensitization to the topic perceives ethical violations in the same way or, perhaps more likely, whether constant exposure to these kinds of representations perpetuates undesired stereotypes.

At the same time, another major part of the open justifications argues that humor leads to an opposite situation if its theme is perceived to be inappropriate, as was incontinence in the Mmax case, infant switching in the NM FruFru case, and the threat of children's being slapped in the Mller case. This trend is in line with other studies, which encourages us to believe that the occurrence and strength of the masking effect depends not only on the type of humor used (e.g., Buijzen & Valkenburg, 2004; Speck, 1991), but also on the degree of the offensiveness of the ad.

Implications for research and practice

The specific context of our studynamely, real advertisements brought before the Austrian Advertising Council and the respective decisions made by councilorsallows an investigation of humor's masking effect beyond laboratory settings, as well as permits a discussion of humor in advertising from the viewpoint of advertising ethics. The results of the study pose implications for all three levels of advertising critique suggested by P. Cunningham (1999).

First, when looking at the decisions of councilors active with the different boards and alliances, we naturally focus on individual decisions made at the micro level. It is of specific relevance to consider the findings of research on advertising's effects and to investigate whether they apply also to key players within advertising critique interest groups, for their decisions directlyaffect not only single ad campaigns, but also shared ethical standards in the advertising industry as a whole (i.e., the meso- level) and subsequently among the public (i.e., the macro level). It is thus also of specific importance to keep councilors informed and trained. In this sense, it is good news that our findings do not support the masking effect of humor among this group of recipients. Nevertheless, though persons active in advertising ethics and sensitive to the subject are not affected by humor or its masking effect, the effect may occur among average audiences, an idea that should be subjected to further investigation.

Second, at the group level, board members from the industry arguably mayalsoattemptto protect the interests of their specific sectors and thus tend to evaluate ads more mildly and less frequently opt to stop a campaign than board members from other areas of society. This assumption should be investigated in future research, since it could clarify whether different types of advertising councilors exist. The data provided to us for the present analyses did not contain information concerning board members' affiliations, which could have also impacted their interests in assessing the ads. Although councilors have to submit to the rules of the code of ethics of the Austrian Advertising Council, future analyses should scrutinize the underlying values and motives of counci- lors and their potential impacts on decisions. Applying the dichotomy between utilitarianism and social contract theory, the analytical distinction between two sorts of councilors can be made: Advertising councilors that wear the "veil of ignorance" (Rawls, 1971) or act as "principled advocates" (Baker, 2008) recognize their responsibility to all consumers and make their decisions from the perspective of the weaker parts of society such as children or minorities; instead, decisions by "pathological partisans" (Baker, 2008) or utilitarian pragmatists (A. Cunningham, 1999) are guided by utility, "leading to the mentality that the ends justify the means, however venal" (Baker, 2008, p. 243), and therefore they rather decide in favor of the advertising industry.

Thus, and third, future studies should also evaluate the nature of advertising councils and their decisions to ensure that they fulfill their responsibilities to the public (i.e., the macro level). According to A. Cunningham (1999), "[I]t is increasingly important that the power granted to business not be abused. In reaping the rewards of this system, advertisers should recognize their moral obligation to continue to support the free market and democratic process" (p. 93). By extension, in investigating advertising councilors' decisions regarding offensive ads and in

identifying potential threats, research and advertising practitioners can collaborate in support of advertising's role and function concerning the public at large.

The low number of cases analyzed in the present research is our study's chief limitation. Our results thus reveal tendencies about which inferential implications can be drawn only with due caution. Future research should involve larger samples of offensive advertisements, which would also allow the inclusion of more variables in the analysis. For example, the role of the type of gender discrimination (e.g., discrimination against men versus that against women) might affect decisions regarding the offensiveness of ads. In our study, all explicitly sexist and humorous ads sexually demeaned and objectified women. Future research should also consider the type of humor used (e.g., Beard, 2008a; Gulas & Weinberger, 2006) and how that relates to the offensiveness.

Considering advertising's function in society and its effects on perceived gender roles and norms, examinations of regulatory authorities and their work is of specific interest. The present study provides an important contribution to advertising ethics by investigating the use of humor in advertisements about which recipients complained, by examining the impact that humor has on advertising councilors' decisions, by analyzing the roles of sexism and councilors' gender in inter- preting these ads, and by discussing the consequences of our findings for advertising research and practice.

Note

1. The keywords used were lach*, witz*, humor*, satir*, spass*, spa*, lusti*, berspitz*, schmunzel*, komi*, comi*, persi*, ironi*

References

Alden, D. L., Mukherjee, A., & Hoyer, W. D. (2000). The effects of incongruity, surprise and positive moderators on perceived humor in television advertising. Journal of Advertising, 29(2), 1-16.

Baker, S. (2008). The model of the principled advocate and the pathological partisan: A virtue ethics construct of opposing archetypes of public relations and advertising practitioners. Journal of Mass Media Ethics: Exploring Questions of Media Morality, 23(3), 235-253. doi:10.1080/08900520802222050

Baker, S., & Martinson, D. L. (2001). The TARES test: Five principles for ethical persuasion. Journal of Mass Media Ethics: Exploring Questions of Media Morality, 16(2-3), 148-175. doi:10.1080/08900523.2001.9679610

Barnes, J. H., & Dotson, M. J. (1990). An exploratory investigation into the nature of affensive television advertising.

Journal of Advertising, 19(3), 61-69.

Beard, F. K. (2008a). Advertising and audience offense: The role of intentional humor. Journal of Marketing Communications, 14(1), 1-17.

Beard, F. K. (2008b). How products and advertising offend consumers. Journal of Advertising Research, 48(1), 13-21. doi:10.2501/S0021849908080045

Bless, H., & Schwarz, N. (1999). Sufficient and necessary conditions in dual-mode models: The case of mood and information processing. In S. Chaiken & Y. Trope (Eds.), Dual process theories in social psychology (pp. 423-440). New York, NY: Guilford Press.

Bonifield, C., & Cole, C. (2007). Advertising to vulnerable segments. In G. J. Tellis & T. Ambler (Eds.), The Sage handbook of advertising (pp. 430-445). London, England: Sage.

Breitmeier, B. G., & Ogmen, H. (2000). Recent models and findings in visual backward masking: A comparison, review, and update. Perception and Psychophysics, 62(8), 1572-1595.

Buijzen, M., & Valkenburg, P. M. (2004). Developing a typology of humor in audiovisual media. Media Psychology, 6

(2), 147-167.

Christians, C. G. (2007). Utilitarianism in media ethics and its discontents. Journal of Mass Media Ethics, 22(2-3), 113-131.

Chung, H., & Zhao, X. (2003). Humour effect on memory and attitude: Moderating role of product involvement.

International Journal of Advertising, 22(1), 117-144.

Cline, T. W., & Kellaris, J. J. (2007). The influence of humor strength and humor-message relatedness on ad

memorability. Journal of Advertising5, 536(1),-67.

Cunningham, A. (1999). Responsible advertisers: A contractualist approach to ethical power. Journal of Mass Media Ethics: Exploring Questions of Media Morality, 14(2), 82-94. doi:10.1207/S15327728JMME1402_2

Cunningham, P. (1999). Ethics of advertising. In J. P. Jones (Ed.), The advertising business (pp. 499-513). London, UK: Sage.

de Vries, D. A., & Peter, J. (2013). Women on display: The effect of portraying the self online on women's self- objectification. Computers in Human Behavior, 29(4), 1483-1489. doi:10.1016/j.chb.2013.01.015

Drumwright, M. E. (2007). Advertising ethics: A multi-level theory approach. In G. J. Tellis & T. Ambler (Eds.), The Sage handbook of advertising (pp. 398-416). London, England: Sage.

Drumwright, M. E., & Murphy, P. E. (2009). The current state of advertising ethics. Industry and academic perspectives. Journal of Advertising, 38(1), 83-107.

Eisend, M. (2009). A meta-analysis of humor in advertising. Journal of the Academy of Marketing Science, 37(2), 191-203. Fredrickson, B. L., & Roberts, T.-A. (1997). Objectification theory: Toward understanding women's lived experi- ences and mental health risks. Psychology of Women Quarterly, 21(2), 173-206. doi:10.1111/j.1471-6402.1997.

tb00108.x

Freedman, J. L., Sears, D. O., & Carlsmith, J. M. (1978). Social psychology (3rd ed.). Englewood Cliffs, NJ: Prentice Hall. Friend, C. (2004). Social contract theory. In J. Fieser & B. Dowden (Eds.), Internet encyclopedia of philosophy.

Retrieved from http://www.iep.utm.edu/soc-cont/

Fry, M.-L., & Polonsky, M. J. (2004). Examining the unintended consequences of marketing. Journal of Business Research, 57(11), 1303-1306.

Gass, R. H., & Seiter, J. S. (2003). Persuasion, social influence, and compliance gaining. Boston, MA: Allyn and Bacon.

Gelb, B. D., & Pickett, C. M. (1983). Attitude-towards-the-ad: Links to humor and to advertising effectiveness. Journal of Advertising, 12(2), 34-42.

Goldberg, M. E., & Gorn, G. J. (1987). Happy and sad TV programs: How they affect reactions to commercials. Journal of Consumer Research, 14, 387-402.

Grabe, S., Ward