Dana L Alden, Wayne D. Hoyer, & Choi Lee Identifying Global and …php.scripts.psu.edu/faculty/n/x/nxy906/COMPS... · 2008-03-05 · Dana L Alden, Wayne D. Hoyer, & Choi Lee Identifying

Dana L Alden, Wayne D. Hoyer, & Choi Lee

Identifying Global and Culture-Specific Dimensions of Humor in

Advertising: A MultinationalAnalysis

Humor is a commonly used communication tool in advertising in the United States, but U.S. marketersknow little about its use and effectiveness in foreign markets. Such limited knowledge hinders interna-tional managers' ability to determine which aspects of humorous communications are likely to be ame-nable to global standardization and which should be adapted to local expectations. The authors examinethe content of humorous television advertising from four national cultures: Korea, Germany, Thailand,and the United States. Findings indicate that humorous communications from such diverse national cul-tures share certain universal cognitive structures underlying the message. However, the specific contentof humorous advertising is likely to be variable across national cultures along major normative dimen-sions such as collectivism-individualism.

HUMOR is one of the most widely employed mes-sage techniques in modem American advertising.

Indeed, several of the most memorable American tele-vision advertising campaigns (e.g., "Bud Light" and"Joe Isuzu") have incorporated humor as a centralcomponent of their communication approach. Giventhe widespread use of humor by the advertising in-dustry, academic researchers in the United States havesought to identify potential benefits (e.g., reduced

Dana L. Alden is Assistant Professor of Marketing, University of Hawaiiat Manoa. Wayne D. Hoyer is Professor of Marketing, University of Texasat Austin. Choi Lee is Assistant Professor of International Business andMarketing, Hong Ik University, Seoul, Korea. The authors thank Profes-sor Hans Bauer and Mr. Harald Eisenacher at the Koblenz School ofCorporate Management, Koblenz, Germany, and Professor GuntaleeWechasara at the Faculty of Commerce and Accountancy, Chulalong-korn University, Bangkok, Thailand, for their assistance in the collectionof data. Thanks are also extended to Robert T. Green for comments ona previous version of the manuscript. Partial support for the project wasprovided by the Research Advisory Committee of the College of Busi-ness Administration, University of Hawaii and the Faculty Academic De-velopment and Research Committee of the Graduate School of Busi-ness, University of Texas at Austin.

counterargumentation and enhanced affect toward thead and brand; Scott, Klein, and Bryant 1990) as wellas potential problems (e.g., impaired recall; Gelb andZinkhan 1986) associated with the use of humor inadvertising.

Though preliminary evidence suggests that hu-morous advertising can be effective in foreign markets(cf. Weinberger and Spotts 1989), few studies havefocused on ways in which humorous content variesacross national cultures. Because information is lim-ited, it is unclear which aspects of humorous televi-sion advertising, if any, can be globally standardizedand which should be adapted to match local expec-tations. Seeking commonalities as well as differences,we attempt to identify dimensions in humorous tele-vision advertising that could be global as well as thosethat are likely to vary across nations by examiningadvertising from (former West) Germany, Thailand,South Korea (hereafter referred to as Korea), and theUnited States. We begin by reviewing relevant re-search from the domestic and international advertisingstreams.

64 / Journal of Marketing, April 1993Journal of MarketingVol. 57 (April 1993), 64-75

The Advertising Research Streamon Humor

Domestic Research

Communication managers in the United States havegenerally assumed that humor enhances advertising'seffectiveness (Madden and Weinberger 1984). To de-termine whether this assumption is true and, if so, why,advertising research in the U.S. has centered on threetopics: (1) analysis of humor effects on recall, eval-uation, and purchase intention (cf. Zhang and Zink-han 1991), (2) study of mediating factors such as rep-etition of the ad (Gelb and Zinkhan 1985), social settingin which the ad is viewed or heard (Zinkhan and Gelb1990), and prior attitude toward the brand (Chatto-padhyay and Basu 1989), and (3) examination ofwhether humor influences consumers more throughcognitive processes such as enhanced recall (Zhangand Zinkhan 1991) and reduced counterargumentation(Gelb and Zinkhan 1986) or through affective mech-anisms such as transfer of liking for the ad to the brand(Aaker, Stayman, and Hagerty 1986; Zinkhan and Gelb1990).

One overall conclusion drawn from these streamsis that humor is more likely to enhance recall, eval-uation, and purchase intention when the humorousmessage coincides with ad objectives, is well-inte-grated with those objectives, and is viewed as appro-priate for the product category. Under such circum-stances, humorous advertising is more likely to "secureaudience attention, increase memorability, overcomesales resistance, and enhance message persuasive-ness" (Scott, Klein, and Bryant 1990, p. 498; see alsoKrishnan and Chakravarti 1990).

Cross-National Research

Several cross-national studies of international adver-tising in general have been undertaken. For example,researchers have examined print and television ad-vertising from various national markets for similari-ties and differences in (1) levels and types of infor-mation (Dowling 1980; Hong, Muderrisoglu, andZinkhan 1987; Madden, Caballero, and Matsukubo1986), (2) reflection of cultural attitudes toward con-sumption (Mueller 1987; Tse, Belk, and Zhou 1989),and (3) portrayal of sex roles (Gilly 1988). With theexception of Tse, Belk and Zhou (1989), whose sam-ple did not include U.S. ads, these researchers foundsignificant differences between U.S. and foreign ad-vertising on key variables of interest.

Other researchers have documented ways in whichmultinational firms attempt to globally standardize ad-vertising. For instance, Peebles, Ryans, and Vemon(1978) distinguish between firms' use of "prototype"standardization (same ad with only translation and

necessary idiomatic changes) and "pattern" standard-ization in which the overall campaign is designed (e.g.,theme) for application in several national markets withsome adaptation of content and execution (Walters1986). Killough (1978) differentiates between "buy-ing proposals" that state the basic offer and "creativepresentations" that package the buying proposal. Onthe basis of reports from senior executives involvedin more than 120 multinational campaigns, Killoughconcludes that buying proposals can be used success-fully across cultures without modification more oftenthan creative presentations, which tend to interact withlocal cultural factors (see also Onkvisit and Shaw 1987).

Though such issues have been examined in a cross-national context, only one study appears to have lookedat the use of humor in advertising within other na-tional markets. Comparing television advertising inthe U.S. and the U.K., Weinberger and Spotts (1989,p. 39) report that a significantly greater percentage ofads in the U.K. (35.5% vs. 24.4%) were character-ized by humorous intent. In both countries, humor wasemployed most often with "low involvement/feelingproducts" and least often with "high involvement/feeling" products. Knowledge of such differences isclearly important. However, another dimension yet tobe examined may serve as a construct common to hu-morous advertising in multiple national markets. Thatdimension is the potential similarity in the cognitivestructures underlying humorous television advertisingfrom around the world.

Meyers-Levy and Tybout (1989) demonstrate theusefulness of the cognitive structure approach in a do-mestic consumer behavior context. They examined theevaluative effects of moderate differences or "incon-gruities" between new product information and infor-mation for the overall product category cued frommemory. They report that moderate incongruity fromexpectations produced more favorable evaluations ofnew product information than did congruity or ex-treme incongruity. As they note, "the very process ofresolving the incongruity is thought to be rewardingand thus may contribute to the resulting positive af-fect" (p. 40; see also Mandler 1982).

Cognitive principles similar to those described byMeyers-Levy and Tybout seem likely to be involvedin the structures and processing of humorous adver-tisements. More specifically, Meyers-Levy and Ty-bout's results appear to parallel those that would bepredicted by theories from psychology and linguistics.These theories specify incongruity and incongruityresolution as central to generating the positive affectthat often accompanies humor (cf. Herzog and Larwin1988). For example, Raskin's (1985, p. I l l ) theoryof humor argues that jokes often produce a mirthfulresponse by including cognitive, structural contrastsbetween expected and unexpected situations (e.g., in

Humor in Advertising / 65

the Bud Light beer commercial, the character arrivingwith a flashlight versus the other character exclaim-ing, "I said Bud Light!").

Structural analysis of humorous advertising fromseveral national cultures could help determine whethercognitive principles such as those hypothesized byRaskin are global or culture-speciflc. To explore po-tential applications of these theories to cross-nationalanalysis of humorous advertising, we now turn to thepsychological and linguistic literature on humor.

Psychological and LinguisticPerspectives on Humor

A major school of humor has focused on the cognitivestructures and processing pathways that are central toa humorous response (Herzog and Larwin 1988). Keyto such theories is the notion of incongruity or devia-tion from expectations. One group of theorists arguesthat incongruity is a necessary and sufficient conditionto produce humor (Suls 1983). In line with this po-sition, Nerhardt (1970) found that the greater the un-expected deviation from normally expected occur-rences, the greater the humor response. A second grouphypothesizes that incongruity alone is not always suf-ficient to produce a humor response. Rather, "ac-cording to this account, humor results when incon-gruity is resolved; that is, the punch line is seen tomake sense at some level with the earlier informationin the joke" (Suls 1983, p. 42). A key tenet of thisschool is that humor is a form of problem solving, asincongruity without resolution leaves listeners con-fused or frustrated because they do not "get the joke."

Not all problem solving, however, is humorous.Incongruity-resolution theorists suggest that a humor-ous response depends on (1) rapid resolution of theincongruity, (2) a "playful" context, that is, with cuessignifying that the information is not to be taken se-riously, and (3) an appropriate mood for the listener(Suls 1983). Support for this position has been pro-vided by several studies of different humorous stimuli(cf. Herzog and Larwin 1988; Oppliger and Sherblom1988; Wicker et al. 1981). Suls (1983) concludes thatboth incongruity and incongruity-resolution styles ofhumor exist, but that the latter predominates, partic-ularly for verbal humor. In addition, even researcherswho advocate greater integration of motivational andcognitive models of humor acknowledge a central rolefor incongruity theory within their proposed frame-work (cf. Kuhlman 1985).

From a linguistics perspective, Raskin (1985) sug-gests a script-based semantic theory, "a linguistic the-ory which interposes a cognitive step in the perceptionof what's funny" (MacHovec 1988, p. 92). This the-ory states that a verbal or written communication isconsidered a joke when the "text . . . is compatible

fully with two distinct scripts and the two scripts areopposite in certain definite ways such as good-bad,sex-no sex, or real-unreal." The third element, thepunchline, "switches the listener from one script toanother creating the joke" (Raskin 1985, p. 34-35).More often than not, according to Raskin, the hu-morous scripts will be opposite in terms of a "real"and an "unreal" situation. For example, consider thefollowing joke (Raskin 1985, p. 106).

An English bishop received the following note fromthe vicar of a village in his diocese: "My lord, I re-gret to inform you of my wife's death. Can you pos-sibly send me a substitute for the weekend?"

Here, the joke initially evokes the real situation ofa vicar wanting a substitute vicar because his wife hasjust died. The unreal script involves a vicar wantinga substitute wife. In addition, there is a playful op-positeness in the two scripts on which the humor turns.That opposition involves the contrast between the ex-pected "piousness" of a religious figure and the un-expected implied sexual interaction between the vicarand his "substitute" wife. Following the incongruity-resolution model, one can hypothesize that the incon-gruity of a vicar seeking a substitute "wife" is re-solved with the realization that the vicar undoubtedlymeans one thing but has inadvertently implied an-other.

According to Raskin, contrasts such as these canbe more finely categorized as (1) actual/existing andnonactual/nonexisting, (2) normal/expected and ab-normal/unexpected, and (3) possible/plausible andfully/partially impossible or much less plausible. Forthe first subtype, consider the preceding joke. Here,it is actually the case that the vicar wants a substitutefor himself, but it is not actually the case that the vicarwants a substitute for his wife. Thus the humorouscontrast involves actual versus nonactual. An exampleof an expected/unexpected contrast is seen in the fol-lowing joke.

A doctor tells a man, "Your wife must have absoluterest. Here is a sleeping tablet." "When do I give itto her?", the man asks. "You don't," explains thedoctor, "you take it yourself."

In this case, the contrast involves the normal orexpected action of a doctor prescribing medication foran ill person versus the abnormal or unexpected pre-scription for the healthy but talkative spouse. Play-fulness in the joke is captured in the contrast betweenthe expected care-giver role and the unexpected irri-tation-inducing role of the spouse.

The third contrast specified by Raskin involves apossible or plausible versus an impossible or muchless plausible situation. For example (p. 47):

Samson was so strong, he could lift himself by hishair three feet off the ground.

66 / Journal of Marketing, April 1993

Though it seems likely that Samson was strong enoughto lift another person off the ground, it would be im-possible for him to lift himself off the ground.

Raskin's theory can be interpreted within the in-congruity-resolution school of humor. First, as withincongruity-resolution theory, Raskin's theory positsa switch from "bona-fide" communication to a play-ful, nonthreatening mode (p. 140). Incongruity is thenestablished by the presence of two partially or fullycontrasting scripts that are compatible with the text,as discussed previously (i.e., possible/impossible, etc.).Finally, a "trigger, obvious or implied" (e.g., apunchline), helps the listener or reader resolve the in-congruity by fully realizing the oppositeness of thesituation. Thus, from the incongruity-resolution school,Raskin's theory can be supplemented with the hy-pothesis that the sudden realization of oppositenessquickly reduces the listener's felt tension and de-creases "arousal back to base-line," creating pleasurein the process (Suls 1983, p. 44).

Raskin's script-based humor theory along with in-congruity and incongruity-resolution theories couldprove helpful to understanding cognitive structures thatmay characterize humorous advertising around theworld. Though developed for verbal humor, Raskin'sscript-based semantic theory may well predict the typesof incongruent contrasts one is likely to find in hu-morous advertising, whether verbal or visual.

Application of Humor Theory in aCross-National Context

Cognitive factors underlying humorous communica-tion in the U.S. may also be found in humorous ad-vertising from other national markets. For example,cross-cultural researchers report evidence suggestinguniversal use of cognitive categories and summaryrepresentations for storage and application of the con-tinuous stream of information to which human beingsare exposed (e.g.. Pick 1980; Rosch 1977). Globaluse of such structures supports a central assumptionof the incongruity school of humor—that people de-velop expectations based on category norms that arecapable of being violated, sometimes in a humorousfashion.

Within the humor stream itself, several scholarsconclude that humor is indeed universal and that in-congruity is one of its central cognitive-structuralprinciples. Fry (1987, p. 68) notes that humor was apart of life in dynastic Egypt and that "contemporaryrecords in the Old Testament speak of laughter, joy,and amusement." As Berger (1987, p. 6) states:

Humor is . . . all pervasive; we don't know of anyculture where people don't have a sense of humor,and in contemporary societies, it is found every-where—in film, on television, in books and news-papers, in our conversations, and in graffiti.

Similarly, anthropologists have found that "jokingrelationships" involving "joking, teasing, banter, rid-icule, insult, horseplay, usually, but not always, in-volving an audience" are present in both traditionaland more industrialized societies (Apte 1983, p. 185).From the Amba people of southern Africa (Apte 1983)to machine operators in the United States (Fine 1983),such joking relationships appear to be commonplace.Joking behavior has even been observed in primates(Fry 1987).

Furthermore, "incongruent, outrageous or deviantmanifestations of personalities, behavior and so forth,are also important in such joking activities" (Apte 1983,p. 186). For example, evidence for the universal im-portance of incongruity in humor is found in "contrarybehavior" (e.g., "sitting on animals backwards whileriding"), which has been reported to be a major com-ponent of ritual humor among American Indians,tribespeople in Africa, and villagers in India (Apte1983, p. 190). Suls (1983) takes the argument a stepfurther when he states that most humor around the worldhas an incongruity-resolution structure. As evidence,he cites Shultz (1972), who examined verbal humorin the folklore literature of non-Western societies, andreports (p. 47):

The presence of incongruity and resolution featureswas found in the vast majority of materials (for ex-ample, of 242 Chinese jokes examined, 210 pos-sessed incongruity and resolution).

Hypotheses 1 and 2: GlobalPrinciples

The foregoing review suggests that humor is univer-sal. Furthermore, the cognitive-structural character-istic, incongruity, appears likely to be present in muchof the humor around the world. Hence, incongruitymay well be a major global component of humorousadvertising. Though Raskin's (1985) theory does notspecify whether the frequencies of his three hypoth-esized contrasts vary by national culture (i.e., whetherculture A's humor will emphasize expected/unex-pected contrasts whereas culture B's will emphasizereal/unreal), it appears to predict that the contrastswill be discernible in a given national culture's humorin some proportion. On the basis of this theory, wepropose our first hypothesis.

H]! Most television advertising from diverse national mar-kets in which humor is intended exhibits incongruentcontrasts.

In addition to establishing the presence of incon-gruity in humorous advertising, a goal of our study isto identify specific types of contrasts. If these con-trasts can be identified, researchers and practitionerswill better understand which aspects of the ad can bestandardized and how such standardization can occur.


For example, finding that none of the humorous adsfrom several national markets employ Raskin's pos-sible/impossible contrast would suggest that this formof incongruity may not work well in global advertis-ing campaigns that intend to be humorous. In contrast,if Raskin's theory is to be relevant to international ad-vertisers, one or more of the contrasts it predicts (i.e.,actual/nonactual, expected/unexpected, and possi-ble/impossible) should be identifiable in substantialnumbers in television advertising from several diversenational markets. Our second hypothesis proposes thatthe three specific contrasts predicted by Raskin areidentifiable in television advertising from differentmarkets.

H2: Across diverse national markets, three specific typesof contrasts (actual/not actual, expected/unexpected,and possible/impossible) are identifiable in televisionadvertising that is intended to be humorous.

Though we expect national markets to differ in theproportion of humor ads that stress one or more of thethree types of contrasts, there is little prior theory onwhich to base any related predictions. Therefore, ourinvestigation of differences in the relative use of thespecific contrasts is exploratory.

Hypotheses 3 and 4: Culture-Specific Differences

As McCracken (1986, p. 75) notes, "advertising is aconduit through which meaning constantly pours fromthe culturally constituted world to consumer goods."As a result, an important goal of advertising is to bringthe cultural world and the good together in a "specialharmony" that enables the viewer to see "this simi-larity and effect the transfer of meaningful properties"(McCracken 1986, p. 75). Because the "content ofads mirrors a society" (Tse, Belk, and Zhou 1989),one would expect actual message content (versus theunderlying structure) to reflect the culture in which itappears. Hence, despite possible similarities in hu-morous structures and principles across national cul-tures, significant differences seem likely to be foundin the situations, settings, and themes used to conveyhumor. Furthermore, such differences seem likely toreflect major national culture distinctions such as thosedocumented both in previous advertising research (e.g.,Gilly 1988; Mueller 1987; Tse, Belk, and Zhou 1989)and nonadvertising research (e.g., Hofstede 1983).

Hofstede (1983), for example, found that nationalcultures could be differentiated on several dimen-sions. Two of the dimensions he identified were"individualism-collectivism" and "power distance."Looking at the first dimension, Triandis et al. (1988)note that subordination of individual goals to the goalsof a few large in-groups is central to collectivist cul-

tures. Individualist cultures, in contrast, tend to becharacterized by multiple in-groups that are smallerand less demanding of their members. Reflecting thesedifferences, intended humor ads from cultures high incollectivism (e.g., Thailand and Korea) should in-volve larger groups of relatively close associateswhereas those from cultures low in collectivism (e.g.,Germany and the U.S.) should involve smaller groupsor no group at all. Therefore, we hypothesize:

H3: The number of individuals or characters playing majorroles in ads in which humor is intended is greater inhigh collectivism (low individualism) cultures than inlow collectivism (high individualism) cultures.

Hofstede's second dimension, power distance, in-volves the extent to which power within a nationalculture is unequally distributed (Ronen 1986). Na-tional cultures high on power distance tend to be hier-archic in their interpersonal relationships and decisionmaking whereas those low on power distance tend tobe more egalitarian. Advertising should differ on thisdimension, with high power distance cultures exhib-iting more relationships between characters that areunequal and low power distance cultures exhibitingmore relationships that are equal. Our fourth hypoth-esis follows.

H4: Relationships between central characters in ads in whichhumor is intended are more often unequal in high powerdistance cultures than in low power distance cultures,in which these relationships are more often equal.

Method

Sampling National Cultures

To improve reliability while enhancing generalizabil-ity, we chose two sets of countries that had similarcharacteristics within each set but differed between setson several important dimensions. The United Statesand Germany made up the first set. These nations aresimilar inasmuch as both are Western, developed na-tions having high scores on Hofstede's (1983) indi-vidualism-collectivism dimension (i.e., low on col-lectivism) and low scores on the power distancedimension (see Table 1). The second set, Korea andThailand, are similar to each other yet different fromthe first set inasmuch as both are Asian, rapidly de-veloping nations having high scores on collectivismand high scores on power distance. Though hypoth-esized cognitive-structure principles (H, and H2) werepredicted to hold across all four nations, the contentof humorous advertising was expected to differ be-tween the matched pairs of nations on Hofstede's in-dividualism and power distance dimensions (H3 andH4).


TABLE 1Scores and Ranks of Countries on Collectivism

and Power Distance Dimensions'

United StatesGermanyThailandKorea

Power

Score

40356460

Distance

Rank"

16103127

Individualism/Collectivism

Score

91672018

Rank

50361311

•Based on Hofstede (1983)."Rank is based on 50 countries in sample; for example, U.S.is 16th from the lowest on power distance and Thailand is 31stfrom the lowest. Germany is 6th highest on individualism butonly 14th on collectivism whereas Korea is the 11th highest onindividualism but 39th from the lowest on collectivism.

Sampling Ads Within Country

Randomized cluster samples of national brand TV adsshown on major networks in each country were col-lected. Local advertising and duplications of nationalbrand ads were eliminated along with ads that con-tained more than 50% sales promotion information(e.g., a tie-in promotional ad for Pepsi with a localsupermarket). Ads for the same brand that differed in50% or more of the content remained in the sample.As other sampling plan details varied slightly amongcountries, we describe each plan.

Sampling in the U.S. was conducted over threedays in early November 1990. On each day (randomlychosen), one of the three major privately owned na-tional networks (randomly assigned to each day) wasrecorded (6 a.m. to midnight). All ads were thenlogged. We obtained a total of 497 unduplicated adsfor national brands from the three major Americannetworks. In Thailand and Korea, a similar procedurewas followed. However, in those countries, the tapesand master lists of cluster-sampled ads (three days fromthree stations with days randomly selected) were ob-tained from market research firms that monitored TVand radio advertising. Also in both countries, one ormore government-owned stations that carried ads fornational brands were included. In Thailand, ads wererecorded during April 1990, resulting in 351 undu-plicated ads. In Korea, recording was done duringFebruary 1990, resulting in 520 unduplicated ads.

Finally, in Germany, the two major national tele-vision channels are strongly regulated by the govern-ment (Clemens 1987). As a result, advertising on thesechannels is very limited in terms of frequency andcontent. With the advent of cable television, however,German viewers are now exposed to a wider varietyof programming options and advertising (Clemens1987). Therefore, to provide a representative sampleof German advertising, three privately owned and op-erated channels that carry ads for national brands were

sampled over a three-day period during October 1990.As in the United States, a log of all advertising wascreated. Duplicates, promotional ads, and local brandads were eliminated, leaving a total of 244 ads foranalysis.

Humorous Ad Identification

Three native, bilingual coders in Germany and fourin each of the other countries were instructed in theirown language and in English on how to rate ads interms of humor. Following Weinberger and Spotts(1989, p. 40), we did not ask judges to determinewhether they personally felt the ad was humorous; in-stead, the humorous intent of the ad was coded in aneffort to reduce subjectivity. In other words, becausecertain ads seem likely to appeal more to specific seg-ments of the culture than to others, coders did notjudge how funny each ad was, but only whether hu-mor was intended.

An ad was assumed to contain intended humor whenat least three coders agreed. In all countries, inter-judge agreement (calculated as the percentage of threeor more agreements that humor either was or was notintended) exceeded 80% (cf. Sujan 1985). In the U.S.,80 ads were judged by three of four coders to containintended humor. In Germany 48 ads and in both Thai-land and Korea 51 ads were judged by three or morecoders to contain intended humor. To reduce the sub-sequent in-depth coding task for the U.S. sample, onlyads on which all four coders agreed were used, re-sulting in 52 ads for analysis.

In-Depth Coding Procedures

Three new native coders used a standard coding formto evaluate intended humor ads in each country. Allcoders received extensive training prior to the actualcoding task. Much of this training was conducted inthe coders' native languages, though foreign research-ers were usually present. Originally written in En-glish, coding forms were subsequently double back-translated to assure maximal equivalency, except inGermany where the coders' high English proficiencyallowed use of the original forms. Each intended hu-mor ad was viewed two to three times and coders thenindependently evaluated the ad. Subsequent viewingwas allowed when coders had questions about the ad'scontent, style, or some other aspect. The coding formstook 10 to 15 minutes to complete for each ad. Forall items in all country samples, interjudge agreementexceeded 85%. Disagreements were resolved amongthe coders without the involvement of the investiga-tors beyond simple clarification of coding guidelines.

Measures

First, coders were asked to indicate (yes/no) whetherthe ad contained any of the contrasts such as those


specified by Raskin (1985)—actual/nonactual, ex-pected/unexpected, or possible/fully or partially im-possible. Second, coders were asked to determine whichspecific contrasts were present in the ad. In addition,when more than one contrast was identified in the ad,coders were asked to determine which contrast wasmost emphasized (i.e., the main type). These mea-sures were used to evaluate the study's two global hy-potheses (Hi and H2).

Next, coders were asked to determine how manypeople or characters were featured in the ad. To beincluded, characters had to be actively involved in thead's plot. For example, if the story was set in a su-peimarket checkout stand and involved interaction be-tween the shopper and the checker, with other indi-viduals walking by in the background but not interactingwith the main characters, the coders would indicatethat two characters were featured in the ad. On thebasis of Triandis et al.'s (1988, p. 325) statement that"the emphasis is usually on people more than on taskin collectivist cultures; and the reverse happens in in-dividualist cultures," we operationalized H3 by as-suming that a larger proportion of ads in collectivistcultures (Korea and Thailand) would feature three ormore characters whereas a larger proportion in indi-vidualist cultures (U.S. and Germany) would featuretwo or fewer characters.

Finally, coders indicated whether the people orcharacters featured in the ad were portrayed as havinggenerally equal or unequal status. This variable wasdefined to include explicit and/or implicit differencesbetween two or more individuals or characters in termsof age, wealth, education, power, and general knowl-edge. In ads featuring one individual or character, sta-tus differences between individuals or characters wereassumed to be absent. Following Hofstede (1983), weoperationalized H4 by predicting that cultures with highpower distance scores (Korea and Thailand) would havea larger proportion of humorous ads containing indi-viduals of unequal status than cultures with low powerdistance scores (U.S. and Germany), which would havea larger proportion of humorous ads containing indi-viduals of equal status.

ResultsOur study strongly supports both H| and H2 on theglobal presence of incongruent contrasts in humoroustelevision advertising. In all four countries, a majorityof the humorous television ads contained one or moreincongruent contrasts such as those described in Ras-kin's (1985) theory. As shown in Table 2, higher pro-portions of ads with contrasts were found in Germany(92%) and Thailand (82%) than in the United States(69%) and Korea (57%). Chi square tests for differ-ences in the proportions of ads with any one of the

TABLE 2Number of Ads With Any of Three Contrasts by

Country

U.S. Germany

Any of Three Contrasts Present"NumberPercent"

No ContrastNumberPercent

Total

3669

Present163152

4492

48

48

Thailand

4282

91851

Korea

2957

224351

°Ads may have more than one contrast."Percentages rounded.

TABLE 3Types of Humor Contrasts by Country

U.S.

Actual/NonactuaPNumber 4Percent" 7

Expected/UnexpectedNumber 26Percent 48

Possible/ImpossibleNumber 24Percent 44

Germany

814

4071

814

Total Number of Contrasts54 56

Thailand

48

2856

1836

50

Korea

412

1648

1339

33

°Ads may have more than one type of contrast.""Percentages rounded.

three contrasts are significant (p < .05) for three ofsix paired nation comparisons: United States versusKorea, Germany versus Thailand, and Korea versusThailand. Despite such differences, the fact that al-most 60% of the humorous ads in all four nations con-tain contrasts such as those specified by Raskin sug-gests that incongruent cognitive structures may bepresent globally in a majority of television ads thatare intended to be humorous.

In support of H2, the specific types of contraststheorized by Raskin were identified in all four coun-tries (see Table 3). Interestingly, there is relativelylittle variation in the proportions of contrast types acrossnational cultures. For example, in all four countrysamples, 15% or fewer of the contrasts were found tobe actual/nonactual. Though in the German sample70% of contrasts were expected/unexpected, the per-centages of expected/unexpected and possible/im-possible were fairly consistent and evenly divided acrossthe three other country samples. Thus, advertising inwhich humor is intended appears to feature greaterproportions of certain types of contrasts. Furthermore,these proportions may be relatively consistent acrossnational cultures, though some variation undoubtedlyis present.

Pattems similar to those just discussed emerge when


one looks at the main type of contrast in ads with oneor more contrasts (see Table 4). First, very few ac-tual/nonactual contrasts were identified as the maincontrast type in each country (i.e., fewer than 10%).Second, distributions of the main types of humor con-trast in the U.S., Thai, and Korean ads do not varysignificantly (p > .96). However, overall, the distri-bution of the main type of contrast across all fourcountry samples is not independent of country (x^ [6]= 21.61, p < .001). This result appears to be due tothe fact that substantially more expected/unexpectedcontrasts (84.1%) constituted the main type of con-trast featured in the German ads. There is a fairly evensplit between expected/unexpected and possible/im-possible main contrasts in the three other national cul-ture samples.

In addition to supporting H| and H2, our findingsprovide evidence on behalf of H3. As seen in Table5, the two nations high on Hofstede's collectivism di-mension (Korea and Thailand) had a substantial num-ber of humorous ads with three or more central char-acters whereas the two that were low (U.S. andGermany) had substantially fewer ads with three ormore characters. Within each matched country sam-ple, there was no significant difference between theproportion of ads with three or more and the propor-

TABLE 4Main Types of Contrasts in Ads by Country"

U.S.

Actual/NonactualNumber 2Percent" 6

Expected/UnexpectedNumber 16Percent 44

Possible/ImpossibleNumber 18Percent 50

Germany

49

3784

37

Number of Main Contrasts36 44

Thailand

25

2252

1843

42

Korea

27

1448

1345

29

•Include ads with one or more contrasts.""Percentages rounded.

tion with two or fewer characters (for Korea and Thai-land, p > .17; for Germany and the U.S., p > .82).Combining the results for Korea and Thailand, we seethat 75% of the sampled ads contained three or morecharacters and 25% contained two or fewer. In con-trast, only 26% of the ads in the combined sample forGermany and the U.S. had three or more charactersand 74% had two or fewer. Chi square analysis in-dicates that the relationship between differences in thenumbers of characters and the matched country pairsis significant (x^ [1] = 47.53, p < .001).

Finally, H4 is supported. As shown in Table 6,more humorous ads featured unequal status betweenmain characters in the two nations that are high onHofstede's power distance dimension (Thailand andKorea) than in the two that are low on that dimension(U.S. and Germany). In both Thailand and Korea, morethan 60% of the humorous ads portrayed characters ofunequal status and there is no significant differencebetween the two nations on this variable (p > .68).

Differences in the distributions of equal and un-equal status ads are significant for Germany and theU.S. (x^ [1] = 9.32, p < .003). These differencesappear to result from the fact that the percentage ofads judged to contain equal status characters was largerfor Germany (85%) than for the U.S. (58%). How-ever, a majority of ads in both countries featured equalstatus characters. Thus, the two country samples werecombined and results at the aggregate level were com-pared with those for Thailand and Korea.

Combining the results for Korea and Thailand, wefound that 63% of the sampled ads featured relation-ships between characters that were unequal. In sharpcontrast, for the combined U.S. and German sample,71% of the ads featured equal status relationships. Chisquare analysis indicates that the relationship betweenstatus and the combined country pairs is not indepen-dent (x^ [1] = 23.15, p < .001).

Last, the content differences found may be due inpart to differences in products advertised. To test thispossibility, three native coders in each country cate-gorized featured products along three dimensions: (1)

TABLE 5Numbers of Individuals Featured in Television

Ads

U.S. Germany Thailand Korea

TABLE 6Status of Individuals or Characters in Television

Ads

U.S. Germany Thailand Korea

Ads With TwoNumberPercent

or Fewer3873

Ads With Three or MoreNumberPercent

Total Number

1427

of Ads52

3675

1225

48

1020

4180

51

1631

3569

51

Ads WithNumberPercent^

Ads WithNumberPercent

Equal Status3058

Unequal Status2242

Total Number of Ads52

4185

715

48

2039

3161

51

1835

3365

51

'Percentages rounded. 'Percentages rounded.


avoidance nondurable (goods associated with unde-sirable events and activities such as aspirin, cf. Stay-man, Aaker, and Bruzzone 1989), approach nondur-able (goods associated with desirable events andactivities), durable, or service; (2) high, medium, orlow involvement, and (3) pleasure-oriented, function-oriented, or both. Interrater agreement was again 80%or higher in each country and disagreements were re-solved through discussion. Though certain significantdifferences (p < .05) on each product dimension arepresent across countries, the differences are not sub-stantial. For example, in each country, consumer ap-proach nondurable goods comprise the modal cate-gory (i.e., 50% or higher). In addition, lowerinvolvement products are predominant (i.e., 50% orhigher) in every nation's sample, and in three of fourcountries the modal category consists of pleasure-ori-ented products. Thus, though differences in producttypes featured in the ads may have contributed to ob-served differences in terms of the collectivism-indi-vidualism and status measures, it appears unlikely thatproduct-related differences could account for the ex-tent of systematic variation in advertising content doc-umented in our study.

DiscussionThe purpose of our study was to examine the natureand content of humorous appeals across national cul-tures. The first goal was to identify an underlying globalprinciple that might provide a deeper understandingof the basic structure of humorous appeals. Second,we attempted to identify dimensions along which na-tional cultures might meaningfully differ in terms ofspecific television advertising content.

In reference to the first goal, coding results forhumorous ads in four different cultures (Thailand,Korea, Germany, and the U.S.) lend support to Ras-kin's (1985) script-based semantic theory, which fo-cuses on incongruity of scripts as the central elementof humor. In all four cultures, a majority of the adswere classified as containing incongruent cognitivestructures. In two countries, Thailand and Germany,the presence of such contrasts was even significantlyhigher (82% and 92%, respectively). These findingsextend Suls' (1983) conclusion that humor globallyexhibits incongruity to advertising.

The implications of this finding are twofold. First,from a theoretical perspective, it appears that the basiccognitive structure approach underlying humorous ap-peals may not be "culture-bound." Thus, our studysupports and adds to a growing body of research in-dicating that certain aspects of consumer cognition are"universal" (Pick 1980; Rosch 1977). More impor-tant, the seemingly global applicability of the incon-gruity principle suggests that a cognitive structure ap-

proach has the potential to provide a more basicunderstanding of how humorous appeals operate acrossdifferent cultures.

Second, from a global strategy perspective, ourstudy may provide valuable insights to attempts tostandardize humorous appeals across national cul-tures. Specifically, our results suggest that ads con-structed in line with the incongruity and/or incongru-ity-resolution principles may have the ability to generatehumor in diverse national cultures. However, our studyexamined only the presence of incongruent structuresin ads classified as humorous in intent. The study didnot address the issue of the effectiveness of the adsin generating a humorous response. Hence, an im-portant topic for future research would be to examinethe extent to which ads containing "incongruent" ver-sus "incongruent with resolution" cognitive structuresare more or less effective in generating desired affec-tive responses.

It is also interesting that very few actual/nonac-tual contrasts were found in any country. Thus a sub-stantial portion of humorous advertising around theworld appears to rely on expected/unexpected or pos-sible/impossible contrasts. Furthermore, for the UnitedStates, Thailand, and Korea, distributions of these twocontrast types are very similar, with proportions forboth types ranging from approximately 40% to 50%of the total number of main contrasts. The Germanresults are distinct in that so many of the ads (morethan 80%) featured expected/unexpected contrasts.Unfortunately, present theory and research are inad-equate to explain this difference. Similarly, from atheoretic perspective, it is not clear why humorous ad-vertising in Korea exhibited fewer contrasts than thatin any other country. Such unanswered questions clearlyindicate a need for future research into possible cul-tural factors that might relate to a preference for cer-tain types of contrasts as well as humor without con-trasts.

The second goal of our study was to identify di-mensions along which the specific content (as op-posed to stmcture) of the humorous appeals might varyacross national markets. Two dimensions from Hof-stede's (1983) work were chosen for hypothesis de-velopment: collectivism-individualism and powerdistance. The collectivism-individualism dimensionpertains to how much the national culture emphasizeseither the subordination of individual goals to the goalsof a few large groups or the association of the indi-vidual with multiple smaller groups that tend to beless demanding. Results support the hypothesis thatads in countries high on collectivism (Korea and Thai-land) contain more group-oriented situations than adsin countries that stress individualism (Germany andthe U.S.). The suggestion is that consumers in col-lectivist societies are more likely to respond to hu-


morous appeals involving large groups than con-sumers in individualist cultures, where use of smallergroups or individuals may be more appropriate.

Power distance was the second national culture di-mension examined. Cultures high on power distancetend to be hierarchic whereas cultures low on powerdistance tend to be more egalitarian. We examinedthis dimension by looking at the status relationshipsbetween the characters in the ad. The results supportour hypothesis that cultures high on power distance(Thailand and Korea) have more ads with charactersof unequal status than countries low on this dimension(Germany and the U.S.). This finding is refiective ofthe fact that in the U.S. and Germany, more emphasisis placed on equality than is the case in the Asian cul-tures studied.

Though our findings suggest that it may be pos-sible to standardize the basic structure of humorousappeals by employing the incongruity principle, thead's content (i.e., numbers of characters and their sta-tus relationships) may require modification for mar-kets with widely varying values and norms. Hence,our findings are consistent with previous research onthe ways in which multinational firms standardize ormodify advertising across national cultures (Killough1978; Peebles, Ryans, and Vemon 1978). In partic-ular, the results are similar to Killough's (1978) find-ings about the possibilities of standardizing an ad's"buying proposal" (i.e., a structural characteristic akinto the incongruity principle of humor found in ourstudy) but culturally adapting the ad's "creative pre-sentation" (i.e., a thematic content characteristic sim-ilar to differences in the numbers of individuals andtheir status relationships).

Our findings also support Onkvisit and Shaw's(1987, p. 54) conclusion that standardized advertisingstrategies incorrectly assume that communications de-signed for the U.S. market can be used abroad with-out modification. Those authors describe such ap-proaches as "ethnocentric" and argue that they are likelyto be unsuccessful for any one of the following rea-sons: (1) failure to gain attention if the message isirrelevant, (2) failure to be understood, or (3) failureto motivate action. Instead of a globally standardizedadvertising approach, Onkvisit and Shaw recommenda "geocentric" strategy that requires (p. 54):

. . . the advertisement to be designed for the world-wide audience from the outset to appeal to a sharedintercountry denominator while allowing for somemodification to suit each market.

Our Study identifies a potentially important "in-tercountry denominator" in humorous advertising. Thatis, incongruent cognitive structures were found in amajority of television advertising that was intended tobe humorous despite significant differences on major

cultural dimensions. Thus, following Onkvisit andShaw, we suggest that these structures may provide adimension on which to standardize because their useis likely to enhance the communication effectivenessof humorous advertising around the world. At the sametime, the study demonstrates that advertisers are likelyto benefit from "allowing for some modifications tosuit each market" (Onkvisit and Shaw 1987, p. 54).

Finally, while remaining alert to differences, man-agers may be able to group countries strategically onthe basis of national culture dimensions. For example,our findings suggest that appeals emphasizing collec-tivism and unequal status relationships are much morefrequent in Thailand and Korea (and perhaps othercollectivist national cultures) than in more individu-alistic cultures such as the U.S. and Germany. Futureresearch should test the extent to which such valuedimensions may assist development of standardizedhumorous appeals for various groupings of nationalcultures.

LimitationsThe many difficulties associated with cross-nationalresearch have long been recognized (Albaum and Pe-terson 1984) and some of these problems are presentin our study. For example, one of the key challengesin cross-cultural data collection is the attainment ofmeasurement equivalency (Hui and Triandis 1985). Inother words, the data collection is valid only to theextent that the researchers can demonstrate that theconstructs and measures are conceptually and opera-tionally equivalent across the various cultures studied.

In our study, we devoted extreme effort to ensur-ing that the coders clearly understood the constructsand categories and could demonstrate the ability tomake judgments as we intended. Nevertheless, cul-tural biases inherent in the coders could have some-what infiuenced the results and accounted for some ofthe variance between countries. The differences ob-served, however, were on culturally sensitive mea-sures (e.g., individualism-collectivism) and were typ-ically very large. Therefore it is unlikely that all ofthe meaningful variance in categories is attributableto coder biases. Second, the coders were all extremelywell trained and exhibited high interrater agreement.

We also recognize that the samples studied rep-resent only a portion of the ads that appear on tele-vision in the respective countries. Future research em-ploying larger numbers of ads is needed to verify ourfindings. Finally, we examined only one medium (i.e.,television). Future research should examine humorousappeals in other media as well because the culturesstudied differ in terms of exposure to ads in the var-ious media. For example, in Germany, print adver-tising (in particular magazines, newspapers, and bill-


boards) plays a more significant advertising role thanit does in the U.S. (Toyne and Walters 1989).

ConclusionOur study makes an important contribution by dis-covering that humorous television advertising in fournational cultures employs incongruent cognitive struc-tures. Furthermore, the incongruity in humorous ad-vertising was found in every culture to be expressedin identifiable structures such as the expected/unex-pected contrast specified in Raskin's (1985) model.Message content in humorous advertising, however,appears to vary along major national culture dimen-sions such as collectivism/individualism and powerdistance documented by Hofstede (1983). As a result,though the notion of incongruent contrast structures

may ultimately prove capable of serving as a guide tothe development of globally standardized communi-cations, certain aspects of the ad's message may con-tinue to benefit from adaptation to the targeted na-tional culture.

Finally, we emphasize again that our study did notaddress the issue of ad effectiveness. In the future,relationships between use of altemative cognitivestructures (e.g., expected/unexpected vs. possible/impossible) and effectiveness in generating desiredhumorous responses should be examined. For exam-ple, an important question is whether and when in-congruity or incongruity with resolution is more ef-fective as a humorous communication strategy.Regardless, the search for global and culture-specificprinciples in international marketing communicationsappears to be a promising area for future research.

REFERENCESAaker, David A., Douglas M. Stayman, and Michael R. Hag-

erty (1986), "Warmth in Advertising," Journal of Con-sumer Research, 12 (March), 365-80.

Albaum, Gerald and Robert A. Peterson (1984), "EmpiricalResearch in International Marketing," Journal of Interna-tional Business Studies, 15 (Spring/Summer), 161-73.

Apte, Mahadev L. (1983), "Humor Research, Methodologyand Theory in Anthropology," in Handbook of Humor Re-search, Paul F. McGhee and Jeffery Goldstein, eds. NewYork: Springer-Verlag Inc., 183-212.

Berger, Arthur A. (1987), "Humor: An Introduction," Amer-ican Behavioral Scientist, 30 (1), 6-16.

Chattopadhyay, Amitava and Kunal Basu (1989), "Prior BrandEvaluation as a Moderator of the Effects of Humor in Ad-vertising," Journal of Marketing Research, 26 (Novem-ber), 466-76.

Clemens, John (1987), "Television Advertising in Europe: TheEmerging Opportunities," Columbia Journal of WorldBusiness, 22 (Fall), 35-41.

Dowling, Grahame R. (1980), "Infonnation Content in U.S.and Australian Television Advertising," Journal of Mar-keting, 44 (Fall), 34-7.

Fine, Gary Alan (1983), "Sociological Approaches to the Studyof Humor," in Handbook of Humor Research, Paul F.McGhee and Jeffery Goldstein, eds. New York: Springer-Verlag Inc., 159-81.

Fry, William F., Jr. (1987), "Humor and Paradox," AmericanBehavioral Scientist, 30 (1), 42-71.

Gelb, Betsy D. and George M. Zinkhan (1985), "The Effectof Repetition on Humor in a Radio Advertising Study,"Journal of Advertising, 14 (4), 13-20.

and (1986), "Humor and AdvertisingEffectiveness After Repeated Exposures to a Radio Com-mercial," Journal of Advertising, 14 (4), 15-20.

Gilly, Mary C. (1988), "Sex Roles in Advertising: A Com-parison of Television Advertisements in Australia, Mexico,and the United States," Journal of Marketing, 52 (April),75-85.

Herzog, Thomas R. and David A. Larwin (1988), "The Ap-

preciation of Humor in Captioned Cartoons," Journal ofPsychology, 122 (6), 597-607.

Hofstede, Geert (1983), "National Cultures in Four Dimen-sions," International Studies of Management and Organi-zation, 13 (2), 52.

Hong, Jae W., Aydin Muderrisoglu, and George M. Zinkhan(1987), "Cultural Differences and Advertising Expression:A Comparative Content Analysis of Japanese and U.S.Magazine Advertising," Journal of Advertising, 16 (1), 55-62.

Hui, C. Harry and Harry C. Triandis (1985), "Measurementsin Cross-Cultural Psychology," Journal of Cross-CulturalPsychology, 16 (2), 131-52.

Killough, James (1978), "Improved Payoffs From Transna-tional Advertising," Harvard Business Review, 56 (July-August), 102-10.

Krishnan, H. Shanker and Dipankar Chakravarti (1990), "Hu-mor in Advertising: Testing Effects on Brand Name andMessage Claim Memory," Proceedings of the Summer Ed-ucators' Conference, William Bearden and A. Parasura-man, eds. Chicago: American Marketing Association, 10-16.

Kuhlman, Thomas J. (1985), "A Study of Salience and Mo-tivational Theories of Humor," Journal of Personality andSocial Psychology, 49 (1), 281-6.

MacHovec, Frank J. (1988), Humor: Theory, History, Appli-cations. Springfield, IL: Charles C. Thomas Publisher.

Madden, Charles S., Marjorie J. Caballero, and Shinya Mat-sukubo (1986), "Analysis of Information Content in U.S.and Japanese Magazine Advertising," Journal of Advertis-ing, 15 (3), 38-45.

Madden, Thomas J. and Marc G. Weinberger (1984), "Humorin Advertising: A Practitioner View," Journal of Advertis-ing Research, 24 (4), 23-9.

Mandler, George (1982), "The Structure of Value: Accountingfor Taste," in Affect and Cognition: The 17th Annual Car-negie Symposium, Margaret S. Clark and Susan T. Fiske,eds. Hillsdale, NJ: Lawrence Erlbaum Associates, 3-36.

McCracken, Grant (1986), "Culture and Consumption: A The-


oretical Account of the Structure and Movement of the Cul-tural Meaning of Goods," Journal of Consumer Research,13 (June), 71-84.

Meyers-Levy, Joan and Alice Tybout (1989), "Schema Con-gruity as a Basis for Product Evaluation," Journal of Con-sumer Research, 16 (June), 39-54.

Mueller, Barbara (1987), "Reflections of Culture: An Analysisof Japanese and American Advertising Appeals," Journalof Advertising Research, 27 (June/July), 51-9.

Nerhardt, G. (1970), "Humor and Inclinations of Humor:Emotional Reactions to Stimuli of Different Divergence Froma Range of Expectancy," Scandinavian Journal of Psy-chology, 11, 185-95.

Onkvisit, Sak and John J. Shaw (1987), "Standardized Inter-national Advertising: A Review and Critical Evaluation ofthe Theoretical and Empirical Evidence," Columbia Jour-nal of World Business, 22 (Fall), 43-55.

Oppliger, Patrice A. and John Sherblom (1988), "Late NightWith David Letterman: A Humorous Balance," Commu-nication Research Reports, 5 (2), 193—6.

Peebles, Dean M., John K. Ryans, Jr., and Ivan R. Vemon(1978), "Coordinating International Advertising," Journalof Marketing, 42 (January), 28-34.

Pick, Anne D. (1980), "Cognition: Psychological Perspec-tives," \n Handbook of Cross-Cultural Psychology, Voi. 3,Harry C. Triandis and William Lonner, eds. Boston, MA:Allyn and Bacon, 117—53.

Raskin, Victor (1985), Semantic Mechanisms of Humor. Bos-ton: D. Reidel.

Ronen, Simcha (1986), Comparative and Multinational Man-agement. New York: John Wiley & Sons, Inc., 166-9.

Rosch, Eleanor (1977), "Human Categorization," in Studiesin Cross-Cultural Psychology, N. Warren, ed. New York:Academic Press, Inc., 1-49.

Scott, Cliff, David M. Klein, and Jennings Bryant (1990),"Consumer Response to Humor in Advertising: A Series ofField Studies Using Behavioral Observation," Journal ofConsumer Research, 16 (March), 498-501.

Shultz, T. R. (1972), "The Role of Incongruity and Resolutionin Children's Appreciation of Cartoon Humor," Journal ofExperimental Child Psychology, 13, 456-77.

Stayman, Douglas M., David A. Aaker, and Donald E. Bruz-zone (1989), "The Incidence of Commercial Types Broad-

cast in Prime Time," Journal of Advertising Research, 29(June/July), 26-33.

Sujan, Mita (1985), "Consumer Knowledge: Effects on Eval-uation Strategies Mediating Consumer Judgements," Jour-nal of Consumer Research, 12 (June), 31-46.

Suls, Jerry (1983), "Cognitive Processes in Humor Appreci-ation," in Handbook of Humor Research, Jeffery Gold-stein, ed. New York: Springer-Verlag Inc., 39-57.

Toyne, Brian and Peter G. P. Walters (1989), Global Mar-keting Management: A Strategic Perspective. Needham, MA:Allyn and Bacon.

Triandis, Harry C , Robert Bontempo, Marcelo J. Villareal,Masaaki Asai, and Nydia Lucca (1988), "Individualism andCollectivism: Cross-Cultural Perspectives on Self-IngroupRelationships," Journal of Personality and Social Psy-chology, 54 (2), 323-38.

Tse, David K., Russell W. Belk, and Nan Zhou (1989), "Be-coming a Consumer Society: A Longitudinal and Cross-Cultural Content Analysis of Print Ads From Hong Kong,the People's Republic of China and Taiwan," Journal ofConsumer Research, 15 (March), 457—72.

Walters, Peter G. P. (1986), "International Marketing Policy:A Discussion of the Standardization Construct and Its Rel-evance for Corporate Policy," Journal of InternationalBusiness Studies, 17 (2), 55-69.

Weinberger, Marc G. and Harlan E. Spotts (1989), "Humorin U.S. Versus U.K. TV Commercials: A Comparison,"Journal of Advertising, 18 (2), 39-44.

Wicker, F. W., I. M. Thorelli, W. L. Barron III, and M. R.Ponder (1981), "Relationships Among Affective and Cog-nitive Factors in Humor," Journal of Research in Person-ality, 15, 359-70.

Zhang, Yong and George M. Zinkhan (1991), "Humor inTelevision Advertising: The Effects of Repetition and So-cial Setting," in Advances in Consumer Research, Vol. 18,Rebecca H. Holman and Michael R. Solomon, eds. Provo,UT: Association for Consumer Research, 813-18.

Zinkhan, George M. and Betsy D. Gelb (1990), "Humor Rep-etition and Advertising Effectiveness," in Advances in Con-sumer Research, Vol. 17, Marvin E. Goldberg, Gerald Gom,and Richard W. Pollay, eds. Provo, UT: Association forConsumer Research, 438-41.

Reprint No. JMS72I04


Documents

Dana L Alden, Wayne D. Hoyer, & Choi Lee Identifying Global and …php.scripts.psu.edu/faculty/n/x/nxy906/COMPS... · 2008-03-05 · Dana L Alden, Wayne D. Hoyer, & Choi Lee Identifying