, Pages 70-76
, University of Alaska-Fairbanks, U.S.A.
, University of Alaska-Fairbanks, U.S.A.
, The Chinese University of Hong Kong, Hong Kong
The purpose of this paper is to discuss ways in which Geert Hofstede's research on cross-cultural values in an employment context may be applicable to consumer behavior.
In 1980, Geert Hofstede's book, Culture's Consequences: International Differences in Work-Related Values was published. In this book, Hofstede describes the culmination of a research project begun in the 1960's for the HERMES corporation, later revealed to be IBM (Hofstede and Bond, 1988). The original data set, derived from 117,000 questionnaires in 20 different languages comparing 88,000 respondents from 66 countries in 50 occupations, was eventually stabilized to include 53 countries and regions (Hofstede, 1983a).
The survey instrument could be divided into those questions relevant to work satisfaction, perceptions, personal goals and beliefs, as well as demographics. Hofstede's interest was restricted to those questions pertinent to values or "...a broad preference for one state of affairs over others" (p. 389, 1984a). Using theoretical guidance on the mammoth data set, Hofstede extracted the fundamental blueprint of his conceptualization of four basic dimensions, or indices, across which employees of different countries may be meaningfully compared. The four dimensions are power distance, uncertainty avoidance, individualism, and masculinity.
Briefly, power distance is the societal acceptance of hierarchy; uncertainty is the cultural tolerance of ambiguity; individualism is the societal predilection for inter-dependence; and masculinity is the culture's desire for sex-role differentiation. Since this phenomenal beginning, Hofstede (1982; 1983a, b, c; 1984a, b, c; 1985; 1987) and others (e.g., England, 1983; Hortum and Muller, 1989; Jensen, White, and Singh, 1990; Kim, Park, and Suzuki, 1990; Lebas and Weigenstein, 1986) have validated and expounded on his initial insights. However, the discussion, with few exceptions (e.g., Milner, in press), has always taken place in an employment-related context.
Since workers are often consumers too, the purpose of this paper is to present Hofstede's work and to suggest that his ideas on work-related behaviors may be relevant to cross-cultural differences and similarities in consumptive behavior as well. The discussion of the present paper will focus on the dimensions as conceptualizations, and not on the methodological and statistical methods employed to derive the four indexes. For those interested, the abridged (1984c) and complete version (1980) of Culture's Consequences: International Differences in Work-Related Values are excellent in their detail of the research and statistical analysis process used to derive these four dimensions.
THE FOUR DIMENSIONS
The four dimensions and each countries relative standing on these dimensions are featured in Appendix A. "Together, these four dimensions explain about 50 percent of the measured value differences among countries. The remaining differences are country-specific, i.e. they cannot be explained from universal dimensions" (Hofstede, p. 39, 1982). Appendix B presents a table of the four major dimensions, a brief definition of the dimensions, the highest and lowest countries on each dimension, as well as selected implications for consumer behavior.
Power distance is the societal desire for hierarchy. Expressed more eloquently, Hofstede (1984b) defines it as:
Power distance is the extent to which the members of a society accept that power in institutions and organizations is distributed unequally....People in Large Power Distance societies accept a hierarchical order in which everybody has a place which needs no further justification. People in Small Power Distance societies strive for power equalization and demand justification for power inequalities among people when they occur. (p. 83)
Hofstede's research indicates that two predictors of the Power Distance Index score which are relevant to consumer behavior are geography and wealth. Indeed, 43% of the variance in the score can be explained by the geographical latitude of the country's capital. Further, the colder the climate, the smaller the power distance. Using a Darwinian type of explanation whereby social "pecking" orders are leveled and even negated by the need for mere survival, Hofstede speculates that "...the key intervening variable is the need for technology as a condition for survival. Human survival in colder climates presupposes protection against the hardships of nature, which means that only those people survived who were able to master the minimal technical skills necessary for survival. In warmer climates the need for technology was less" (p. 95, 1984c). Thus, where survival is dependent upon technology, social stratification becomes irrelevant.
Regarding wealth, Hofstede states "Greater wealth presupposes higher technology; higher technology calls for higher-educated but also better-paid lower and middle strata; so wealth will be more equally distributed and power also will be more equally distributed" (p. 98, 1984c).
What then are the consequents and correlates of the power distance dimension for consumer behavior? A very obvious hypothesis suggested by Hofstede's research is that there will be a correlation between geographical latitude and the consumption of technology, probably not only in terms of the amount of technology purchased but also in terms of the sophistication of the technology as well. Additionally, although it remains to be seen whether the term "better" should be operationalized in terms of either volume or quality, another hypothesis would be that those consumers in low Power Distance countries (due to either wealth or education) would be better consumers than those in high Power Distance countries. If conspicuous consumption is any indication of social class, then low Power Distance countries, where social class stratifications are less predominant, would not be as likely to have displays of conspicuous consumption, as high power distance countries.
If religious preference may be considered within the domain of consumer choice, then Hofstede's research already indicates some tangible applicability to consumer research. Consider that the Power Distance Index is correlated with a country's dominant religious choice. By their very nature of worshiping a supreme being or acknowledging a higher power, religions are hierarchical; however some may be more hierarchical than others. Specifically, Hofstede (1984c) argues that "Catholicism with the supreme authority of the Pope and the intermediate authority of the priest corresponds more to a large power distance than Protestanism with its general priesthood of the believers" (p. 104). Interestingly, Catholicism is more clearly associated with high power distance index countries and Protestanism is more correlated with low power distance index countries.
The Uncertainty Avoidance Index addresses a society's tolerance of ambiguity and uncertainty. Questions on the survey that deal with uncertainty avoidance address rule orientation, employment stability, and stress. Specifically, Hofstede finds that those who seek to avoid uncertainty endorse responses consistent with the ideas that rules are made to be followed, people should stay with organizations for a lifetime, and that because of its uncertainty, life is stressful. Conversely, "uncertainty-accepting cultures are more tolerant of behavior and opinions that differ from their own; they try to have as few rules as possible, and on the philosophical and religious level, they are relativist, allowing many currents to flow side by side" (Hofstede and Bond, 1988, p. 11)
Mechanisms suggested by Hofstede that societies use to cope with uncertainty include law and religion. Hofstede found that in the Christian countries in the sample, Catholicism was more highly correlated with the desire to avoid uncertainty than Protestanism. Hofstede argues that Catholicism, more so than Protestanism, works to remove life's ambiguities by such things as stressing infallibility of the pope, offering participants greater control over their life after death, etc.
Hofstede also found that high Uncertainty Avoidance Index scores were correlated with extensive legislation. For instance, Hofstede (1984c) notes "We can, for example, oppose Germany (UAI 65) to Great Britain (UAI 35). Germany has an extensive set of laws even for emergencies that might occur ("Notstandgesetz"); Great Britain does not even have a written constitution. Attempts to codify labor-management relations in Britain (the Industrial Relations Act) failed because they are too much against societal norms" (p. 135). Further, in comparing the use of personal identity cards in Western Europe, Hofstede found that higher higher Uncertainty Avoidance countries in Western Europe require identity cards but that lower Uncertainty Avoidance countries did not.
Though having no data, Hofstede (1984c) speculates that uncertainty avoidance will impact a culture's choice of games. Specifically, "an interest in chance games as part of a national culture is likely to be correlated with greater willingness in managers to make risky business decisions. For the time being, I have no data on this issue, but impressionistically, at least British and U.S. managers seem more willing to make risky decisions than, for example, German and French managers" (p. 138).
Hofstede believes that uncertainty avoidance impacts the meaning of time and the desire for precision and punctuality. Specifically,
...more Uncertainty Avoiding traditional societies like Mexico are more hurried than less Uncertainty Avoiding societies where meditation is important (like India)....among the Uncertainty Avoiding developed countries, precision and punctuality come more naturally than among the less Uncertainty Avoiding ones. The success of a country like Japan in the precision industries is supported by the strong Uncertainty Avoidance in its culture (1984b, p. 95)
Such an orientation surely would impact expectations as well as expressions of service in the buyer-seller transaction.
This dimension seems relevant to other consumer behaviors as well. For instance, one would expect that brand loyalty to a given product would be higher in those countries where uncertainty avoidance is high. New product introductions would be expected to be more difficult in high uncertainty avoidance countries as well. Indeed, countries might be placed along the adoption curve such that countries like Singapore (#1), Jamaica (#2), and Denmark (#3) would be expected to be more eager to accept innovations but Guatemala (#48), Portugal (#49), and Greece (#50) more properly considered as laggards with Finland (#20-21), Brazil (#29-30), and Spain (#36-41) expected to occupy a middle-ground of early adopters, early majority, and late majority.
Since uncertainty avoidance would indicate a waryness, intolerance, or dislike of foreigners, researchers studying country of product origin effects on consumer acceptance might find results impacted by the extent to which a country was classified as uncertainty avoidance. Finally, one would expect that those countries high in uncertainty avoidance might have more stringent consumer protection mechanisms like better product and service warranties, "cooling-off" laws, etc.
According to Hofstede,
Individualism stands for a preference for a loosely knit social framework in society wherein individuals are supposed to take care of themselves and their immediate family only. Its opposite, Collectivism, stands for a preference for a tightly knit social framework in which individuals can expect their relatives, clan, or other in-group to look after them in exchange for unquestioning loyalty....I relates to people's self-concept: 'I' or 'we.' (1984b, p. 83)
Individualism is highly correlated with wealth. Hofstede found that 82% of the variance in the Individualism Index Score is accounted for by GNP per capita. Indeed,
The data show convincingly that the arrow of causality goes from wealth to Individualism and not vice versa. If the resources in a country allow people to "do their own thing," they will start doing just that....Only for wealthy countries (all of which tend to be individualistic) is more Individualism associated with slower economic growth and vice versa. If everybody does his or her own thing, the economy grows less quickly than it would if at least some individuals worked for collective purposes. (Hofstede & Bond, 1988, p. 14)
This dimension is probably the least obviously relevant to consumer behavior; however there are still possible applications. For instance, Hofstede suggests that High Individualism Societies will place more emphasis on "autonomy, variety, pleasure, and individual financial security" in contrast to Low Individualism Societies which will value "expertise, order, duty, security provided by organization or clan" (1984c, p. 171). Obviously, financial services for individuals would be more heavily sought out by consumers in high individualism societies than in low individualism societies. The sale of "pleasure" products and services like travel and cultural events such as theatre, opera, etc. would be expected to be higher in the individualistic countries than collectivistic countries.
Since people in collectivistic cultures will more likely live in an extended or tribal structure and possibly be more likely to share living quarters, then there is likely to be an impact on such durable purchases like number of stoves and refrigerators bought. One could predict that less household space would be devoted to decorative items, and more toward functional items like beds, etc.
Without the support of extended families, childcare would be a greater need among more individualistic societies. Since survival is more dependent on individual initiative, perhaps one could predict a greater emphasis on education for all. It could also be hypothesized that decision making styles would be impacted. For instance, Hofstede (1985) says that in collectivistic societies, "the individual is nothing without his/her in-group and will strive for the group interest" (p. 354). Such a statement suggests that joint decision making would be more frequent in these countries than independent decisions.
Finally, there are some implications for retailer-consumer interactions. Specifically, Hofstede (1984b) notes that in collectivist cultures,
Relations, friends, tribesmen get better deals than strangers and this is the way it should be....Considerations of personal trust and relationships should have precedence over business considerations....In collectivist cultures with their tightly knit and predetermined social framework, there is generally an extensive set of expectations of how people should behave towards each other....Maintaining harmony consists in avoiding anybody's loss of face....In most collectivist cultures, therefore, it is not a virtue to be open and direct. (p. 88-89)
Such statements suggest that in a collectivist culture, once a relationship is established between a retailer and a customer, then the retailer can expect a higher store loyalty from the customer. Similarly, the customer can expect the service and product quality rendered to be higher than at an unfamiliar outlet. Further, consumer complaint behaviors might be impacted. For instance, if directness is not appreciated, perhaps suggestion boxes might be more heavily utilized in these countries.
As stated previously, the masculinity dimension measures the preference for sex-role distinctions to be made between men and women in a particular culture. Specifically, "the factor score...on this factor has been used as the basis for a country Masculinity Index (MAS) which measures to what extent HERMES respondents in a country (of both sexes) tend to endorse goals usually more popular among men (high MAS) or among women (low MAS)" (p. 176, 1984a) Hofstede states,
Masculinity stands for a preference in society for achievement, heroism, assertiveness, and material success. Its opposite, Femininity, stands for a preference for relationships, modesty, caring for the weak, and the quality of life....Some [masculine] societies strive for maximum social differentiation between the sexes....Other societies strive for minimal social differentiation between the sexes. This means that some women can take assertive roles if they want but especially that some men can take relationship-oriented, modest, caring roles if they want to. The minimum-social-differentiation societies in comparison with their opposite, the maximum-social-differentiation societies, will permeate their institutions with a caring, quality-of-life oriented mentality. Such societies become "welfare societies" in which caring for all members, even the weakest is an important goal for men as well as women. (1984b,p. 84)
Therefore, this dimension addresses not only which gender's values will be most valued but also the differentiation between the sexes.
As far as can be discerned, this dimension, using Hofstede's conceptualization, is the only dimension thus far to be utilized outside of the work-related context. Specifically, Milner (in press) states: "Though originally written for a work behavior context, Hofstede's research seems particularly appropriate for gender-positioning studies. For instance, the extent to which a culture could be characterized as emphasizing the differences between men and women would affect the positioning of a product and the consequent promotion of that product." Continuing on, she states, "According to Bartos (1989), the traditional product categories marketed to women are food, fashion, grooming, and homemaking but now are beginning to include travel, automobiles, and financial services. Hofstede's research would indicate that those countries that are identified as strong gender differentiators would have less gender-usage crossover for traditional products than those countries identified less strongly." That is, it would be more difficult to get women to use men's products or men to use women's products in those countries where there is strong differentiation.
Also as stated previously, Hofstede's framework could be utilized in research examining the promotion of consumer products. For instance, studies of sex role stereotypes in advertising are highly prevalent (e.g., Gilly, 1988; Lysonski, 1985; Robbins and Paksoy, 1989; Whipple and Courtney 1985). To date, however, no real framework exists for organizing and explaining these studies, particularly the cross-cultural studies. As an example of how Hofstede's work could be applied, consider Gilly's 1988 study, "Sex Roles in Advertising: A Comparison of Television Advertisements in Australia, Mexico, and the United States" in which her findings indicate that "Australian advertisements show somewhat fewer sex role differences and Mexican advertisements show slightly more sex role differences that U.S. advertisements" (p. 75).
In explaining her results, Gilly suggests the contrast with previous research (Edgar and McPhee, 1974) indicating that Australian ads were more sex-typed than U.S. ads may be due to the fact that Australian advertisers have been receptive to various "pressure" groups suggesting improvements be made in the portrayal of women in ads. The surprising lack of greater differences between the U.S. and Mexico, "...a country perceived to be much more traditional than our own..."(p. 84), is tentatively suggested to be due to the great influence of U.S. products and their consequent ads in Mexico. Gilly's findings however are exactly what Hofstede would predict. First, his research suggests there is a myth concerning the U.S. as the "great liberator of woman kind." Indeed, out of 50 countries, the U.S. ranking is 36. In Hofstede's framework Gilly chose only traditional countries, but still the findings of her research correspond exactly to his rankings. That is, Australia is 35, the US is 36, and Mexico is 45.
Now a very interesting test using Gilly's approach and Hofstede's framework would be to compare the television advertising contents of a Scandinavian country (Sweden =1; Norway=2; Finland=7; Denmark=4) with her Mexican data. Or to get away from Anglo and Hispanic comparisons, the Mexican data could be compared with data gathered from another Hispanic culture like Chile which is ranked #8 on Hofstede's Masculinity dimension. Indeed, an intriguing suggestion of Hofstede's data is that just because a country is Hispanic or even Muslim-influenced like Turkey, which is rated #20-21, does not mean they are less egalitarian with respect to men and women than an Anglo or Christian country like the U.S. which is rated #36.
Interactions between dimensions
Thus far, the dimensions have only been described individually or as main effects. However, the more potent predictive power probably lies in their interactions. For instance, "both Power Distance and Individualism affect the type of leadership most likely to be effective in a country....Masculinity and Uncertainty Avoidance affect people's motivations: Competition is more effective in a masculine culture, and personal risk is more acceptable if Uncertainty Avoidance is low" (Hofstede and Bond, p. 14).
Using sex role portrayals in TV advertisements as an example again, one might predict that in promoting children's products in masculine countries there would be more frequent "mother" commercials but in comparison, feminine countries more be more likely to feature "parent" or "dad" commercials; in collectivist countries, it would be predicted that more commercials feature extended families, but in individualistic countries, more commercials featuring the nuclear family would be the expectation. Therefore the interactions of these dimensions would then predict that advertisements in high masculine/collective countries like Venezuela (#48/#4) would more likely feature mother/extended family ads; Costa Rica (#5-6/#8), a feminine/collective countries would have more parent/extended family ads. Denmark (#4/#42), a feminine, individualistic country would be more likely to feature parent/nuclear family ads. The U.S. (#36/#50), a masculine/individualistic country could be counted on for mother, nuclear family ads. Unfortunately, data collected in previous sex role research is not reported in a way necessary to show support or nonsupport for the previous hypotheses; therefore, at this point, it is mere speculation.
The Fifth Dimension: Confucian Dynamism
Hofstede has never been one to assert that the four dimensions of masculinity, individualism, uncertainty-avoidance, and power distance are completely exhaustive of all useful cross-cultural value categories possible. There seem to be other dimensions. Indeed, recently Hofstede and Bond (1988) have posited the existence of a fifth dimension, which they label "Confucian Dynamism." Confucian Dynamism has the advantage over the previous four dimensions of being positively correlated with a country's economic growth. Of interest to consumer behaviorists is the fact that one of the values inherent in this dimension is thrift, or in other terms, non-consumption. Indeed, Hofstede and Bond point out that "the value of 'thrift' leads to savings, which means availability of capital for reinvestment, an obvious asset to economic growth...." (p. 18)
Possibly of even greater interest to consumer behaviorist, however, is the method with which this dimension was originated. Literally, this dimension comes from survey research done by Chinese researchers on values. When analyzed, three dimensions similar to power distance, individualism, and masculinity were derived. The dimension, which Hofstede and Bond now label Confucian Dynamism, was also apparent. From the perspective of the Western mind, the questions necessary to derive this dimension would never have been asked. Similarly, the dimension, "uncertainty avoidance," was not a result in the Chinese values survey. The questions necessary to even obtain this dimension would not have occurred to Eastern researchers. Hofstede and Bond (1988) sum it up:
Besides the three previously mentioned dimensions common to both West and East (Power Distance, Individualism/Collectivism, and Masculinity/Femininity), we found one uniquely Western dimension: Uncertainty Avoidance. As we argued, this dimension deals with a society's search for Truth; uncertainty-avoiding cultures believe in an absolute Truth, and uncertainty-accepting cultures take a more relativist stance. We also found one uniquely Eastern dimension, Confucian Dynamism; we believe that this dimension deals with a society's search for Virtue. It is no accident that this dimension relates to the teachings of Confucius; as we described them earlier, he was a teacher of practical ethics without any religious content. He dealt with Virtue, but left the question of Truth open. (p. 19)
In light of such findings, there is little doubt as to why Hofstede (1984c; Hofstede and Bond, 1988) is such a strong advocate of the notion that even the very theories themselves by which we derive our ideas and practices are cultural creations and can be so culture bound that while useful in the originating culture, they are rendered irrelevant or at least not very useful when transferred to other cultures.
As stated previously, Hofstede does not suggest that the value categories derived in his research are the only cultural values of consequence. Further, he is not even necessarily the originator of the conceptualizations behind the specific values he investigates; his work has incorporated the ideas of others. Nor is his reporting of the data as complete as one would wish it to be. The results as reported are not applicable to individual differences or even to with-in country regional differences. The data is aggregate and applies only to those gross "cultural" differences as defined by geographical borders.
Beyond these shortcomings, however, his research on work-related values in a cross-cultural context does seem applicable to consumptive behavior in a cross-cultural context. Obviously, studies are needed to directly assess this assumption that data gathered on value dimensions in the work world are relevant to values in a consumption context. Further, this research would need to be made compatible with the already-existing research on values and consumptive behavior.
Beyond these considerations, however, the appeal of Hofstede's research lies fundamentally in the fact that countries are already categorized on dimensions, not necessarily exclusive or exhaustive, but nevertheless dimensions recognized as important. For cross-cultural researchers without the necessary access to requirements like research participants, colleague connections, and the funds for validating predictive scales on similar values of interest, Hofstede's results offers an economical way to make hypotheses about predicted results when gathering inter-country data.
COUNTRIES AND THEIR RELATIVE RANKINGS
SUMMARY TABLE OF HOFSTEDE'S APPLICATION TO CONSUMER BEHAVIOR
Bartos, Rena (1989), Marketing to Women around the World, Boston, MA: Harvard Business School Press.
Edgar, Patricia and Hilary McPhee (1974), Media She, Melbourne: Heinemann.
England, George W. (1983), "Japanese and American Management: Theory Z and Beyond," Journal of International Business Studies, Vol. 13, No. 2, 131-142.
Gilly, Mary C. (1988), "Sex Roles in Advertising: A Comparison of Television Advertisements in Australia, Mexico, and the United States," Journal of Marketing, Vol. 52 (April), 75-85.
Hofstede, Geert (1980), Culture's consequences: International Differences in Work-Related Values. Beverly Hills, CA: Sage Publications.
Hofstede, Geert (1982), "Cultural Pitfalls for Dutch Expatriates in Indonesia: Lessons for Europeans, Part 1," Euro-Asia Business Review, Vol. 1, No. 1, 37-41.
Hofstede, Geert (1983a), "Dimensions of National Cultures in Fifty Countries and Three Regions," in Expiscations in Cross-Cultural Psychology, J.B. Deregowski, S. Dziurawiec, and R.C. Annis, eds., Lisse, Netherlands: Swets & Zeitlinger, 335-355.
Hofstede, Geert (1983b), "The Cultural Relativity of Organizational Practices and Theories," Journal of International Business Studies, Vol. 14, No. 2, 75-89.
Hofstede, Geert (1983c), "Cultural Pitfalls for Dutch Expatriates in Indonesia: Lessons for Europeans, Part 2 & 3," Euro-Asia Business Review, Vol. 2, No. 1, 38-47.
Hofstede, Geert (1984a), "The Cultural Relativity of the Quality of Life Concept," Academy of Management Review, Vol. 9, No. 3, 389-398.
Hofstede, Geert (1984b), "Cultural Dimensions in Management and Planning," Asia Pacific Journal of Management, January, 81-99.
Hofstede, Geert (1984c), Culture's consequences: International Differences in Work-Related Values. (Abridged Edition), Beverly Hills, CA: Sage Publications.
Hofstede, Geert (1985), "The Interaction Between National and Organizational Value Systems, Journal of Management Studies, Vol. 22, No. 4, 347-357.
Hofstede, Geert (1987), "The Applicability of McGregor's Theories in South East Asia," Journal of Management Development, Vol. 6, No. 3, 9-18.
Hofstede, Geert, and Michael Harris Bond (1988), "The Confucius Connection: From Cultural Roots to Economic Growth," Organizational Dynamics, Vol. 16, No. 4, 4-21.
Hortum, Mustafa and Lloyd H. Muller (1989), "Management and Culture," Supervision, Vol. 50, No. 1, 14-17.
Jensen,Thomas D., Donald D. White, and Raghavendra Singh (1990), "Impact of Gender, Hierarchical Position, and Leadership Styles on Work-Related Values," Journal of Business Research, Vol. 20, No. 2, 145-152.
Kim, Ken I, Park, Hun-Joon, and Nori Suzuki (1990), "Reward Allocations in the United States, Japan, and Korea: A Comparison of Individualistic and Collectivistic Cultures," Academy of Management Journal, Vol. 33, No. 1, 188-198.
Lebas, Michael, and Jane Weigenstein (1986), "Management Control: The Roles of Rules, Markets and Culture," Journal of Management Studies, Vol. 23, No. 3, 259-272.
Lusonski, Steven (1985), "Role Portrayals in British Magazine Avertisements," European Journal of Marketing, Vol. 19, No. 7, 37-55.
Milner, Laura M. (in press), "Multinational Gender Positioning: A Call for Research," in B. Englis and F. Baker (Eds.), Global and Multinational Advertising, New York: NY: Lawrence-Erlbaum Associates.
Robbins, Stephanie and Christie H. Paksoy (1989), "A Comparative Study of German and U.S. Magazine Advertisements: Has Global Standardization Been Achieved?" in Marketing: Positioning for the 1990's, Robert L. King, ed., Charleston, SC: Southern Marketing Association, 339-342.
Whipple, Thomas W. and Alice E. Courtney (1985), "Female Role Portrayals in Advertising and Communication Effectiveness: A Review," Journal of Advertising, Vol. 14, No. 3, 4-17.