Academics and practitioners alike have debated the desirability of standardized vs. locally tailored international advertising campaigns. Though this controversial practice has attracted a great deal of attention over the last two decades, we seem little closer to complete agreement than then the issue was first joined.

Proponents of the standardized approach argue that, for any given product, consumers' interests everywhere are basically the same and that therefore, by making advertising decisions at headquarters and transferring a standardized advertising strategy from country to country, an international company can enhance its economic effectiveness.

The opposite view of standardization strategy ecphasizes obvious dissimilarities among countries, especially those relating to consumer goods, and argues in favor of using internationally differentiated avertising strategy.

From a theoretical standpoint, the validation of the standardization Concept could have broader implications which may necessitate a re-examination of the general marketing process, The basis for the need to re-examine the marketing process involves the fact that advertising standardization implicitly assumes consumer homogeneity while the concept of market segmentation, in contrast, assumes consumer heterogeneity. There market segmentation, in contrast, assumes consumer heterogeneity. There appears to be a need for examinating the validity and applicability of each strategy in order to dry to resolve this apparent contradiction.

The purpose of this research is to provide empirical evidence for whether or not to use standardized advertising based on a comparison of evaluative criteria for automobiles and toothpaste among American, Japanese and Korean consumers. This research also provides empirical evidence for whether the demographic and psychographic variables can be used as segmentation criteria for standardized advertising among these countries.

This research objective was accomplished with a survey of U.S., Japanese and Korean consumers. The sample sizes were 200 from the United State, 189 from Japan, and 200 from Korea. Each sample completed the questionnaire in their home country. In order to assure linguistic consistency in the questionnaires, the Korean version was translated into English and Japanse respectively, and translated back to Korean, a common method in cross-cultural surveys.

The respondents in each country were asked to rate the importance of several attributes for each of the products along a five point Likert sacle which ranged from "Extremely important" to "Pay no attention to it." Twenty attributes were rated for automobiles and 16 for toothpaste.

The analysis of the data was performed in the following steps. First, in an effort to examine the comparability of the sample, x²-tests were used to determine if the groups varied in their sex, marital status, and occupational status. Second, analysis of variance was then used to determine whether the differences in the means among the three countries were statistically significant. Third, on each of the evaluative criteria for which significant differences were found to exist among the three countries, scheffe tests were calculated to determine if significant differences exist between Korea and the United States, Korea and Japan, and/or the United States and Japan. Fouth, two-way MANOVA was used to examine variance due to country, the demographic variables and the interaction between the two variables. And two-way MANOVA was also used to examine variance due to country, the psychographic variables, and the interaction between the two variables.

The major findings of empirical analysis are as follows:

① In an effort to examine the comparability of the samples, x²-test indicated no significant differences among the three samples in terms of occupational status. In addition, an F-test of the differences in the mean ages of the three samples was not statistically significant. Thus, these domographic variables appear comparable for the three samples. But there were differences in terms of sex and marital status. These differences should be kept in mind when interpreting the results.

② A MANOVA testing of means for automobiles and toothpaste showed that the differences among the three samples were significant in the overall multivariate analysis. A closer examination of the results via the univariate ANOVA showed that the differences derived from a comparison of means were generally significant. Differences in 17 of the 19 automobile significant. Therefore, the results of the study appear to contradict the idea that international advertising should be standardized due to the similarities among world-wide consumers. The manager should be award that people in different countries may vary significantly in their purchasing motivation.

③ The results of two-way MANOVA indicated that the influences of demographic variables such as sex, occupation, age and psychographic variables such as modern life-style were significant in explaining purchasing motivation, but the effect was not strong enough to wipe out national influences. Cunsequently, there is little indication to suggest that a strategy of cross-country segmentation focusing on demographic and psychographic variables is likely to be appropriate either for automobiles or toothpaste.

This study, however, has the following limitations.

① The samples employed make it impossible to generalize the specific findings to the entire pupulations of the representative countries. Thus, the findings provide an indication of the extent to which the consumers of these countries differ, but nor necessarily the exact manner in which they differ.

② The findings of the study have no applicability beyond these two products.

③ Examination of appropriate research procedures in making crossnational study is needed. Although relatively little bias appears to be introduced by using equivalent measurement instruments, it is not clear whether such an assumption can be made in relation to concepts such as opinion leadership or modern life-style.