Special Session Summary Beyond Broad Cross-Cultural Classifications: Implications For Understanding Consumer Responses

Sharon Shavitt, University of Illinois, Urbana-Champaign
Julie Anne Lee, University of Hawaii at Manoa
[ to cite ]:
Sharon Shavitt and Julie Anne Lee (2002) ,"Special Session Summary Beyond Broad Cross-Cultural Classifications: Implications For Understanding Consumer Responses", in NA - Advances in Consumer Research Volume 29, eds. Susan M. Broniarczyk and Kent Nakamoto, Valdosta, GA : Association for Consumer Research, Pages: 46-48.

Advances in Consumer Research Volume 29, 2002     Pages 46-48



Sharon Shavitt, University of Illinois, Urbana-Champaign

Julie Anne Lee, University of Hawaii at Manoa


Marketing, advertising, and consumer behavior are global phenomena, yet the cross-cultural aspects of consumer behavior received surprisingly little research attention until recently. Fortunately, the last several years have witnessed a dramatically increased focus on the role of cultural differences in social cognition and consumer response. So much important work has been published and so much has been learned that cross-cultural issues in general are now mainstream topics, not emergent ones. The great majority of work in this area has contrasted two broad cultural types, individualism and collectivism, independence and interdependence, East and West, etc. We argue that there is a need to investigate more refined conceptualizations and classifications of cultural types (see Maheswaran and Shavitt, 2000). Although the broader classifications have tremendous value, they also obscure important distinctions among the cultures within those classifications. This is the emergent topic covered in the present session.

The first paper, by Julie Lee, Howard Marmorstein, and Richard Brislin, demonstrates how sub-dimensions of individualism and collectivism may allow for more precise predictions of consumer behaviors than would be afforded by broader measures, such as independent/interdependent self-construal (Singelis, 1994). Their results show a greater discrimination between different consumer behaviors than what was found using Singelis’s broader measure.

The second paper, by Sharon Shavitt, Jing Zhang, and Timothy Johnson, demonstrates the role of another distinction within individualism and collectivism: horizontal versus vertical orientation. Their results show differences in the persuasiveness of appeals within individualism/collectivism categories that fit with theoretical assumptions about verticals’ focus on status and horizontals’ focus on uniqueness (individualists) or social relationships (collectivists). They also show that these persuasion differences are not predicted by broader measures of individualism/collectivism or independent/interdependent self-construal.

The third paper, by Richard Bagozzi, Nancy Wong, and Mahesh Gopinath, also utilizes measures of individualism and collectivism diensions. Their results show differences in the way people experience emotions with pride and negative emotions correlating positively for collectivist and negatively for individualist based cultures. In addition, collectivist values were found to predict the experience of positive emotion whereas individualist values were predictive of negative emotions.

The papers in this session outline the use of state-of-the-art cultural constructs and measures, and offer examples of how these multidimensional constructs can be used in three different substantive areas of consumer behavior, including the internal representation of emotions, the influence of others, and the persuasiveness of advertising. Finally, the papers offer examples of how cultural classifications, once thought to be unidimensional, may in fact interact, to help to predict exceptions to generalizations.



Julie Lee, University of Hawaii at Manoa

Howard Marmorstein, University of Miami

Richard Brislin, University of Hawaii at Manoa

Despite the theoretical and empirical advances that have been made in the research into individualism and collectivism, numerous problems with the measurement of the central constructs still exist. Researchers have highlighted issues including the catch-all nature of the constructs (e.g. Schwartz, 1990), confounding of measures (Bierbrauer, Meyer & Wolfradt, 1994), low reliabilities (Singelis et al., 1995), and lack of predictive validity. In addition, several researchers have found unusually high correlations between scales measuring individualism and those measuring collectivism in Asian but not Western cultures. These disturbingly high correlations may be due to the multidimensional nature of the scales being used. Conceptually this is not surprising, in view of the fact that Triandis (1995) derived four universal sub-dimensions from the literature on individualism and collectivism. Specifically, these include the sub-dimensions of self-concept, goal-orientation, cognition, and relationship-value. As an example, Singelis (1994, p. 581) asserted that an interdependent self-construal "emphasizes (a) external, public features such as statuses, roles, and relationships, (b) belonging and fitting in, (c) occupying one’s proper place and engaging in appropriate action, and (d) being indirect in communication and reading others’ minds". Notably, this sub-dimension of self-concept overlaps both the goal-orientation and cognition dimensions outlined above.

These problems have led several researchers to call for a refinement of the concepts and instruments, especially when measuring the relationships between allocentric (more collectivist) and idiocentric (more individualist) tendencies and behavior. Accordingly, the current paper builds upon the individual level constructs of idiocentrism and allocentrism by clarifying their conceptualization for use in consumer behavior and creating measures for this purpose. To do so, the first step in this research was to develop a set of theoretically derived items to tap the dimensions of idiocentrism and allocentrism identified by Triandis (1995). Second, an empirical examination of the resulting scales was conducted. Finally, a comparison of the proposed measures with a widely used existing scale is provided.

To pursue the aforementioned ojectives, a survey was designed and administered to a sample of students from Hong Kong (n=178) and the U.S.A. (n=173). The survey included demographic measures, items reflecting Triandis’s four universal sub-dimensions of individualism and collectivism, Singelis’s items, and a selection of questions to measure closeness to the family and likelihood of consulting them in a purchase decision. As this paper focuses on the family context, only the measures relating to the family were included in the analysis.

First, correlation and regression analyses were used to examine the relationship-value sub-dimension across a variety of self-reported behaviors. Across the entire sample, being more relational (e.g., more likely to stick with family even if unhappy with them) is significantly correlated with spending more non-working time with family (r=.13, p<.05), feeling more emotionally close with family members (r=.34, p<.001), and being more likely to consult family members when making a major decision (r=.34, p<.001). Also as expected, the proposed measure of family relationship value was negatively related to likelihood of consulting others, including acquaintances (r=-.195, p<.00), classmates (r=-.25, p<.001), neighbors (r=-.25, p<.001), and romantic partners (r=-.13, p<.01).

Next, we conducted a multiple regression analysis to assess our predictions concerning consumers’ inclinations to seek and follow advice. Two of Triandis’s four sub-dimensionsCrelationship value and cognitions (normative versus attitude) serve as the independent variables. The dependent measures were likelihood of asking and taking advice from family members, summed across a variety of 18 different product types. In support of our hypotheses, we found that people who were more relational were more likely to ask family for advice (b=.15; t=6.28, p<.001). In addition, they were more likely to take the advice if they were more relational (b=.12; t=5.10, p<.001) and more normative (b=.05; t=2.18, p<.05).

Finally, we contrast our findings with those obtained using Singelis’s (1994) independent and interdependent self-concept scales for the same sample. We found that Singelis’s (1994) interdependence score was not significantly correlated with the amount of nonworking time spent with the family. Interestingly, both the independent and interdependent self-concept scales were significantly correlated with emotional closeness with all family members. Lastly, the interdependent self-concept was predictive of whether or not family members would be asked for advice and whether or not it would be taken, but less so than using the sub-dimensions developed in this paper. In sum, the proposed measures provide greater discrimination between idiocentrism and allocentrism and are more predictive of the consumer behaviors that were examined.

This paper proposes that articulation and measurement of the sub-dimensions of individualism and collectivism is necessary. A set of items was developed to measure the four dimensions suggested by Triandis. An empirical test of two of the dimensions showed that better discrimination of the constructs was achieved and that a number of related consumer behaviors were better explained. The research also highlights the need for further refinement of the measures of the sub-dimensions and tests of their application to understanding consumer behaviors outside of the family context.



Sharon Shavitt, University of Illinois, Urbana-Champaign

Jing Zhang, University of Illinois, Urbana-Champaign

Timothy P. Johnson, University of Illinois at Chicago

Previous research has pointed to clear distinctions in how people from individualist and collectivist cultures respond to persuasive appeals (e.g., Aaker & Maheswaran, 1997; Han & Shavitt, 1994; Zhang & Gelb, 1996). These studies have demonstrated that the distinction between individualistic and collectivistic societies is crucial to the cross-cultural understanding of consumer behavior. However, important distinctions remain within the broader Individualism/Collectivism categories. For instance, within the framework of Individualism/Collectivism, Triandis and Gelfand (1998) recently introduced a further distinction between societies that are horizontal (valuing equality) and those that are vertical (emphasizing hierarchy). A vertical orientation is associated with a focus on enhancing one’s status. In contrast, a horizontal orientation is associated with a tendency to focus on expressing one’s uniqueness and self-reliance (horizontal individualists) or a focus on social relationships and interdependence (horizontal collectivists; Triandis, 1995).

To date, the implications of this horizontal and vertical distinction for consumer responses are unknown. The present research examined the relation between horizontal and vertical types of individualism and collectivism and the persuasiveness of appeals, operationalizing these variables at the level of cultural values within the U.S. One hundred and ninety-eight participants from a large Midwestern university completed measures of cultural values, including Triandis, Chen, and Chan’s (1998) scenario measure of horizontal and vertical individualism and collectivism. Then, they were asked to write an ad (for either a fictitious brand of furniture or a website) that they would find persuasive. Ads that participants wrote were later coded for the degree to which they used particular types of appeals, including status, relationship, and uniqueness/self-reliance appeals. Next, participants reported their ad and brand attitudes in response to a series of fictitious ads in various product categories (e.g., car, vacation resort, chocolate). The appeals used in these ads were pretested and manipulated along the same dimensions used to code the ads, including appeals to status ("enjoy the royal treatment you deserve"), relationships ("they feel your love all day long" ) and uniqueness/self-reliance ("if I am the director of my own life, then I am going to set my own route" ). Appeal type was a within-subject variable. Thus, all participants read and responded to a set of ads, each of which used a different type of appeal.

Regression analyses revealed differences in the persuasiveness of appeals as a function of participants’ cultural values. Horizontal collectivism (HC) was associated as expected with more favorable (r=.17, p<.05) attitudinal responses to ads that stressed relationship benefits. Vertical collectivists (VC) were more likely to write ads that used status appeals (r=.20; p<.05). Thus, although both HC and VC cultural values reflect a collectivist orientation, their differing orientations to status and hierarchy were revealed in differences in the persuasiveness of appeals. Moreover, horizontal individualism (HI) was associated, as expected, with a tendency to write ads that used uniqueness and self-reliance appeals (r=.16; p<.05), and was associated with unfavorable attitudes toward (r=-.14; p<.05), and a reduced tendency to write (r=-.16; p<.05), ads that used status appeals. Although the horizontal/vertical distinction predicted persuasion outcomes in expected ways, broader measures of individualism/collectivism (Gaines, et al., 1997; Triandis, et al. 1998) and independent/interdependent self-construal (e.g., Singelis, 1994) did not predict the persuasiveness of these types of appeals. These findings point to important persuasion differences within individualism/ collectivism categories that track distinctions between horizontal and vertical cultural orientations. They emerged despite the fact that participants were from a U.S. sample whose experiences with advertisements and products were largely shared. Implications for cultural differences at the societal levl, as well as for the measurement and operationalization of cultural categories, are discussed.



Richard Bagozzi, Rice University

Nancy Wong, University of Hawaii

Mahesh Gopinath, Tulane University

Leading researchers in psychology maintain that positive and negative emotions should be highly negatively correlated (e.g., Barrett and Russell, 1998; Green, Goldman, and Salovey, 1993; Russell, and Carroll, 1999). For example, the more one feels happy, the less one feels sad, and vice versa. This outcome has been called the "bipolarity condition" of positive and negative affect in the literature. In the most comprehensive analysis to date of the literature on bipolarity, Russell and Carroll (1999, p. 25) found "little or no substance to the psychometric challenge to bipolarity" and concluded that "for the routine assessment of affective feelings, bipolar response formats are justified". In recent years, strong bipolarity has been found when terms are selected for tests of bipolarity that are seemingly opposites (Zautra, Potter, and Reich, 1997) and when measurement error in observations is corrected for attenuation (e.g., Green et al., 1993: Barrett and Russell, 1998).

One problem with research to date is that most research has been conducted with participants from Western cultures. A second problem is that most research has combined men and women into a single aggregate. Bagozzi, Wong, and Yi (1999) proposed that one of three patterns of correlation between positive and negative affect could result, depending on the interaction between culture-specific heuristics and gender-specific knowledge and skill in the use and interpretation of emotion words. The three patterns are (1) a dialectic pattern between positive and negative affect, such that positive covariation is expected, (2) a bipolar pattern of negative covariation, and (3) a pattern of independence. In two studies conducted in the US, China, and Korea, these predictions were born-out. Positive and negative affect correlated r=.53 for Chinese women, r=-.61 for American women, and r=-.12 and r=-.27 for Chinese and American men, respectively. Thus, positive and negative affect were predicted to be positively correlated for women in interdependent-based cultures, negatively correlated for women in independent-based cultures, and nearly independent for men in interdependent- and independent-based cultures.

A shortcoming of the findings in Bagozzi et al. (1999) is that no mechanism for the cultural effect was tested. Rather, country of origin and residency were used to operationalize the cultural differences. Second, only two countries showed the predicted effects strongly (a third country, Korea, showed the predicted pattern as well but not as strongly). The present study expands inquiry to include ten new countries (United States, China, Thailand, South Africa, Kenya, Tanzania, Malawi, Zambia, Swaziland and Nigeria), which present a combination of independent and interdependent cultures. More importantly, we measure specific orientations within each culture to explain the presumed cultural heuristics: namely, empathy, individualism-collectivism, and self-consciousness. Next, we expanded the range of positive and negative emotions to include instances not tested before (e.g., pride, humility, optimism, pessimism). Last, we varied the target of appraisal (i.e., self-generated, negative, positive event).

The results show that pride and negative emotions (anger, sadness, pessimism, fear, guilt/shame; plus humility) correlate positively for interdependent-bsed cultures and negatively for independent-based cultures (thus providing a conceptual replication of past findings). This occurred when self-generated affect was investigated. When common positive or negative event was administered, correlations between pride and negative emotions were influenced toward bipolarity in all samples. Similar results occurred for correlations between happiness and sadness. A number of other interesting gender and culture differences were found. Finally, the findings could be explained, in part, by scores on individualism-collectivism, empathy, and self-consciousness.


Aaker, J.L. and D. Maheswaran (1997), "The Effect of Cultural Orientation on Persuasion," Journal of Consumer Research, 24, 315-328.

Bagozzi, R.P., N. Wong and Y. Yi (1999), "The Role of Culture and Gender in the Relationship Between Positive and Negative Affect," Cognition and Emotion, 13: 641-672.

Barrett, L.F. and J.A. Russell, (1998), "Independence and Bipolarity in the Structure of Current Affect," Journal of Personality and Social Psychology, 74: 967-984.

Bierbrauer, G., H. Meyer and U. Wolfradt (1994), "Measurement of Normative and Evaluative Aspects in Individualistic and Collectivistic Orientations: The Cultural Orientation Scale (COS)," in Individualism and Collectivism: Theory, method, and applications, ed. Uichol Kim, Harry C. Triandis, Cigdem Kagitcibasi, Sang-Chin Choi, and Gene Yoon, pp. 189-199, Sage Publications, Thousand Oaks, California.

Gaines, S. O., Jr.; W. D. Marelich, K. L. Bledsoe, W. N. Steers, et al. (1997), "Links Between Race/Ethnicity and Cultural Values as Mediated by Racial/Ethnic Identity and Moderated by Gender," Journal of Personality & Social Psychology, 72(6), 1460-1476.

Green, D.P., S.L. Goldman and P. Salovey (1993), "Measurement Error Masks Bipolarity in Affect Ratings," Journal of Personality and Social Psychology, 64: 1029-1041.

Han, S. P. & S. Shavitt (1994), "Persuasion and Culture: Advertising Appeals in Individualistic and Collectivistic Societies," Journal of Experimental Social Psychology, 30, 326-350.

Maheswaran, D. & S. Shavitt (2000), "Issues and New Directions in Global Consumer Psychology," Journal of Consumer Psychology, 9(2), 59-66.

Russell, J.A. and J.M. Carroll (1999), "On the Bipolarity of Positive and Negative Affect," Psychological Bulletin, 125: 3-30.

Schwartz, S. H. (1990), "Individualism-Collectivism Critique and Proposed Refinements," Journal of Cross-Cultural Psychology, 21 (June), 139-157.

Singelis, T. M. (1994), "The Measurement of Independent and Interdependent Self-Construals," Personality and Social Psychology Bulletin, 20 (5), 580-591.

Singelis, T. M., H. C. Triandis, D. P. S. Bhawuk, and M. Gelfand (1995), "Horizontal and Vertical Dimensions of Individualism and Collectivism: A Theoretical and Measurement Refinement," Cross Cultural Research, 29 (August), 240-275.

Triandis, H. C. (1995), Individualism and collectivism. Boulder, CO: Westview Press.

Triandis, H. C., X-P. Chen and D. K BS Chan (1998), "Scenarios for the Measurement of Collectivism and Individualism," Journal of Cross-Cultural Psychology, 29 (2), 275-289.

Triandis, H. C. & M. J. Gelfand (1998), "Converging Measurement of Horizontal and Vertical Individualism and Collectivism," Journal of Personality and Social Psychology, 74 (1), 118-128.

Zautra, A.J., P.T. Potter and J.W. Reich (1997), "The Independence of Affects is Context-Dependent: An Integrative Model of the Relationship Between Positive and Negative Affect," in M.P. Lawton and K.W. Schaie (eds), Annual Review of Gerontology and Geriatrics, 17, 75-102.

Zhang, Y. & B. D. Gelb (1996), "Matching Advertising Appeals to Culture: The Influence of Products’ Use Conditions," Journal of Advertising, 25 (3), 29-46.