Consumers' Cultural Differences As Local Systems of Tastes: a Critique of the Personality/Values Approach and an Alternative Framework


Douglas B. Holt (1994) ,"Consumers' Cultural Differences As Local Systems of Tastes: a Critique of the Personality/Values Approach and an Alternative Framework", in AP - Asia Pacific Advances in Consumer Research Volume 1, eds. Joseph A. Cote and Siew Meng Leong, Provo, UT : Association for Consumer Research, Pages: 178-184.

Asia Pacific Advances in Consumer Research Volume 1, 1994      Pages 178-184


Douglas B. Holt, The Pennsylvania State University


International consumer research lacks a useful approach for empirically examining cross-national differences in consumption patterns that arise from cultural differences. This paper argues that the predominant approach used in international consumer research to study cultural differences-the personality/values approach-is inadequate for this task. Three weaknesses of this approach are detailed: universalism and the negation of meaning, translation and the problem of action, and holism and the level-of-analysis problem. To address these weaknesses, an alternative approach developed by synthesizing and adapting contemporary cultural theory-"cultural systems of taste"-is proposed. As opposed to the nomothetic, univocal, universalizing perspective of traits and values, cultural systems of tastes are ideographic, meaning-based, and consumption-focused.

Culture is considered by many to be the central concept in international consumer research. While economic, political, demographic, and ecological factors each play a significant role in explaining consumption patterns cross-nationally, much of the explanatory weight is usually directed toward culture. But while culture assumes the brunt of the explanatory load, consumer researchers have had little empirical success describing cultural differences in consumption patterns across countries (Douglas and Craig 1992).

Given this problem, it is perhaps surprising that little effort has been devoted to conceptualizing the relationship between culture and consumption in the international area of research, even though the cultural approach has made substantial inroads in consumer research (e.g., Levy 1981; McCracken 1986; Mick 1986; Sherry 1987). I argue that a major reason for this lack of success is that almost all empirical cross-cultural consumer research makes use of a conception of culture-the personality/values approach-that is inadequate for the task of uncovering differences in consumption patterns driven by culture. Instead, I argue that, following contemporary cultural theory, we should develop techniques that treat culture as ideographic symbolic systems, which-based on the substantive interests of consumer research-we can specify as "cultural systems of tastes." I use McCracken's (1989) study of home decor to demonstrate the usefulness of this approach.


In international consumer research there exists an enormous rift between how culture is viewed in non-empirical works and how it is measured in empirical research. For example, in textbooks, culture is usually defined as a catch-all term that includes every aspect of human life. Reading consumer research and international marketing textbooks, one is quickly led to believe that social scientists became frustrated with and ceased thinking about culture around 1952. It is in this year that one finds Kluckholn's oft-cited (in marketing) comments that there exist at least 164 definitions of culture in the literature. Interpreting this finding as confirmation that no useful conception of culture exists, textbook authors then typically turn to a vague, all-inclusive definition: "culture includes every part of life" (Cateora 1990), "culture reflects the human aspect of a person's environment: it consists of beliefs, morals, customs, and habits learned from others" (Jeannet and Hennessey 1992), "culture is the learned ways of group living and the group's responses to various stimuli" (Toyne and Walters 1993). These definitions cast such wide nets that it is often unclear how "culture" differs from "human experience." Certainly none provide enough specificity to offer a useful means of examining consumption phenomena.

In stark contrast, most empirical studies in international consumer research (e.g., Kamakura and Mazzon 1991; Hong, Muderrisoglu, and Zinkhan 1987; Madden, Caballero, and Matsukubo 1986; Tansey, Hyman, and Zinkhan 1990; Yau 1988; see Tse, Belk, and Zhou 1989 for a somewhat different approach, though still susceptible to most of this critique) operationalize culture as consumers' placement on a set of universal values (such as those postulated by Rokeach or the List of Values) or personality traits (such as those postulated by Hofstede and Triandis). Thus, cultural differences in consumption are almost always conceptualized in nomothetic terms. [In international consumer research, it has recently become popular to describe the difference between nomothetic and ideographic approaches to culture in terms of "etic" and "emic." It should be noted that, based on their original and predominant usage stemming from linguistics and anthropology, etic and emic are inappropriate synonyms for nomothetic-ideographic. Emic refers to the "native's" understanding -- that is the phenomenological reality of the people under study. Etic refers to any other type of understanding of these people, but most often references academic theories. Thus, descriptions using either an ideographic or a nomothetic conception of culture could serve as an etic interpretation of a given population. The only example located in the major consumer research and marketing journals that includes an ideographic tact is Wallendorf and Arnould's (1988) cross-cultural analysis of favorite objects. Note that in this article, they attempted to measure materialism as a nomothetic construct and found that the construct didn't hold up across cultures.] That is, following the lead of cross-cultural psychology, cultural differences are conceived as variance in the weightings given by people in different societies to universal, invariant constructs. Those taking the values tack often use Rokeach's (1973, 5) perspective on values: "A value is an enduring belief that a specific mode of conduct or end-state of existence is personally or socially preferable to an opposite or converse end-state of existence." Like others who have presented typologies of values or goals or needs (e.g., Vernon and Allport's "derived categories," Murray's "needs"; Parsons' "need-dispositions," Maslow's "need hierarchy," Hofstede's "worker values"), Rokeach asserts that there exists a finite set of values (he argues for eighteen terminal and eighteen instrumental values) that are global in reach and that vary across individuals in terms of their relative importance. The most sophisticated recent use of Rokeach's work in consumer research is Kamakura and Mazzon's (1991) development of "value systems"-which consider value-sets rather than just highest-ranked values-to identify cohesive value-based segments in Brazil. However, even this method yields R-squared values in predictions of consumer preferences averaging only about 2%, consistent with previous efforts using LOV and VALS (Kamakura and Mazzon 1991).

Recently, Clark (1990) has attempted to resuscitate the personality approach to culture, resurrecting the "national character" concept popular in anthropology in the World War II period and later adapted by cross-cultural psychologists to examine cultural differences. Clark draws primarily from Inkeles and Levinson (1969) to define national character in psychological terms as "the pattern of enduring personality characteristics found among the populations of nations" (Clark 1990, 66). In Clark's article, there is some ambiguity as to what national character is intended to reference; sometimes it is patterns of observable behaviors and sometimes that it is a collection of personality traits. Nor does the article distinguish between personality traits and values, although it seems to imply that the latter is a subcategory of the former. However, it appears that this personality-centered view of national character shares all of the basic assumptions of the values approach: (1) there exists a reasonably small set of traits (values) that significantly structure human action, (2) these traits are objective, universal and invariant constructs that hold the same meaning for all people, and so (3) variation in the importance of these traits drives variation in consumption patterns. [It should be noted that in this article Clark constructs a highly-misleading argument to impugn what he terms the "culture-centered" approach. In constructing his argument, Clark (1990) makes two highly-questionable meta-scientific claims regarding the desirability of the personality approach as compared to what he terms the "cultural approach" to characterizing a nation's people: that the personality approach is an objective measure, not so theory-laden since "individual traits are simply observed, enumerated, tabulated, and aggregated," whereas culture-centered approaches tend to be intuitive and subjective, are theory-laden, and are not falsifiable (Clark 1990), 67-69); and also that cultural approaches are untenable because "culture is a fundamental unit of analysis ...[and it] has proven notoriously difficult to measure." I am unable to recognize in Clark's discussion of the "cultural approach" any semblance of the predominant social science approaches to examining cross-cultural difference. While the paper is positioned as a "review," it describes only a handful of studies, all dated prior to 1950. Even in this time frame, it makes no mention of the most famous "national character" study ever conducted -- Max Weber's examination of the personal characteristics that emerged in the world's great religions. It appears that the reason for this oversight is that Clark took the review very literally -- if researchers do not explicitly use the term "national character," even though theory are making cross-national comparisons of personality, they aren't included. Since the term fell into disuse after World War II, this is where Clark's story ends. But there exists a plethora of rigorous and theoretically-sophisticated cross-national anthropological and sociological research conducted in the years since that takes a cultural approach (to take a few recent examples: Schweder 1991; Kohn 1989; Lipset 1990; Lamont 1992).

Let us consider the presumed merits of the personality approach relative to this contemporary research stream. Contra Clark, consumer researchers have for some time recognized that all empirical studies are theory-laden (Peter and Olson 1983). Personality traits are never simply observed as Clark would have us believe; rather, they are a priori constructions of the researcher who creates a measurement instrument that allows the researcher's informal theorizing to be captured in ta forma manner. Clark seems to assert that personality measures are somehow more closely grounded in empirical data than cultural studies. However, if anything, cultural studies are much more inductive. For example, field work conducted by cultural anthropologists and sociologists frequently seeks to develop "emergent themes" or "grounded theory" that are constructed through intensive analysis of large quantities of discursive and observational data. In contrast, Clark's approach requires that a survey instrument be developed that requires the researcher to make a priori decisions regarding which set of personality traits (of the near-infinite traits possible) will capture the differences of interest. Clearly, this selection process is highly-subjective and is not falsifiable. While Clark's second criticism -- that culture has proven notoriously difficult to measure -- is certainly true, this does not in any way absolve his approach from the same critique. As he notes, personality has also suffered from measurement woes, so much so that one rarely finds consumer researchers studying personality. That both concepts have faced similar criticisms is not surprising since many social theorists would argue that they are quite similar (although the terms can be distinguished in that different theoretical questions and methodological strategies are associated with each).]


Certainly there are no right or wrong ways of conceiving of culture in an absolute sense, but, from a pragmatic viewpoint, different conceptions are more or less useful for the tasks to which they are applied by consumer researchers (Peter and Olson 1983). The cross-cultural psychological work on values and personality from which consumer researchers have drawn have proved enormously enlightening in describing the basic characteristics of humankind across all populations. For example, the recent work of Schwartz and Bilsky (1987, 1990) provides compelling evidence for the existence of seven distinct "motivational types" (achievement, enjoyment, maturity, prosocial, restrictive conformity, security, and self-direction) in a multiple-country sample.

But always, the relevant question remains: Is this approach useful for the descriptive tasks that we, as international consumer researchers, ask it to perform? To examine this issue, we must first set out the analytic tasks for which consumer researchers rely on the concept of culture and, then, examine the suitability of the predominant conception of culture for addressing these tasks.

Consumer research typically looks to culture to describe and explain divergent consumption patterns across large populations such as nations, regions, or ethnic groups. Culture is often used specifically to explain differences in consumption that cannot be accounted for by material differences such as in economic resources or geography. While these other factors can explain variance based on different requirements for or resources to acquire basic material needs (e.g., why the Japanese have smaller refrigerators than Americans [Fields 1983]), they are not able to provide explanations for differences in preferences for goods that are more-or-less equivalent in terms of material functionality. Since, in late-capitalist countries, the functional differences between goods is often negligible, the weight of explanation for consumption differences falls to culture (Sahlins 1976). The question then becomes, does the values/personality approach to culture satisfactorily describe these culturally-structured differences in consumption patterns? Generally, I believe that the values/personality approach is oriented toward examining cross-cultural similarities rather than differences. As a result, the approach is not adequate for explicating the often-nuanced differences in consumption patterns that consumer researchers seek to discover. In this critique, I discuss three important weaknesses of the values/personality conception of culture that limit its usefulness for describing differences in consumption patterns: universalism and the negation of meaning, translation and the problem of action, and holism and the level-of-analysis problem.

Universalism and the Negation of Meaning. The values/personality approach asserts that everyone shares the same limited set of traits/values/needs, and that cultural differences arise from variation in their ranking or patterning. While this nomothetic approach to culture fell out of favor some time ago in anthropology and sociology, it is retained by psychologists interested in culture, likely because this view is congenial to the paradigmatic goal of mainstream psychology-to understand psychic universals. However, the weight of empirical research and cultural theory suggests that such an approach is inadequate for the tasks it is asked to accomplish in consumer research.

Etic measures of personality and values severely underdetermine consumption patterns because the intensive data reduction strategy used in this approach results in abstracting away those aspects of culture that are most significant in organizing consumption. All universalizing analytic approaches must generalize the phenomena under investigation to a high degree in order to find similarities. An important question, then, given consumer research's interests in culture as outlined above is: Can the cultural differences that remain after this data reduction adequately describe differences in consumption patterns or has the descriptive power of the data been reduced away? Shweder and Bourne (1984) refer to the reductive feature of universal constructs as the "higher-order generality rule." They cite a study of connotative meanings by Osgood, which takes a nomothetic approach similar to the personality/values approach, that demonstrates the potential problems of this research strategy:

...Osgood concluded that all peoples appraise objects and events in terms of three universal dimensions: good versus bad (evaluation), strong versus weak (potency), and fast versus slow (activity). These universals were discovered, in part, by moving to a level of discourse so general that "God" and "ice cream" were descriptively equivalent: both were perceived as good, strong, and active. (p. 115)

Similarly, the usefulness of the values/personality approach is questionable because it lacks the specificity necessary to extract many key differences in consumption patterns. Those favoring the nomothetic approach sometimes argue that ideographic approaches necessarily lead to eclectic studies that don't serve to cumulate knowledge (Clark 1990). Implicit in this argument is the assumption that theoretical approaches necessarily involve constructs that are universal and invariant. The problem with this understanding of cultural constructs is that most theoretical approaches to culture preclude both of these assumptions.

For example, Clark (1990) suggests that Inkeles and Levinson's (1969) three dimensions of national character-relation to authority, conceptions of self, and primary dilemmas and conflicts-are primary determinants of consumption patterns on the basis that "they appear to be universally occurring and phenomenologically real" (p.70). While one may find these dimensions universally, they are not the only universally occurring dimensions of import (e.g., consider temporal orientation, metaphysical beliefs, and nature of family relations), nor is there any a priori reason to believe that variation along these dimensions drives consumption patterns. Even more important, though, is the problem that this approach assumes that the only constructs of interest are universals. However, there is little evidence to suggest that such universals are satisfactory for explaining cultural variation. In fact, the "culture and personality" approach lost influence in anthropology and sociology by the 1960s because lists of universals were found to be empirically inadequate in describing the growing body of cross-cultural ethnographic data (D'Andrade 1992). The dissatisfaction with personality measures motivated the development of alternative approaches that focused primarily on how societies differ in terms of their "local" interpretation (Geertz 1983) of universal values. Values, in this view, are not defined as variations in rankings on a small universal set, but as culturally-specific constructions. This position does not necessarily reject the existence of psychic universals such as those portrayed in values lists (Quinn 1992), but it does suggest that the cultural patterning of behavior is driven by qualitative differences in how goals are understood and pursued rather than quantitative differences in ranking. Since the motivational force of universals such as values are to a great degree shaped by the systems of meaning through which the value is understood, ranked differences in global values provide less explanatory power than the particular cultural systems used to interpret them.

One explanation for why the universals measured in the personality/values approach lack significant explanatory power, then, is that they are divested of meaning. Meaning is considered by most cultural theorists to be the central feature of culture. For example, everyone has a need for belonging; consumption preferences hedge on how one translates the abstract concept of belonging into particular thoughts and actions. In other words, the actions are structured not by the level of need for belonging, but rather by what belonging means. And meanings of belonging are shaped by the local cultural system that shapes how, when, where, why, and with whom one desires to belong. Similarly, everyone significantly values security, but there are tremendous differences across groups in how security is perceived (e.g., peace of mind, physical safety, sense of self, kinship network).

Translation and the Problem of Action. Another weakness of the personality/values approach relative to the goals of cross-cultural consumer research is that it emphasizes the end-states that people prefer and, therefore, overlooks how people reach these end-states. In the values approach, cultural structuring is conceived in economistic terms as generic declarations of a person's moral order (e.g., I value a sense of belonging) that act as goals shaping specific actions. Personality measures model cultural structuring in terms of general dispositions or psychological states (e.g., uncertainty-avoidance, relation to authority). Both approaches assume that consumption patterns are directly and unproblematically "inscribed" or "translated" from these goals and states. This position is consonant with the utilitarian-economic assumption (also found in structural-functionalism, exchange theory, behavioral decision modeling, and most consumer behavior texts) that actions are rational (or irrational) responses to well-defined, stable sets of goals and traits.

While persons no-doubt hold such goals (or values, or needs), there is ample evidence suggesting that this view provides an underdetermined, attenuated model of how action is structured. Prominent social theorists including Douglas (1973), Goffman (1974), and Bourdieu (1977) have demonstrated that the structuring force of culture comes not so much from following explicit rules or pursuing explicit goals, but rather in doing what feels appropriate or natural, what is common-sense given one's understanding of the situation. Culture is composed not just of goals that direct action, but also constitutive presuppositions that allow us to define and make sense of experience. In fact, cultural models often structure action apart from, and sometimes even in conflict with, the values to which we adhere (Strauss 1992).

Let us now examine this problem of translation in terms of the research goals of international consumer research. Consider a popular construct in cross-cultural consumer research-security/risk (found in Inkeles and Levinson, Rokeach, and Hofstede). How does a consumer's desire for security or degree of risk-avoidance translate into consumption patterns? First we need to resolve how a consumer makes sense of this abstract concept. As described above, risk can take on many meanings. But even with knowledge of the cultural meaning of risk, we have not yet sufficiently plumbed this term from a cultural perspective. We must also consider what modes of action, or "cultural models," are reasonable and desirable given our understanding of risk. The primary cultural meaning of risk for some groups (e.g., cultures emphasizing the individual and mortality) may be the potential for physical harm, with death as the ultimate threat. But how is one to act (e.g., as a consumer) given this understanding of risk as a cultural category? Many alternatives are possible, all of which may provide functionally equivalent resolution to one's desire for risk avoidance. For example, one can live in an area where crime is low, once can get regular physical exams, one can purchase a security system and watchdog, and so on. How do people choose between such alternatives? Cultural theory suggests that people have models for appropriate actions that they use to act on their understandings of the world. Because they focus on end-states rather than processes, personality traits such as risk-avoidance do not capture the specific type of risk-avoidance action that the consumer pursues. For example, the most effective ways for Americans to reduce physical risk would be to cease to drive automobiles, to exercise regularly, and to significantly reduce the percentage of fat in their diet. However, these are not "reasonable" alternatives for most Americans. Since personality/values rankings cannot describe why different groups pursue different actions to satisfy common desires, they are not sensitive enough to successfully analyze cross-national differences in consumption.

Holism and the Level-of-Analysis Problem. A final problem with the values/personality approach in terms of cross-national consumer research is its assumption that the measured values and traits have equal and unlimited scope. Persons are described as having a set of characteristics which they apply with equal enthusiasm across all of their life activities. This conception of person-structure assumes the traditional psychological notion that persons hold unitary, consistent, obdurate self-concepts that are reflected in similarly cohesive, consistent patterns of behavior. But the self is an historical phenomenon; the contemporary highly-reflexive, atomistic view of self is primarily a modern, Western invention. In recent years, influenced by theories of post-modernity, the static, holistic conception of self has lost favor across the social sciences. Current conceptions of self-found across psychology, sociology, and anthropology-suggest a much more mutable conception, viewing the self as a contingent, situational structure. People now must grapple with the reality that who they are and how they act varies extensively across situations and time, that many aspects of how we perceive ourselves are contradictory, that one has no "real" or "inner" self but simply many possibilities.

A view of the self similar to the values/personality approach is found in traditional anthropological studies that describe the culture of a society in a singular, generic sense by showing how the numerous institutional spheres of the society-e.g., economic and political systems, religion, family-are structured by a single, coherent cultural system. However, as many contemporary anthropologists recognize, this holistic assumption is dubious in complex, industrialized societies where it is virtually impossible to identify a single, homogeneous "Culture" that structures all of the society's sub-groups and institutional spheres. Bourdieu has formalized this idea in his conception of the "field" - institutional domains within a society whose structures, while homologous, vary significantly in terms of content. Thus, for example, Bourdieu (Bourdieu and Wacquant 1992) distinguishes the fields of education, consumption, religion, science, art, and economy as relatively autonomous planes of action.

Given the situational or role-specific nature of the structuring of action, which corresponds with the many different institutional spheres in which individuals participate, explicating global personal structures may well be a futile strategy. Rather, it may be more effective and efficient to focus the analysis upon the field of interest. For the purposes of consumer research, specifically examining the institutional sphere of consumption likely will provide more specified descriptions of cultural differences than inferring these differences from global measures.


The lack of empirical success in describing consumption patterns using the personality/values approach is not primarily due, as is often suggested, to the inherent intractibility of culture or to imperfect methodology. Rather, as argued above, the problem lies in the inadequacy of this approach for extracting differences in consumption patterns across groups. To date, the disciplines that have devoted significant attention to cultural differences across groups-anthropology and sociology-have had a negligible impact on how culture is studied in international consumer research. Similarly, the theoretical advances made by anthropologists, sociologists, and semioticians over the past decade in consumer research (e.g., Levy 1981; Mick 1986: Sherry 1987; McCracken 1988; Wallendorf and Arnould 1989) have received little attention as well. Here, I develop a framework for examining the cultural aspects of consumption that draws from contemporary cultural theory. This approach provides a conceptual framework for examining cultural differences in consumption patterns that addresses the three weaknesses of the personality/values approach.

Following Geertz (1973), culture can be viewed as a system of symbols that acts as a "lens" for making sense of the world and as a "blueprint" that disposes one to act in what is perceived to be a reasonable and natural manner given this sensemaking. Cultural systems are predominantly "local" constructions (Geertz 1983)-that is, since culture evolves within specific groups (which can be examined at many levels: nations, regions, ethnic enclaves, occupations, organizations, even families), meaning too must be organized at the group level. While Geertz and other hermeneutic anthropologists often assume the existence of a singular cultural system operating across all activities within a homogeneous society, many cultural researchers today loosen this assumption to examine the multiple and diffuse cultural systems that operate across a given population.

Geertz' two famous metaphors for culture-lens and blueprint-each have their own research tradition. Linguists, structuralists, semioticians, ethnoscientists, and others have focused on understanding the systems of meaning (the lenses) that underlie human understanding. In particular, research focusing on meanings emphasizes how the assignment of meaning creates categories (of things, people, situations, roles, time, etc.) which act as "symbolic boundaries" (Douglas and Isherwood 1979; Zerubavel 1992). Through defining and categorizing our environment, we give it meaning. Another important property of cultural categories is that categories are imbued with moral value (or lack thereof). Thus, through categorizing we also separate the desirable from the less than desirable (Douglas 1966; Lamont 1992).



Cultural categories, however, are not synonymous with action. Pierre Bourdieu has emphasized what he terms the "habitus"-the culturally-constituted implicit skills that are reflected in everyday actions. His seminal study of how the cultural aspects of consumption vary across class (Bourdieu 1984) is a prominent example of how this view can be applied to consumption. Similarly, cognitive anthropologists have recently developed an extensive literature on "cultural models"-the natural and reasonable ways of acting in particular situations. Research examining models for categories such as marriage and success has demonstrated that these models of action have motivational force.

Culture, then, can be conceived as a local symbolic system consisting of cultural categories (implying symbolic boundaries and moral valuation) and cultural models (structuring how one should proceed in a given situation). Further, since we are interested in examining a specific phenomena-consumption-it is helpful to restrict the domain of this concept. That is, instead of considering all of the dimensions of a cultural system, we can examine the cultural system as it applies to consumption-what I here term "cultural systems of tastes" (see Gans 1971; Bourdieu 1984). Taste, as the term is used here, is conceived as the cultural system people apply to make sense of, evaluate, and partake of the universe of consumption objects (goods, services, events). Cultural systems of tastes, then, organize: the cultural categories consumers use to understand what is consumed (e.g., automobiles may be understood in terms of sportiness, elegance, sturdiness, functionality, sense of freedom and adventure, and masculinity/femininity); how these categories are substantiated in goods and services (e.g., the design, colors, and features that connote masculinity in a car); and the cultural models that structure how one consumes (e.g., the way in which American high-school age males use cars as a primary prop for leisure and dating activities).

Compared to the personality/values approach, the cultural systems of tastes approach proposed here gives primacy to the local understanding of universal ends, rather than the degree to which people hold these ends. Specifically, the "cultural operator" (i.e., the mechanism that structures differences in consumption patterns) in the personality/values approach is the weightings of universals, while, in the system of tastes approach the cultural operator is the ideographic system of tastes which organizes one's understanding of and evaluation of consumption objects, and structures one's preferences and actions (see Figure).

Studying Cultural Systems of Tastes: McCracken's "Homeyness" Study

In order to demonstrate the usefulness of this "cultural systems of taste" approach, it may be helpful to develop an example in some detail. An empirical study using an analytic approach similar to that suggested here is McCracken's (1989) fine-grained examination of "homeyness." Let us assume that, from an international consumer research perspective, we are interested in developing a cultural understanding of differences in group preferences for a home. The explanatory significance of "home" is considerable for consumer research since one's understanding of home structures preferences for house style, furniture, flooring, paints and wallpaper, accessories, and so forth. First, how might one approach this question from the values/personality perspective? In this approach, commonly-cited values are thought to have relevance for home preferences-e.g., from Rokeach (1973) we could select family security, a comfortable life, and social recognition. Typical measures of these values would provide rankings or semantic differential scores so that the results would show how the groups-of-interest differ in terms of the importance assigned these values. But would these results be useful in explaining differences in preferences for a home? This seems unlikely. What type of house, furniture, and accoutrements would one expect from a group ranking "a comfortable life" as an important value? Certainly, there are numerous ways of making a home comfortable, depending upon what "comfortable" means to the group and how these meanings are translated into consumption objects and consumer actions. As argued above, this approach severely underdetermines consumption patterns because it reduces away meaning.

How would a study of home and consumption differ using the "cultural systems of taste" approach developed here? While in an abstract sense, one's home certainly reflects values of security, social recognition, and a comfortable life, how a home reflects these properties (i.e., the specific consumption patterns that reflect these values) is not determined by the values themselves, but by how these values are understood. The cultural operator that accomplishes this translation from abstract value to specific consumption preferences is a person's "system of tastes," of which the home can be viewed as one important substantiation. While it is likely that a number of cultural categories are important in linking tastes to home preferences, McCracken examines what may well be the key category, what he terms "homeyness." How does McCracken's manner of describing homeyness, then, improve on Rokeach's rankings of "a comfortable life"? Instead of assuming that homeyness is an objective, univocal construct, McCracken treats the term culturally. This leads him to analyze what homeyness means to his Canadian sample and how they act out their understanding of homeyness in choosing and decorating their house. Understanding homeyness first requires defining it relationally-that is, understanding what it is not. McCracken's informants viewed homeyness in terms such as "looks as though someone lives there," "informal," "comfortable, "cozy," and "welcoming," as opposed to houses that are "pretentious," "formal," "stark," "elegant," and "cold." Further, McCracken is able to explicate eight symbolic properties of homeyness. For example "the authentic property" describes homeyness as an environment that is natural, personalized and therefore individualistic-it signifies the authenticity of the home-dweller relative to mass institutions. This authentic quality is created through a variety of details such as decorating with objects of personal significance (photos, crafts, heirlooms), arranging furniture so that it is not overly symmetric, combining furniture and decor in eclectic combinations, and so forth. McCracken's study demonstrates, then, that the cultural dimension of consumption arises not from the relative importance of abstract ends (be they values, personality traits or needs). Rather, cultural differences are driven by how these universals are made meaningful and acted upon by a group.


In this paper, I argue that the personality/values approach to understanding cultural differences in consumption suffers from three significant weaknesses: universalism and the negation of meaning, translation and the problem of action, and holism and the level-of-analysis problem. Together, these three problems limit the ability of empirical research grounded in this approach to make useful cross-cultural comparisons of consumption patterns. To address these weaknesses, I leverage contemporary cultural theory to develop what I term the "cultural systems of tastes" approach to culture and consumption. The cultural systems of tastes concept has several key features that improve upon the personality/values approach:

(1) Cultural differences in consumption are conceptualized as differences in how consumers make sense of and act upon desired end-states, rather than as relative differences in how these end-states are ranked. In other words, this approach assumes cultural differences arise because goals/values/needs assume different meanings in different cultures, while personality/values assumes that goals/values/needs have a single, universal meaning and so people vary in terms of the importance assigned to the meaning.

(2) The "cultural systems of tastes" concept takes into consideration the variety of cultural models that consumers may apply to pursue a given end-state, rather than assuming that end-states determine actions.

(3) The "cultural systems of tastes" concept focuses specifically on explicating the cultural factors underpinning consumption patterns rather than inferring consumption patterns from global structures.

This critique and alternative perspective suggest that to successfully describe the cultural dimension of cross-national consumption patterns, international consumer research needs to shift its emphasis from nomothetic psychological constructs to ideographic systems of tastes.


Bourdieu, Pierre (1979), Outline of a Theory of Practice, New York: Cambridge University Press.

Bourdieu, Pierre (1984), Distinction, Cambridge, MA: Harvard University Press.

Bourdieu, Pierre and Loic Wacquant (1992), An Invitation to Reflexive Sociology, Chicago: University of Chicago Press.

Cateora, Philip (1990), International Marketing, 7th Edition, Homewood, IL: Irwin.

Clark, Terry (1990), "International Marketing and National Character: A Review and Proposal for an Integrative Theory," Journal of Marketing, 54(4), 66-79.

D'Andrade, Roy (1992), "Schemas and Motivation," in Human Motives and Cultural Models, eds. Roy D'Andrade and Claudia Strauss, New York: Cambridge University Press.

Douglas, Mary (1966), Purity and Danger, London: Routledge.

Douglas, Mary (1973) ed.,Rules and Meanings, Baltimore: Penguin Education.

Douglas, Mary and Baron Isherwood (1979), The World of Goods, New York: Basic Books.

Douglas, Susan P. and C. Samuel Craig (1992), "Advances in International Marketing," International Journal of Research in Marketing, 9, 291-318.

Fields, George (1983), From Bonsai to Levis, New York: MacMillan.

Gans, Herbert (1971), Popular Culture and High Culture: An Analysis and Evaluation of Taste, New York: Basic Books.

Geertz, Clifford (1973). The interpretation of cultures. New York: Basic Books.

Geertz, Clifford (1983) Local Knowledge, New York: Basic Books.

Goffman, Erving (1974), Frame Analysis, New York: Harper and Row.

Hong, J.W., A. Muderrisoglu, and G.M. Zinkhan (1987), "Cultural Differences and Advertising Expression: A Comparative Content Analysis of Japanese and U.S. Advertising," Journal of Advertising, 16,55-69.

Inkeles, Alex and Daniel J. Levinson (1969), "National Character: The Study of Modal Personality and Sociocultural Systems," in The Handbook of Social Psychology, Vol.4, Cambridge, MA: Addison-Wesley.

Jeannet, Jean-Pierre and Hubert Hennessey (1992), Global Marketing Strategies, 2nd Edition, Boston: Houghton Mifflin.

Kamakura, Wagner A. and Jose Afonso Mazzon (1991), "Value Segmentation: A Model for the Measurement of Values and Value Systems," Journal of Consumer Research, 18(2), 208-218.

Kohn, Melvin, ed. (1989), Cross-National Research in Sociology, ewbury Park, CA: Sage.

Lamont, Michele (1992), Money, Morals, and Manners, Chicago: University of Chicago Press.

Levy, Sidney (1981), "Interpreting Consumer Mythology," Journal of Marketing, 45, 49-63.

Lipset, Seymour Martin (1990), Continental Divide: The Values and Institutions of the United States and Canada, New York: Routledge.

Madden, C.S., M.J.Caballero, and S. Matsukubo (1986), "Analysis of Information Content in U.S. and Japanese Magazine Advertising," Journal of Advertising, 15, 38-45.

McCracken, Grant (1988), Culture and Consumption: New Approaches to the Symbolic Character of Consumer Goods and Activities, Bloomington: Indiana University Press.

McCracken, Grant (1989), "`Homeyness': A Cultural Account of One Constellation of Consumer Goods and Meanings," In Interpretive Consumer Research, ed. Elizabeth Hirschman, 168-183. Provo, UT: Association for Consumer Research.

Mick, David Glen (1986), "Consumer Research and Semiotics: Exploring the Morphology of Signs, Symbols, and Significance," Journal of Consumer Research, 13 (September), 196-213.

Peter, J. Paul, and Jerry Olson (1983), "Is Science Marketing?" Journal of Marketing 47, 111-125.

Quinn, Naomi (1992), "The Motivational Force of Self-Understanding: Evidence from Wive's Inner Conflicts," in Human Motives and Cultural Models, eds. Roy D'Andrade and Claudia Strauss, New York: Cambridge University Press.

Rokeach, Milton (1973), The Nature of Human Values, New York: Free Press.

Sahlins, Marshall (1976), Culture and Practical Reason, Chicago: University of Chicago Press.

Schwartz, Shalom and Wolfgang Bilsky (1990), "Toward a Theory of the Universal Content and Structure of Values: Extensions and Cross-Cultural Replications," Journal of Personality and Social Psychology, 58(5), 878-891.

Schwartz, Shalom and Wolfgang Bilsky (1987), "Toward a Universal Psychological Structure of Human Values," Journal of Personality and Social Psychology, 53(3), 550-562.

Shweder, Richard (1991), Thinking Through Cultures: Expeditions in Cultural Psychology, Cambridge, MA: Harvard University Press.

Shweder, Richard and Edmund Bourne (1984), "Does the Concept of Person Vary Cross-Culturally?" in Culture Theory: Essays on Mind, Self, and Emotion, eds. Richard Shweder and Robert Levine, New York: Cambridge University Press.

Sherry, John F., Jr. (1987), "Heresy and the Useful Miracle: Rethinking Anthropology's Contributions to Marketing," Research in Marketing Vol. 9, 285-306, JAI Press.

Strauss, Claudia (1992), "What Makes Tony Run? Schemas as Motives Reconsidered," in Human Motives and Cultural Models, eds. Roy D'Andrade and Claudia Strauss, New York: Cambridge University Press.

Tansey, R, M.R. Hyman, and G.M. Zinkhan (1990), "Cultural Themes in Brazilian and U.S. Auto Ads: A Cross-Cultural Comparison," Journal of Advertising, 19(2), 30-39.

Toyne, Brian and Peter Walters (1993), Global Marketing Management, 2nd Edition, Boston: Allyn and Bacon.

Tse, D.K., R.W. Belk, and N. Zhou (1989), "Becoming a Consumer Society: A Longitudinal and Cross-Cultural Content Analysis of Print Ads from Hon Kong, the People's Republic of China, and Taiwan," Journal of Consumer Research, 15, 457-472.

Wallendorf, Melanie and Eric Arnould (1988), "My Favorite Things: A Cross-Cultural Inquiry into Object Attachment, Possessiveness, and Social Linkage," Journal of Consumer Research, 14(4), 531-47.

Yau (1988), "Chinese Cultural Values: Their Dimensions and Marketing Implications," European Journal of Marketing, 22(5), 44-57.

Zerubel, Eviatar (1991), The Fine Line: Boundaries and Distinctions in Everyday Life, New York: Free Press.



Douglas B. Holt, The Pennsylvania State University


AP - Asia Pacific Advances in Consumer Research Volume 1 | 1994

Share Proceeding

Featured papers

See More


L12. Should I Stay or Should I Go: When Our Companies Have Eyes for Other Consumers

Na Ri Yoon, Indiana University, USA
Jenny Olson, Indiana University, USA
Adam Duhachek, Indiana University, USA

Read More


To Apologize, or Not to Apologize? That is A Question - How Should an Organization Respond to Executive Employees’ Private Life Misconduct?

Zayed Bin Islam, University of Guelph, Canada
Juan Wang, University of Guelph, Canada
Towhidul Islam, University of Guelph, Canada

Read More


O11. Have Less, Compromise Less: How the perception of resource scarcity influences compromise decisions

Kate Kooi, University of Miami, USA
Caglar Irmak, University of Miami, USA

Read More

Engage with Us

Becoming an Association for Consumer Research member is simple. Membership in ACR is relatively inexpensive, but brings significant benefits to its members.