Consumer Socialization in Different Settings: an International Perspective

Scott Ward, The Wharton School
Donna M. Klees, The Wharton School
Thomas S. Robertson, The Wharton School
ABSTRACT - Consumer socialization researchers are turning their attention beyond childhood, and beyond national and cultural boundaries. This paper provides a conceptualization of childhood socialization variables which may vary across cultures, and which are important for the theoretical and applied aims of research in the area. Data are presented from a study of American, Japanese and English families. Results show cultural variation in children's television watching behavior, but not in age-related patterns of product requesting behavior; an exploratory cluster analysis identifies general and consumption-specific family interaction variables characterizing Japanese and American families.
[ to cite ]:
Scott Ward, Donna M. Klees, and Thomas S. Robertson (1987) ,"Consumer Socialization in Different Settings: an International Perspective", in NA - Advances in Consumer Research Volume 14, eds. Melanie Wallendorf and Paul Anderson, Provo, UT : Association for Consumer Research, Pages: 468-472.

Advances in Consumer Research Volume 14, 1987      Pages 468-472


Scott Ward, The Wharton School

Donna M. Klees, The Wharton School

Thomas S. Robertson, The Wharton School


Consumer socialization researchers are turning their attention beyond childhood, and beyond national and cultural boundaries. This paper provides a conceptualization of childhood socialization variables which may vary across cultures, and which are important for the theoretical and applied aims of research in the area. Data are presented from a study of American, Japanese and English families. Results show cultural variation in children's television watching behavior, but not in age-related patterns of product requesting behavior; an exploratory cluster analysis identifies general and consumption-specific family interaction variables characterizing Japanese and American families.


The purpose of this paper is to assess similarities and differences across cultures of key variables in consumer socialization research. The focus is on childhood consumer socialization --specifically, children 3-10 years old--recognizing that the parameters of socialization research are expanding to include changes throughout the life span.

As many companies seek global markets, and as technological innovations make possible instant communication among countries, a major debate in marketing is whether global markets are truly similar, or whether there are important differences across countries and cultures. Levitt (1983), for example, speaks of a "homogenizing commonality" characterizing global markets, while Wind (1985) argues that marketing practices must reflect subtle but powerful differences across countries and cultures.

For marketing practitioners, the issue is crucial: to what extent must product modifications be made to insure successful marketing from one culture to another--and what product modifications? To what extent can similar marketing programs, advertising appeals, and the like, be used across cultures? How can similarities and differences across cultures be identified and understood so that appropriate products can be designed, and effective marketing programs can be implemented.

For consumer researchers, the issues are different, but no less crucial. Beyond descriptive differences between cultures, are there fundamental differences in consumer behavior processes? Are there differences in consumer needs, motives, goals, expectations, information-processing and sources of satisfaction across cultures which reflect more than structural differences, such as annual income, product availability, literacy levels, media penetration and the like? For consumer socialization researchers, the issues are somewhat more focused: what are the cultural differences in processes that characterize the acquisition and use of consumption-related skills, knowledge and attitudes?

From a theoretical perspective, identifying and understanding consumer socialization differences across cultures would advance the field by specifying the concepts, theories and variables which characterize consumer behavior in different cultures. For example, to what extent are the relative roles and interactions among family, media and peers identified in the more plentiful American research base similar to those in European countries or Far Eastern countries?

From a more applied perspective, the childhood consumer socialization questions are crucial. Should a marketer of "child relevant" products such as snack foods and toys market the same products, with the same appeals, in different cultures--recognizing that there may be some structural differences in the availability and suitability of advertising media, distribution, and-the like? Do children want the same kinds of products worldwide? Do they select and evaluate things they want through their parents in similar ways, with similar outcomes?

We suspect that most consumer researchers would say that there are some differences and some similarities, but would tend toward the view that patterns of consumer behavior are quite different across cultures (despite the enormous success among young people of Gino's Pizza, Kentucky Fried Chicken and McDonald's in cultures as diverse as Great Britain, Japan, and Thailand, which would likely be dismissed as mere "case examples )! The issues would seem to be what differences are important for the theoretical and practical aims of the field, and the objective of this paper is to conceptualize and empirically identify some of those differences.

A few definitions and rough cuts are in order. First, we will use the terms "culture" and "country" or nation interchangeably, recognizing there are differences among these terms. However, data presented in this paper are from families in Japan, England, and the U.S., which are different nations and "cultures' in the sense that they differ in the "totality of socially-based and transmitted behavior patterns, arts, beliefs, institutions and other products of human work and thought characteristic of a community or population (Goslin, 1971).

Second, our focus is on "advanced" cultures, in the sense that all three are relatively urban, industrialized, literate, and wealthy. Moreover, our analysis is limited to countries which feature the role of the nuclear family. While Japan was a great many more extended families than England or the U.S., the family as a socializing agent is important, compared to cultures in which commune or tribal life is more Primary.

The focus on "advanced" cultures permits us to focus on socialization processes and variables without the confounding effects of other variables in less "advanced" cultures. For example, some of the core motives for consumption among Americans--to fulfill diverse needs, show achievement, attain social status, and the like-are likely to be present in other relatively "advanced" cultures, but less prevalent in cultures which are "less developed" in the usual sense of that term. Similarly, product meanings such as comfort and quality may be quite different in less advanced cultures, and the inventory of present and desired goods may also be different (although this difference may be disappearing). (One writer noted that a decade ago the Chinese desired 'four things that go round' a bicycle, a watch, a sewing machine and an electric fan. Now many Chinese aspire to own 'eight new big things' -- a camera, a tape deck, a television set, refrigerator, a motorcycle, a washing machine, an air conditioner and a video recorder; Mann, 1986.) In any case, the basket of goods and services in households in our data base are likely to be similar, compared with a household in a less developed culture.


Consumer socialization researchers are invariably interested in processes by which phenomena change, even if those changes are most often assessed via cross-sectional designs, as opposed to longitudinal research. Additionally, consumer socialization research focuses on the interactive effects of complex stimuli, including institutions (such as school), mass media, as well as social influences. In addition to such stimuli, socialization researchers focus on information-processing, broadly defined as processes by which stimuli are selected, evaluated and used in consumer decisions, and on ultimate responses.

Given the interests of many socialization researchers in stimuli, information-processing and responses, the question is: to what extent do these aspects of consumer socialization vary across "advanced" cultures? Table 1 suggests some research-based possible similarities and differences.




"Stimuli," or influences on socialization processes, may be similar or different across advanced cultures. It appears that children desire similar kinds of products across cultures, although product forms may vary. For example, all children desire food products, but American children like cereals and British children like tea. Children are exposed to advertising for these products, but there are some variations in advertising appeals, media, executions, and in the regulations and structure of broadcast advertising. For example, there are major differences in the broadcast systems and advertising practices, as information in the Appendix indicates for our three cultures.

Moreover, while early studies of children's responses to advertising in England and in the U.S. were remarkable in that their findings were highly similar, (Himmelweit, et al., 1958; Schramm, Lyle and Parker, 1961), it is reasonable to assume that children's viewing behavior varies across cultures. These differences are expected because of variations in broadcast systems and advertising presentations, family norms about viewing and advertising, and the relative novelty of television advertising in countries with mixed commercial and non-commercial broadcast systems. Simply stated, children in different cultures probably watch television differently: they may watch varying amounts of commercial television, pay more or less attention, and co-view with parents at different rates, depending on television advertising's novelty, and its utility for them.

While we do not know of any systematic content analyses of advertising to children in different cultures, we hypothesize that there are differences in the format, content, and types of appeals in advertising to children. Data do indicate differences in the degree of television's household penetration, the amounts of television children watch, and the amount of commercial television to which they are exposed. We also hypothesize - although we find no data on the point--that television is the primary advertising medium to which children are exposed in advanced cultures, but that other media, such as in-store displays and posters vary in their prevalence and influence across cultures.

Other "stimuli," include social influences--particularly of parents, in the case of childhood socialization. Studies have been conducted comparing some aspects of socialization practices by parents in the U.S., England and Japan, although no study has directly compared all three countries. It appears that English children experience more rigid and authoritarian parents than do American children. The latter are subjected to less physical punishment, and enjoy more nurturance and warmth than English children. Perhaps these differences account for English children's greater peer-orientation and less time spent with family members, compared to American children (Devereux, 1969, 1970).

Studies comparing American and Japanese families indicate that the latter are characterized by high maternal warmth, but fathers are more emotionally distant from their children. Japanese families stress parent-child contact, family interdependence, and parents exert more control over their children than do American parents (Ezra and Vogel 1968; Inomota and McGinnies 1971; Shigaki 1983).

Regarding Information Processing" in Table 1, cognitive development researchers have demonstrated the invariance of cognitive stages across cultures (Kohlberg, 1971). However, while cognitive development stages usefully characterize children's age-related processing capabilities and characteristics, we hypothesize that there are cultural differences in the nature and degree of children's values, goals, and interests, which, in turn, should affect aspects of information-processing, such as selection and evaluation of product information.

Finally, regarding "Responses," in Table 1, we hypothesize similarities in age-related patterns of children's requesting behavior. Research among American children suggests that children request fewer products as they grow older, reflecting passive dictation" (i.e., parents come to know children's product and brand favorites, so explicit requests become unnecessary), children's increasing purchasing autonomy, and their developing understanding that they must be selective in their product requests in order to achieve success.

We hypothesize differences in children's responses to socialization stimuli, due in part to family processes, and to structural characteristics of different cultures. For example, we expect that family norms in different cultures vary in the extent to which children's requests are rewarded or not. Moreover, children's opportunities for consumption should vary as a consequence of living patterns and product availability. For example, there are many more retail outlets per capita in countries such as Italy and France than in the U.S., and relatively more children living in urban areas, so products may be more accessible than for American children.


Data were gathered in a diary study in the three countries. Mothers kept daily diaries of their children's television viewing behavior, and their product requesting behavior, for two weeks. The procedure was a replication and extension of an earlier study by Isler, Popper and Ward (1979).

The sample included 84 U.S. families, 118 Japanese families, and 65 English families, stratified by age groups: 3-4, 5-7 and 8-10 years old, roughly corresponding to early and later pre-operational and concrete-operational stages of cognitive development. Families were selected only if they had between two and four children, and had both a mother and a father living at home. This was done to avoid particular problems which may attend very small or very large families, or one-parent households. Families were drawn from middle socioeconomic strata. The study is exploratory, due to the small sample size and lack of truly random sample selection procedures.

Each respondent was trained to unobtrusively observe and record the purchase requests and television viewing behavior of one child in the household between the ages of 3 and 10 years of age. At the end of the two-week period, a separate questionnaire was administered to gather additional information, on parent-child communication, conflict etc.

Families were initially recruited by telephone. A personal visit then followed to deliver and explain the materials used in the study. Participants were further instructed to telephone the researchers if questions or problems arose during the study. Researchers assured that mothers properly understood the materials and were committed to conscientious completion of the tasks described.

Care was taken in the translation of the diary materials into Japanese and in the wording modifications for the British version. Some "back-translation" was done to ensure accurate translation of the Japanese diary. There was no attrition in the Japanese study; about 6 mothers dropped out or provided incomplete diaries in England, and 8 mothers did not complete materials for the course of the study in the U.S.


Three results are presented here, which bear on the conceptualization on cultural differences presented above.

The first result concerns our hypothesis that children watch television differently in the three cultures. There are several dimensions to the phenomenon of how children watch television. First, there are differences in the total amount of time that children watch. Children in the U.S. and Britain watch equivalent amounts (about 11.3 hours per week) versus 7.3 hours per week for Japanese children. While these differences are significant (F 17.99, df 2, 264, p < .05), the number of hours is low compared to Nielsen reports for American children, perhaps suggesting some systematic under-reporting, but also perhaps reflecting the fact that the diary study was conducted in spring and summer months.

Another aspect of children's television watching behavior is the extent of co-viewing with parents. Overall, children watched with parents 46% of the time. However, Japanese parents and children were more likely to co-view (50% of the time) than American families (37%). English families reported co-viewing levels similar to the Japanese (48% of the time).

Finally, children's viewing behavior was assessed in terms of their attention to advertising, controlling for the amount of time spent watching commercial television. Results (Table 2) show marked cultural differences. Interestingly, while young children show similar levels of attention to advertising, Japanese children, who watch the least television, show the most market decrease in attention with age. There is a significant age by country interaction (p<.05), suggesting that the utility of advertising differs markedly to children at different ages in the three cultures. It may be that American children develop sufficient information selection skills by about age 7, so that they begin to pay less attention to advertising. However, British children pay more attention with age, perhaps because of their greater peer-orientation and independence from the family, and their greater purchasing autonomy. In Japan, perhaps, rules about purchase requests reduce the utility of advertising, beginning at a very young age.



The second analysis concerns our hypothesis that children will not vary across cultures in their age-related patterns of purchase requests: we expect these to decrease with age, regardless of culture. A two-way analysis of variance (age group by country) confirms our expectation (see Table 3). There is no age by country interaction, but the main effects are significant, and the data show a decrease in the number of requests as age increases. Japanese children show lowest levels of requesting behavior across age groups.




To explore cultural differences in social influences on consumer socialization processes, a cluster analysis of two sets of variables was run using the Howard-Harris routine. The first set consisted of Schaefer's measures of Affect, Dependency and Control dimensions of mother-child interaction (Schaefer 1971). The other set of variables consisted of measures of different kinds of consumer socialization practices of parents, based on previous research (Wart, Wackman and Wartella, 1977).

Three clusters were derived, and are described in Table 4. Cluster 1 consists primarily of Japanese children and nearly equal proportions of children from three age groups. Cluster 2 consists primarily of American children, and contains somewhat more 5-7 year olds; Cluster 3 is a small one, with somewhat more American and British children. and 8-10 year-olds.

Detailed findings are presented in Table 4. Overall, there are marked differences, especially between Clusters 1 and 2. Parent-child interaction in Cluster 1 is characterized by relatively quiet and introverted children who are dependent on their parents. These children are from lover socio-economic households (although "low" in this case means toward the lower end of a distribution of middle-SES families). Regarding consumer variables, these children experience the most rules governing consumption behavior, requests and parental yielding. Regarding parental methods of training their offspring in consumer skills, Cluster 1 parents are the least-demonstrative, expecting their children to learn through observation and modeling, and through trial and error.



Parent-child interaction in Cluster 2 is quite different. Cluster 2 families are primarily American, while families in the first cluster are primarily Japanese. Families are more expressive and openly affectionate, and children are more independent. They are obedient and non-manipulative, although Cluster 1 and Cluster 2 families are similar on Schaefer's Control dimension, while Cluster 3 families appear to exhibit the most control problems.

Cluster 2 families are higher income. Regarding consumer socialization variables, there tend to be fever established rules, and parents are far more purposive in their consumer , preferring directed discussions about consumption to trial and error as a teaching mechanism. Children in Cluster 2 watch the least television, and are close to the mean in requests and yielding. Cluster 3 children, by contrast, watch the most television and exhibit the highest levels of requests.


As consumer socialization research expands beyond childhood and beyond the United States, familiar questions are--and should be--asked about the extent to which findings and phenomena may be generalized beyond national/cultural boundaries. Obviously, no sweeping answer can be given. Rather, the question is: among many cultural similarities and differences, which are important to advance our understanding of theoretical and applied questions about consumer socialization processes and outcomes?

We have suggested some differences in stimuli, or influences on these processes, information-processing, and in consumer socialization outcomes, or responses. These would sees to be important aspects of consumer socialization which deserve particular attention from socialization researchers seeking to broaden the scope of their work beyond national/cultural boundaries. Research evidence is available to support some of the similarities and differences on these dimensions that we have proposed in this paper. We have presented some data on similarities and differences of two aspects of "stimuli," and on one aspect of responses.

The two aspects of stimuli included children's television watching behavior, and cluster analysis of parent-child interaction. Differences were found in the amount of television children watch in the three cultures studied here, in the amount of co-viewing with parents, and in attention to television advertising. The exploratory cluster analysis results suggest that there are characteristic family patterns which differentiate Japanese and American families, consisting of general parent-child interaction dimensions, such as affect, dependency and control, and consumption-specific dimensions, such as the existence of consumption rules, parental orientations toward consumer training, and asking and yielding behavior. In general, Japanese families are more circumspect and children are expected to learn through observation and trial and error. American families are more open, expressive, and children are expected to learn aspects of consumption through more purposive and expressive parental behaviors.

Finally, data were presented on one "response" or consumer socialization outcome. In this case, we hypothesized that children would not vary across cultures in their age-related patterns of purchase requests, and this hypothesis was confirmed by the data.

It is our hope that the kinds of dimensions suggested here will prove useful to consumer socialization researchers who turn their attention to other cultures.




Devereux, Edward D., Jr., Urie Bronfenbrenner and Robert R. Rodgers (1969), "Child-rearing in England and the United States: A Cross-National Comparison." Journal of Marriage and the Family 31, (May) 257-270.

Devereux, Edward C., Jr. (1970), "Socialization in CrossCultural Perspective: Comparative Study of England, Germany and the United States." In Reuben Hill and Rene Konig (Eds.) Families in East and West. Paris: Mouton.

Ezra, F. and Suzanne H. Vogel (1968), '-The Patterning of Emotions: Passive Dependency in Japan." In E. Rent Geiger (Ed.) Marriage and the Family. Boston: Little, Brown and Company.

Goslin, D.A. (1971), Handbook of Socialization Theory and Research. Chicago: Rand McNally Co.

Himmelweit, H.T., A.N. Oppenheim, and P. Vince (1958), Television and the Child. London: Oxford University Press.

Inomata, Satoru and Elliott McGinnies (1971), "Social Attitudes Among Japanese and American Teenagers." Psychologia 14, 158-169.

Isler, Leslie, Edward Popper and Scott Wart (1979), "Children's Purchase Requests and Parental Responses: Results from a Diary Study." Cambridge, Mass.: Marketing Science Institute.

Kohlberg, L. (1971), "The Cognitive-Developmental Approach to Socialization." In D. Goslin (Ed.), Handbook of Socialization Theory and Research. Chicago: Rand McNally Co.

Levitt, T. (1983) "The Globalization of Markets," Harvard Business Review, (September/October).

Mann, Jim (1986) 910 Years Later, Mao's China Disappearing," Philadelphia Inquirer, (September 7) 3-E.

Schaefer, Earl S. (1971) "Development of Hierarchical, Configurational models for Parent Behavior and Child Behavior," in J.P. Hill, (Et.) Minnesota Symposia on Child Psychology, Minneapolis: University of Minnesota Press.

Schramm, W., J. Lyle and E.B. Parker (1961) Television in the Lives of Our Children. Stanford, California: Stanford University Press.

Shigaki, Irene S. (1983) "Child Care Practices in Japan and the United States" Young Children, 38 (May) 13-24.

Ward, Scott, Daniel Wackman and Ellen Wartella (1977), How Children Learn to Buy: The Development of Consumer Information Processing Skills. Beverly Hills: Sage Publishing, Inc.

Wind, Yoram and Susan Douglass (1985) "The Myth of Globalization," Philadelphia: Wharton School Marketing Department Working Paper 86-025.