Wives, Husbands, and Children: Three Studies of Consumer Roles

ABSTRACT - Despite several aspects of their conceptual frameworks and methodology which could be strengthened, the three papers on consumer roles discussed here offer some insights furthering our understanding of marital roles in decisions and adolescent consumer socialization. Each study suggests a promising direction for further research.


Donald Granbois (1979) ,"Wives, Husbands, and Children: Three Studies of Consumer Roles", in NA - Advances in Consumer Research Volume 06, eds. William L. Wilkie, Ann Abor, MI : Association for Consumer Research, Pages: 377-380.

Advances in Consumer Research Volume 6, 1979      Pages 377-380


Donald Granbois, Indiana University


Despite several aspects of their conceptual frameworks and methodology which could be strengthened, the three papers on consumer roles discussed here offer some insights furthering our understanding of marital roles in decisions and adolescent consumer socialization. Each study suggests a promising direction for further research.


In his excellent synthesis and critique of household decision making research, Harry Davis commented that the many studies showing that two or more family members often are involved in joint purchasing and consumption behavior have had little impact on the "mainstream" of consumer research. He argued that this may help explain why associations found between dependent variables reflecting group behavior (such as brand purchase sequence patterns, automobile brands purchased, etc.) and predictor variables (such as attitude structure, personality, and information processing rules) gathered from individuals are often disappointingly weak. (Davis, 1976). One criterion for Judging the preceding papers by Burns and Ortinau and by Douglas is suggested by Davis' observation. Put simply, we can ask: "Do the findings and/or hypotheses suggested by the studies imply ways to strengthen research on topics typically structured by models of individual behavior?"

The potential contribution of the paper by Moschis and Moore seems to illustrate Davis' suggestion that we study the consequences of family consumer roles. Their independent variable, household communication structure, is analogous to the role structure types serving as dependent variables in many family studies. The "consequence'' (which is seldom pursued in typical consumer family role structure studies) is the impact on the consumer socialization process of children.

One function of the discussion which follows is to react to what seems to be strengths and weaknesses in each paper in terms of their conceptual frameworks, methodologies, and findings. A second, and perhaps more important role of the paper is to elaborate and speculate upon suggestions for further work which are stimulated by the papers.


Writers in consumer behavior have been far more prone to argue the potential importance of cross-national research studies than to actually conduct them. It is not surprising (but surely disappointing) to find only one truly comparative study in Douglas' literature review. Not only is it of significance from an applied perspective to understand similarities and differences where marketing or other influence programs may be transferred to other cultures by American firms and organizations. It also can be argued that better understanding of patterns in our own culture (especially those so universal as to be "taken for granted") is facilitated by comparative studies highlighting contrasts with distinctly different behaviors. The Douglas study is an ambitious exploratory study on a highly relevant and promising (but sadly under-researched) topic.

Her decisions to work with language-defined groups, to match English-speaking with French-speaking groups (whose behavior might intuitively be expected to be different but not too different), and to use more than one group in each language category all seem commendable and nicely defended in her discussion. Her objective might be described as explanatory: to discover patterns of differences in husbands' and wives' consumer roles which appear to be attributable to cultural differences between the two language groups. The notion of hierarchical objectives shows that explanation is a necessary achievement before prediction and influence can be successfully undertaken, and also warns that description must first be accomplished. Douglas' study makes some contributions toward this necessary description, but does not go far enough to permit more than tentative hypothesizing about French-English cultural differences as explanations for differing family role structures.

This conclusion is clarified by some aspects of her conceptual framework, variable formulation, and research methodology which may have contributed to her somewhat disappointing results. (Data analysis is not an element of any of the three studies requiring much comment. Like many current consumer studies, the papers here illustrate the application of analytical procedures far more sophisticated than the conceptual models and data involved. This state of affairs exists perhaps because consumer researchers benefit from methodological developments borrowed from a large number of behavioral and even non-behavioral disciplines). To discover needed modifications is indeed a major role of exploratory studies, and my comments mostly reflect and occasionally elaborate many things Douglas herself learned from the study.

Her lack of prior hypothesizing as part of developing a conceptual framework is somewhat surprising, given the familiarity of the two cultures of interest. Broad differences in values and resulting consumer behaviors among four English-speaking groups have been summarized by Sommers and Kernan (1967), and one would expect to find similar generalizations about French-language groups in the literature of cultural anthropology. Exploratory research need not be totally atheoretical, and some notion of possible expected differences could have facilitated some desirable narrowing of the scope of the study. The very ambitious framework resulted inevitably in much variability not attributable to cultural differences.

Although it isn't explicitly presented, Douglas' conceptual framework assumes variability in husband-wife roles along several dimensions:

a) roles may change over time, representing trends such as wives gaining importance due to growing career and work roles;

b) husbands' and wives' responses may differ due to biases and perceptual differences;

c) roles may vary across decisions and activities;

d) role differences may exist among demographic and socio-economic groupings;

e) role patterns may reflect cultural differences between French- and English-language groups;

f) role patterns may reflect differences among sub-cultural groups within each language group.

Since the study is the first of its kind, trend changes cannot be addressed. Differences in responses between spouses have been documented and discussed in several American studies. Douglas anticipates this in her design, but does not analyze husband-wife differences. Cultural differences are, of course, the main concern of the study. Commendably, aggregate role differences are not her target; rather, similarities and differences between cultures in each specific activity or decision were sought. Several writers have commented on the weakness of typologies of family role structure based on arbitrarily-selected activities and decisions lumped together to support numerical "indexes" of role structure. However, there is no evidence of systematic thought behind the selection of the 17 behaviors selected. Comprehensive frameworks of consumer behavior activities have been offered by Gredal (1966), Arndt (1976), and Reynolds and Wells (1977), which could perhaps be consulted in revising the list of behaviors asked about.

American research offers support for the notion that role structure differences relate statistically to certain socio-economic and demographic variables (such as family life cycle stage), although nothing like strong, easily-interpreted patterns seem to exist. Here the scope of the study could have been fruitfully narrowed, perhaps by purposely selecting subjects representing a relatively homogeneous segment defined both socio-economically and demographically.

Douglas' key dependent variable is captured by the question, "who was typically responsible," asked for each decision and activity. She follows Davis and others in referring to this as "involvement," a source of possible confusion since other researchers follow Krugman's (1965) conceptualization of involvement as a measure of importance or strength of concern. Besides response categories indicating "mainly husband" and "mainly wife," she (commendably) adapts from typologies of role structure the notion that each individual activity can be conducted jointly or by husband and wife acting individually with about equal frequency. It is unfortunate that pretesting didn't allow the perfection of an unambiguous question format here, since individual decisions conceivably could be performed autonomically for any of several reasons. Explanations may be situational (dry cleaning may be picked up by the person whose schedule for the day provides easiest access to the dry cleaner), may reflect fairness or equity (highly pleasant or strongly disliked activities may be done an equal number of times by each person), or may represent conscious avoidance of conflict that experience may reveal tends to accompany joint decisions or activity performance.

Methodological features weakening the study (which are understandable, given the ambitious scope and exploratory nature of the study) are the small, convenience samples; the very low response rates; and the pooling of husbands' and wives' responses. The latter seems to have little logical justification and may represent a violation of the "independent response" requirement of many statistical techniques.


The potential contribution of the research questions pursued by Burns and Ortinau is less apparent than the cross-cultural insights sought by Douglas, and their study is obviously less ambitious in scope. My interpretation of their objective is that they hoped to find explanation through interpreting and attributing meaning to groupings found to exist within essentially descriptive data. Unlike cross-cultural research, where one at least implicitly assumes the values and lifestyles which distinguish one cultural grouping from another will impact on the behavior being studied, the Burns-Ortinau approach starts with no tentative, expected patterns at all. The researchers indicate in their literature review that variations in role structure related to specific products and sub-decisions within each total purchase decision have been found earlier, but they did not derive hypotheses from the findings of earlier studies for testing. What they have done is to replicate earlier research in a new setting without showing either in their discussion of their research design or in their presentation of results how earlier knowledge is extended, modified, refuted or reinforced by their findings.

However, their paper represents a real contribution by developing a new way to explore patterns of marital participation in sub-decisions associated with durable goods purchases. Since the study is clearly intended to be exploratory, my comments on it should be interpreted as positive suggestions for designing further research, which I hope the authors will undertake.

The conceptual framework here, like that of Douglas, encompasses several decision areas, but is limited to five durable products. The six sub-decisions relevant for each of the products can be classified into the schemes suggested by Gredal (1966) and Arndt (1976). Arndt, for example, distinguishes between budget allocation, generic and variant selection decisions. Gredal's scheme identifies major and minor, individual and collective classifications of durable goods, two distinctions which, if recognized, might have encouraged the researchers to modify their selection of products so that each category would be represented. The study then would have provided the setting for putting to empirical test the role structure hypotheses advanced by Gredal.

The four household goods used include two furniture and two entertainment items, but no appliances. Nelson Foote (1966) has argued the importance of distinguishing between "time-using" and "time-saving," suggesting another desirable modification of products asked about. Except for dinettes, the products chosen all frequently exhibit multiple-unit ownerships within households, a factor likely to complicate the respondents' choice of response. Whose car, whether the television set in question is the family room console or a bedroom portable, whether the stereo is for parents or children, all may have important influence. Similarly, whether the hypothetical purchase would represent a first-time acquisition or a replacement could well affect expected marital roles to the extent these are influenced and modified by experience. Finally, while Douglas' selection of activities seems intuitively to be reasonably concentrated among those where adult preferences could be expected to dominate, three of the five products Burns and Ortinau selected -- auto, television, and stereo -- might well be subject to strong influence by children. While both studies repeat the unfortunate precedent established by earlier studies of ignoring children's roles, the omission seems especially dangerous here.

The dependent variable, wife's perception of "how she believes she and her husband would decide," requires respondents to imagine a hypothetical future event, a practice usually discouraged in discussions of questionnaire design. More troublesome is the ambiguity of the verb "decide." Seventeen years ago, James Morgan's (1961) review of existing role structure studies found nine alternative dependent variables. He concluded that none was satisfactory, since dimensions of power or dominance, motives, process and learning were likely to be confounded by single measures. The variable here surely suffers from this confounding, and one can only speculate about how respondents interpreted the meaning of "decide."

Each response explicitly relates to a single future purchase, complicating the use of a response category indicating an autonomic pattern over all the replications of the decision. However, the very infrequent repetition of the decisions of interest here makes this a less-serious problem than in the Douglas study.

Finally, the small convenience sample and the extreme heterogeneity of responses are methodological features the researchers themselves emphasize, although one could well argue the study is weakened rather than strengthened by the wide range of ages, education levels, lengths of marriage, occupations, etc. represented by the respondents. The study is far too small to permit rigorous examination of demographic differences.


Unlike the others, Moschis and Moore seek to explain individual consumer behavior, but their paper shares the others' concern with family variables. Family communication structure, a typology previously applied in research on the political socialization process, is investigated as a determinant of adolescent consumer socialization and media exposure patterns.

At least two important contributions are promised by research on this topic. Obviously, their findings provide additional understanding of the consumer socialization process of children, showing the role of family influence instead of the more commonly-investigated impact of mass media and peers. To the extent that family communication structures can help explain differences in adolescent socialization (and their paper is suggestive but incomplete in developing plausible reasons why competence is highest in pluralistic homes) this kind of investigation may eventually help in the development of consumer education programs directed at parents as well as children. Better understanding of children's responses to mass media (especially advertising) is of obvious relevance both for advertising management and for regulatory policy.

A second implication relates at least indirectly to the two papers just discussed. The role of children in family consumer behavior is still very poorly understood and often ignored. Moschis and Moore's conceptualization could be extended to predict the role of children's contributions to family decisions in each of the four family types identified in the communications structure paradigm. Such an extension could obviously be facilitated by further understanding of parents' communication and power structures in each of the types. For example, are egalitarian norms and Joint marital decision structures most common in pluralistic families, and do adolescents in such homes participate more in consumer decisions for family-oriented products than do children in families of the other types?

The support found for the hypothesized patterns of association between pluralistic structure and four indicators of adolescents' "sophistication" as consumers is provocative. One can't help but wonder what other aspects of adolescent consumer socialization might be similarly related to family communications structure. As in the other studies, the dependent variables here are incompletely explained and appear to be somewhat arbitrarily chosen. Far more investigation will be required before the full implications of pluralistic communications structure are realized.

Further research exploiting the other three cells of the paradigm also appears to be called for. Moschis and Moore provide evidence supporting three hypotheses about the relationships between socio-orientation and motives for media exposure and materialism, which appear to offer useful insights. McLeod and Chaffee (1972) report children from consensual families may be more "open" to persuasive communication, for example, and this is suggestive of only one of many possible directions for future hypothesis development and testing.


Douglas has already identified several needed studies in further investigating cross-cultural differences in marital role structure, each of which seems worthy of attention. My paraphrasing of her excellent discussion of these topics results in the following list, ranked in order of priority with respect to desirable timing (not necessarily importance):

1) Depth analysis of role norms and underlying beliefs and values found in each cultural group of interest;

2) Development of equivalent and comparable categories of socio-economic and demographic categories across the cultural groups to be studied;

3) Investigation of hypotheses developed in (1) above through study of carefully-chosen samples drawn from socio-economic and demographic strats in each of the cultural groups to be studied.

Although Burns and Ortinau have not spelled out their ideas for future studies quite as explicitly, the several suggestions made for improving their conceptual framework serve as a program for further developing their study of major decisions. Similarly, the usefulness of Moschis and Moore's FCS paradigm should be explored along the lines suggested in the discussion of their paper.

In all three studies, further conceptual elaboration of the dependent variables was shown to be desirable, and it can only be further emphasized here that concepts and findings from studies of individual consumer behavior may be of great value in this process. Work on information processing, types of evaluation processes, attitude formation and structure, and the relationships between values and choice criteria all may provide help.

The problems of sorting out power, influence, and participation and bias and perceptual problems growing out of self-reporting unfortunately still plague the use of questionnaires in role structure studies. However, it seems we can at least pose our questions so as to deal with explicitly-defined past purchases and intended actual future acquisitions or decisions rather than hypothetical situations. The area still lacks good methodological studies incorporating contrasting techniques. For example, time-activity analysis and observation studies of interaction both appear to be viable ways of studying marital roles in problem-solving and decision-making.

Until means are discovered for providing adequate financial support for research in the area (and here ACR can perhaps someday be involved in a clearinghouse role, bringing together potential contributors to research on specific topics of interest and researchers), we need to be especially careful to avoid overly-ambitious designs. Except when socio-economic or demographic characteristics are the special focus of a study, it makes sense to study one stratum of the population at a time.


Johan Arndt, "Reflections on Research in Consumer Behavior," in Beverlee B. Anderson (ed.), Advances in Consumer Research, Vol. III. Cincinnati: Association for Consumer Research, 1976, 213-221.

Harry L. Davis, "Decision Making Within the Household," Journal of Consumer Research, 2 (March, 1976), 241-260.

Nelson N. Foote, "The Time Dimension and Consumer Behavior," in Joseph W. Newman (ed.), On Knowing the Consumer. New York: Wiley, 1966, 38-46.

Karen Gredal, "Purchasing Behavior in Households," in Max Kjaer-Hansen (ed.), Readings in Danish Theory of Marketing. Kobenhaun: Einar Harcks Forlag, 1966, 84-100.

Herbert E. Krugman, "The Impact of Television Advertising: Learning Without Involvement," Public Opinion Quarterly, 29 (Fall, 1965), 349-356.

Jack M. McLeod and Steven H. Chaffee, "The Construction of Social Reality," in J. T. Tedeschi (ed.), The Social Influence Process. Chicago: Aldine-Atherton, 1972, 50-99.

James N. Morgan, "Household Decision-Making," in Nelson N. Foote (ed.), Household Decision-Making. New York: New York University Press, 1961, 81-102.

Fred D. Reynolds and William D. Wells, Consumer Behavior. New York: McGraw-Hill, 1977.

Montrose Sommers and Jerome Kernan, "Why Products Flourish Here, Fizzle There," Columbia Journal of World Business, March-April 1967.



Donald Granbois, Indiana University


NA - Advances in Consumer Research Volume 06 | 1979

Share Proceeding

Featured papers

See More


Ritual Scholarship in Marketing: Past, Present and Future

Cele Otnes, University of Illinois at Urbana-Champaign, USA
Linda Tuncay Zayer, Loyola University Chicago, USA
Robert Arias, University of Illinois at Urbana-Champaign, USA
Arun Sreekumar, University of Illinois at Urbana-Champaign, USA

Read More


Morality Matters in the Marketplace: The Influence of Morally Based Attitudes on Consumer Purchase Intentions

Andrew Luttrell, Ball State University
Jacob Teeny, Ohio State University, USA
Richard Petty, Ohio State University, USA

Read More


The Interaction Effect of Food Variety and Simulation of Eating on Consumers' Calorie Estimation

Liang Shen, Shanghai Jiao Tong University
Fengyan Cai, Shanghai Jiao Tong University
Ying Yu, Huazhong Agricultural University

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.