Advances in Consumer Research Volume 14, 1987 Pages 283-287
CHILDREN'S INFLUENCE IN FAMILY DECISIONS AND CONSUMER SOCIALIZATION: A RECIPROCAL VIEW
Karin M. Ekstrom, Washington State University
Patriya S. Tansuhaj, Washington State University
Ellen R. Foxman, Washington State University
[This paper represents the initial conceptual development for a cross-national project funded by the International Marketing Program to Assist Commodity Trade (IMPACT) at Washington State University.
In both family decisions and consumer socialization studies, research has to a large extent focused on the influence of parents on children. Because family communication and learning are not unidirectional, taking a reciprocal view of family decision making and consumer socialization will be a more realistic research perspective. That is, studying children's influence on parents, in addition to parents' influence on children, will yield a better understanding of these family phenomena. This paper reviews research on parent-child influence in family decisions and consumer socialization, and suggests propositions for future research on parent-child influence in both areas.
Many studies have shown that children's purchasing decisions are influenced by parents through learning (e.g., Douglas 1983; Filiatrault 1980; Moschis and Churchill 1977, 1978; Szybillo et al. 1977). Fewer studies have dealt with how parents may acquire consumer skills from their children. We feel that adopting a reciprocal view of family decision making will yield a fuller understanding of both interactive consumer learning in the family and family decision making as a whole.
The decision making process occurring in the parent-child dyad has been studied under the heading "consumer socialization". Since this is a relatively young field of study with multidisciplinary roots (i.e., family sociology, social psychology, child development and consumer behavior) there are several definitions of socialization. Researchers have had difficulty in arriving at a definition that is sufficiently precise to be empirically useful while maintaining its appropriateness to the wide range of behaviors, events and social conditions to which it is commonly applied (Gecas 1981; Tallman et al. 1983). In marketing, socialization has been defined as the process of learning consumer-related skills, knowledge and attitudes (Ward 1974). We adopt a somewhat broader definition where socialization refers to "the ways in which individuals learn skills, knowledge, values, motives and roles appropriate to their position in a group or a society" (Bush and Simons 1981, p.134).
Informal observation readily provides instances of purchase decision situations in which children have greater product knowledge than their parents. In such situations, they can logically be expected to express that knowledge and thus contribute to decision outcomes by two routes-direct expression of their preferences, and influencing parental preference by communicating new knowledge (reverse socialization). For example, in families considering the purchase of a home computer, children may have the most product knowledge because they encounter and use computers on a daily basis at school. Parents who are homemakers or who do not use computers in their jobs are likely to have little knowledge of computers and as a result are perhaps more open to children's contributions in the decision making process.
Recent changes in demographic and household structure may have significantly increased children's impact on parental decisions and general involvement in family decision making. For instance, the rising number of single-parent households may lead to increased decision involvement for children. It is estimated that by 1990 only 69 percent of the children in the United States will be raised by both parents (Glick 1984). Currently, approximately one-fifth of all families with children are headed by women as single parents (Kantrowitz 1986). Additionally, many households today have both parents working full time. These alternative family structures face on the whole the same number and type of family decisions to make, but are likely to have less parental time to devote to these decisions than a standard nuclear family. Such "time poor" families may permit or actively encourage increased child participation in family decision making.
The purpose of this paper is to:
(1) review research on the parent-child dyad of family purchase decisions as well as consumer socialization: and
(2) advance propositions for future research on parent-child decisions from the perspective of "reverse socialization," the process by which parents acquire consumer skills and knowledge from their children.
We believe that family decision making and consumer-socialization are closely linked; that is, decision making in the parent-child dyad often serves as a vehicle for consumer socialization of both parents and children. We begin with a brief discussion of the types of influence focused upon in family decision making and consumer socialization studies. Consumer variables that have been fount to be related to family decisions and consumer socialization are then reviewed, and propositions are presented regarding the effect these variables are expected to have on reverse socialization. Finally, methodological considerations are briefly discussed.
TYPES OF PARENT-CHILD INFLUENCE
The types of influence examined in family decision making and socialization studies can be broadly differentiated. Influence in family decision m=king studies characteristically focuses on whether an individual's wishes are acted upon in a particular decision situation. Studies attempt to measure preference, decision dominance, and occasionally the exercise of social power (e.g., Cox 1975; Olson 1969; Olson and Rabunsky 1972; Shuptrine and Samuelson 1976). Previous consumer socialization studies, on the other hand, view influence from the perspective of a socialization agent. As such, they attempt to measure communication and the learning of knowledge and skills (e.g., Moschis and Churchill 1977, 1979; Moschis et al. 1984; Wart 1974; Wart et al. 1977). As a socialization agent, a child influences the parents behavior by teaching them new knowledge and consumer skills. A child's influence can actually lead to internalized and lasting changes in values, self-concept and consumption behavior on the part of the parents. The "yielding" discussed in parent-child decision making studies (i.e. Atkin 1978; Berrey and Pollay 1968; Roberts et al. 1980; Wart et al. 1977) is merely compliance with the child's request, and may not involve internalization of new values, skills, or roles.
Forming a purchase preference can be viewed as a learning process. When family decision making studies request individuals to state their purchase preferences or purchase influence, they are asking for information on the outcome of a process, not the process itself. Such questions are more simple to ask and answer than those which request individuals to remember the preference formation process and report on the roles others (i.e., socialization agents) played in it.
Children and Family Decisions
Host studies examining children's influence fall into this category. Influence is measured by opinions, statements of personal preference, and decision outcomes; not by learning or knowledge acquisition. Atkin (1978) observed children between the ages 3 and 12 in 20 supermarkets. In two-thirds of the cases, the child expressed a desire for a particular cereal; and in most of those cases, the parent agreed to the cereal selection. The authors concluded that parents were more likely to yield to product requests when the product was primarily for the child's consumption. Roberts et al. (1981) found mother's attitudes and perceptions of children's influence to be related to economic, health, and child-rearing philosophy factors (i.e., liberal- versus conservative-orientation). Larger numbers of preference statements by children were associated with a higher rate of product usage in the family. In this study as well as the others discussed above, children's influence was measured by their statements of personal preference and the extent to which these matched purchase outcomes.
Children and Socialization
Moschis and Moore (1979, 1984) discuss consumer socialization based on two types of learning. The first, social learning, refers to socialization as a function of environmental influences on a person. Various socialization agents in the environment (such as parents, media, and peers) act as learning media. The second type of learning, cognitive development, is a cognitive psychological process of adjustment to one's environment. Essentially, cognitive development refers to learning which occurs as a function of information processing maturation. Moschis and Moore (1979, 1984) combine both the cognitive and the social learning perspectives in their studies.
Another type of child's influence on parents concerns what we propose to label as "reverse socialization," or parents' acquisition of consumer skills and product information from their children. Most research in child development has been from a unidirectional viewpoint in which the effects of parents' behavior on children's socialization has been studied. This assumes a simple asymmetrical causality model where parents shape children (Cromwell and Olsen 1975). The child, however, may also shape his/her parents. Bell (1968,1971) has criticized the view that the parent is the initial agent of culture and the child is the object. Be views parent and child as a social system in which each participant's responses constitute stimuli for the other. The effect of children in socialization of their parents is emphasized in his study although it is from a unidirectional viewpoint.
The importance of considering the reciprocal nature of parent and child roles was stressed by Brim (1957). Also, the norm of reciprocity was found in an empirical study on intergenerational relationship and family development performed by Rill et al. (1970). Brim (1966, 1968) discussed socialization as a learning process which occurs throughout a person's lifetime since cognitive development never ceases (though the amount of learning is greater at an earlier age than at a later age). Bronfenbrenner (1979), Zigler & Child (1973), and Tallman et al. (1983) have emphasized that socialization occurs in the course of a relationship and involves learning as well as teaching. Although parents have the socially prescribed role of teachers while children are viewed as learners, it may be more realistic to describe parent-child socialization as one in which parents and their children are both teachers and learners. We therefore believe it is important to study not only how children and adolescents are socialized by their parents, but also how parents learn from their children through a reciprocal socialization process.
Examining television as a socialization agent, Churchill and Moschis (1979) found a significant relationship between amount of intrafamily communication regarding consumption and the extent to which adolescents hold economic motivations for consumption. A reciprocal relationship in terms of children socializing their parents after watching television has been discussed to a lesser extent, but not empirically demonstrated. Moschis and Mitchell (1986) found no significant relationship between an adolescent's exposure and a more influential role than parents in mentioning the need for the product, discussing the purchase of products, deciding what should be purchased, or actually buying the product.
Factors Affecting Reciprocal Socialization
This section examines factors expected to affect the reciprocal socialization process. First, previous studies on family communication between parents and children in the consumer behavior context will be discussed. Then, background factors including family structure and socio-economic characteristics will be reviewed. Personal resources, product-related factors, and satisfaction with decision outcome are also examined. As previous research is reviewed, we offer new propositions regarding how each factor is expected to affect children's influence on parents.
Family Communication Between Parents and Children
Family communication is expected to affect children's influence in family decision making. McLeod and Chaffee (1972) developed a typology that characterizes parent-child communication structure. In a socio-oriented communication environment the child avoids controversy and does not argue, since he or she does not want to risk offending others. In a concept-oriented communication structure, the child is encouraged to develop his or her S own ideas. The presence or absence of these two communication patterns describes four family types: laissez-faire families exhibits low levels of both patterns; protective families stress socio-orientation and avoid conflicts; pluralistic families have a concept-orientation and encourage communication; and consensual families emphasize both the socio-oriented and the concept-oriented view.
Moschis (1985) suggested that a concept-oriented family communication structure would foster greater participation in family decisions, while a socio-oriented structure would deter participation. An empirical study by Moschis et al. (1984) found that pluralistic families were more likely to use positive reinforcement and less likely to engage in negative reinforcement. Parents in consensual families used both types of reinforcement, while protective families were more likely to use negative reinforcement. Laissez-faire families used little reinforcement of any type. The authors also found that a socio-oriented communication pattern was negatively related to an adolescent's influence on his or her parents in deciding on a purchase and in actual purchases.
The studies discussed above demonstrate a relationship between family communication patterns and how children are socialized. Considering socialization as a reciprocal process, family communication pattern is also expected to explain how parents are socialized by their children. This has been dealt with in a few studies but only indirectly. As noted above, Roberts et al. (1981) found that liberal versus conservative child-rearing orientations affect the extent to which children are described by their mothers as having influenced purchase decisions.
It is expected that a child in a socio-oriented family environment will be less likely to influence his or her parents, or act as a socialization agent with respect to them, than a child in a concept-oriented communication environment. The reason for this is that a child in a socio-oriented home is not expected to interfere and argue for his or her own ideas as readily as a child in a concept-oriented family, who is expected to develop his or her own ideas. The following proposition is therefore proposed:
P1: Children whose family communication pattern is characterized by a high concept-orientation will influence (socialize) their parents more than children whose family communication pattern is characterized by a high socio-orientation
When studying family decision making, it is important to recognize and take into account ongoing changes in family structure. The statistics mentioned earlier on the rising number of single-parent families imply changes in children's relative influence in household decision making. A child in a single parent home might be allowed to participate in family decisions to a greater extent than a child in a nuclear family. The single parent may, for example, be more receptive to advice and opinions regarding his or her own clothes and products intended for the home.
Kourilsky and Murray (1981) examined the effect of economic reasoning on satisfaction within the family and found that single-parent families exhibited a higher level of economic reasoning and satisfaction compared to two-parent families. A possible explanation for this lies in the reallocation of decision tasks in families that have only one parental decision maker. Sinkula (1984) pointed out that quasi-households (or "broken families") are dynamic because separated parents might maintain a number of households during the separation period, take responsibility for the children only part of the time, and experience passages through the revised family life cycle at varying times. This should be recognized when research is tone regarding single parent households.
Qualls (1981, 1984) discussed the impact on changing sex roles in family decision making. He suggested that changes in family member perceptions regarding sexual gender and division of labor have caused a shift in how decision making responsibility is allocated. He argued that a family member's sex role orientation determines the degree of family decision interaction: a sex-role modern family showed a more egalitarian distribution of family decisions, while a sex-role traditional family reach decisions through past role classifications. Heffring (1980) recommended more research on sex role orientations and family ideology, pointing out that sex-role egalitarian families are more likely to diwide tasks, whereas sex-role traditional families are more likely to delegate tasks. The limited research on family structure, then, suggests that children's influence and participation in family purchase decisions will be limited in sex-role traditional families; children will have greater decision influence in sex-role modern families.
A single parent's suffering from task overload has been discussed by Glasser and Navarre (1965). The single parent might face difficulties in adjusting if the marriage has hat a traditional sex-role orientation. If there is no support system available, the household will be less systematically organized and exhibit a higher level of family disorganization (Hetherington et al. 1977). Children in such families might replace the traditional support system; the resources they contribute may lead to a higher degree of participation and influence in family decision making. Furthermore, single parent mother-headed families often face financial difficulties. Bane (1977) reported that 51.5 percent of children under 18 in female-headed families were in families with incomes below poverty level. Children who earn an income and contribute to the family's income can be seen as having resources. The resources they provide might also lead to a greater influence in family decisions.
Based on the research discussed above, the following relationships between family structure variables and children's influence in family decision making are pro posed:
P2: A child's influence varies with family structure:
1. A child in a single parent family will have more influence on family purchase decisions than a child in a two-parent family.
2. A child in a sex-role egalitarian family will have more overall decision influence than a child in a sex-role traditional family.
A number of researchers have examined the effect socio-economic variables (mainly social class and income) have on socialization and children's influence in family decision making. Moschis and Churchill (1978) showed that family environment influences newspaper readership, which in turn influences the adolescent learning process. Their study confirmed that social class affects socialization. Consumer learning might occur as a result of structural factors that are present in higher socio economic classes (more opportunities for consumption). Different degrees of learning might also occur due to the different socialization practices that are emphasized by different socio economic families.
Moschis and Moore (1979) and Moschis and Churchill (1978) later did show that children from higher socio-economic backgrounds have more opportunity for consumption and are more aware of products available in the market. Roberts et al. (1981) found that economic factors (primarily family financial status) affected the extent to which mothers perceived children as having influenced purchase decisions. Atkin (1978) also found a difference between social classes in children's attempts to initiate cereal purchases. The amount of child initiation of purchase of breakfast cereals was higher for children from the middle class than for children from the working class. However, Moschis and Moore (1985), found no difference in the extent to which middle-class adolescents and lower-class adolescents discussed consumption with parents.
Socio-economic factors may help explain the extent to which children socialize their parents, as well as children's influence in family decision making. Families with higher socio-economic backgrounds may provide better opportunities for influence and be more receptive to their children's opinions. One reason for this is that they have more opportunities for economic consumption, which means that children have a greater absolute number of opportunities to influence consumption decisions. A second possible reason is that middle and upper-middle class families may encourage decision-making approaches that would result in greater child participation in purchase decisions, specifically, joint decision making and autonomic decision making (Evans and Smith 1969). It seems that a family with high education, more financial resources and/or more time might allow a child to express his or her opinions more and therefore exert more influence on decisions in the family. Accordingly, the third proposition is stated as follows.
P3: Children whose families are characterized by a high socio-economic status (i.e., high income, high social class, highly educated parents) will have moore influence in family purchase decisions, both in terms of relative influence and as socialization agents.
Resource theory developed by Blood and Wolfe (1960) asserts that the comparative resources of parents and children will determine who dominates in family decisions. Children's resources may include income contribution, employment status, education, grades, or parental love and affection. The greater the individual's resources, the greater his or her decision making power. In the past, studies based on resource theory have explained the husband's dominance in family decisions on the basis that he earns the most income. However, the change in family structure with more working wives has led to increased influence for wives in family decisions.
Similarly, children's personal resources should affect their relative influence in family decision making. For instance, being the first-born child may be regarded as a personal resource. Churchill and Moschis (1919) expected birth order to be related to the amount of communication between adolescents and parents on purchase decisions. They hypothesized that a first-born adolescent would interact more frequently with his or her parents regarding consumption matters. Their study found a positive but nonsignificant relationship between birth order and family communication: the first-born child does not necessarily communicate more with parents than do other children. Perhaps researchers should also measure specific behavioral influences (e.g., the number of times the child goes shopping with the parents, how active he or she is in suggesting product features or brands).
In terms of financial resources, Moschis and Mitchell (1986) proposed that young people who earn more money outside the home are less likely to mention the need, discuss the purchase, decide whether to buy and also to be independent in purchasing products for individual or family use. Their findings supported the idea that adolescents with more money have more purchasing independence but failed to support expectations regarding the other stages of the purchase process. The greater the child's personal resources, the greater his or her influence in the family decisions. It is therefore proposed that:
P4: A child's relative influence in the family decisions will be positively related to the child's personal resources.
Product Related Factors
Product-related variables expected to affect children's relative influence in family decisions and the extent to which they function as socialization agents for parents include product importance and product knowledge. Consumer socialization studies have shown that parents' influence on children varies depending on a child's knowledge of the product and perceptions of product importance. For instance, Moschis and Moore (1979) found that parents are preferred twice as much as other information sources for purchases of wrist watches and pocket calculators. Parents' influence was found to increase with high-risk products or products that the child knew little about. The same study also found that parents had relatively low influence at the product evaluation stage of the decision process; price and brand name were the most important evaluative criteria at this stage. It appeared that product attributes outweighed social influences, although this varied depending on product type.
Roberts et al. (1981) found that a higher degree of children's influence was associated with a higher rate of product usage by the family. A reasonable explanation for this finding is that children more frequently attempt to influence decisions on products they consider to be important. Also, parents may consider products used frequently to be less important purchases, and r-y be more open to children's influence attempts in these decision are s. Krishnamurthy (1981) also suggested that the person who is more involved with a product will be more active in the product decision process and have a greater effect on the dec is ion outcome.
To summarize, a child is expected to be more influential in family decision making regarding a product he or she perceives as important. The reason for this is that he or she will likely become more actively involved in the family decision process. It is also reasonable to expect, however, that parents' perceptions of product importance and parents' product knowledge will moderate children's influence. For example, parents might be less likely to accept children's influence on products for parents' personal we but more likely to accept children's influence on products for children's exclusive use. The following proposition is therefore stated:
P5: A child will have greater influence in family decision making for product purchase decisions:
5.1. that he or she considers important, and
5.2. about which he or she has a high level of product knowledge.
Satisfaction with Family Purchase Decisions
After discussing how product importance and knowledge relates to involvement in family decision making, it is logical to address the issue of whether increased involvement in the decision process leads to increased satisfaction. Burns' (1976) study found that spouses who are highly involved in a decision derive satisfaction from their influence. Kourilsky and Murray (1981) suggest three situations when a family member may be dissatisfied with the group decision-making process. First, the cost of participation has not been fully explained to the individual or the group so they knew what they had to forego as a result of the pending decision. Second, the family member may not have participated sufficiently in the ranking of benefits. Finally, the family member may believe that his or her self-interest is inconsistent with family cohesiveness. Their empirical findings supported the hypothesis that economic reasoning would increase satisfaction; that is, they found that the economic and psychological involvement concomitant to cost-benefit analysis did contribute to child and parent satisfaction.
The question can then be asked if the child who is allowed to participate in family decision making is more satisfied than a child who is not allowed to participate. Robertson et al. (1985) studied parent-child interaction effects on children' s satisfaction in the age group 6-11 years old. Satisfaction was found to be greatest among older children and children from homes with high parent-child interaction. The findings also indicated that parent-child interaction related positively to satisfaction, especially among younger children. This might imply that older children are less responsive to parent-child interaction and learn more through direct experience and modeling parental or peer behaviors. Older children also showed more disappointment when they did not get what they expected. Additionally, Shaffer (1971) argued that households sometimes plan for satisfaction for the household as a unit and this might not lead to maximized satisfaction for any one member. This possibility should not affect our final proposition, which is stated in relative terms:
P6: For products he or she perceives to be important, a child who participates in family decision making will be more satisfied with the decision than a child who does not participate in the decision process.
Proposing the study of children as family decision participants and socialization agents with respect to parents invites all the problems of studying dyadic family decision making previously discussed (e.g., Burns and Granbois 1980; Davis 1976; Dunsing and Hafstrom 1975; Foxman 1986; and Heffring 1980), plus a number of new ones. As in husband-wife studies, methodological issues that relate to children's roles in the decision process can be considered under four headings: questions regarding the meaning, of influence, questions relating to respondents, questions relating to decision topic, and questions relating to data collection mode.
An individual's influence in family decision making studies has often been conceptualized as a dichotomous construct: if the individual's prior preferences match the decision outcome, the individual is said to have influence, or to dominate; otherwise the individual does not have influence or dominate. Examining studies of parent child decision making suggests a further distinction would be useful: that is, a child's influence through parental "yielding" should be studied separately from children's influence as a result of reverse or reciprocal socialization. "Yielding" indicates parental indulgence rather than child influence and ideally should be measured separately from influence resulting from children's effect on parents' knowledge, consumer skills, or viewpoints. Operationalizing this distinction in a meaningful and reasonably natural way appears to represent a considerable challenge for researchers in this area, however.
In dyadic family decision making studies, a long-standing issue has involved which spouse is the appropriate respondent in family decision making studies and whether it is necessary to obtain responses from both spouses. There is movement in recent studies to obtain data from both spouses, but how to combine or weight this dyadic data is by no means a settled issue.
Cromwell and Olsen (1975) criticize parent-child interaction studies to have focused only either on a parent (usually the mother) or on a child. A small percentage of studies have involved both the mother and the child. There is a need for research which include the entire family. It seems clear that family decision making studies which focus on children's roles should include responses from at least one child and both parents. For example, some of the propositions advanced in this paper, especially the ones relating to reverse socialization, would ideally use a measure of parents' perceived learning from the child socialization agent as a dependent variable, and would seek to predict and explain levels of this variable based on certain characteristics the child perceives himself or herself to have. The issue of combining or weighting respondent data would, of course, continue to be a problem in decision making studies that include children. There are also two special problems associated with this type of study--which age range to study and whether it is Productive to study certain age groups.
Broadly speaking, the term "children" in the U.S. can include all dependent offspring in a family. Because children's cognitive abilities a-e rapidly changing as they mature and because these changing abilities are likely to relate to their influence in the family, it is desirable to include a broad range of ages in child decision making studies. To do so, however, necessarily makes such studies administratively more difficult and analytically more complex. Age would at least have to be controlled for in the analysis of empirical data from a broad age range of children. In cases where cognitive development stages can be identified and appear to be closely related to the phenomenon being studied, it would be advisable to group responses by stage and analyze by group.
Also, there is a real question whether younger children are useful respondents. They may not understand either the relationships or the concepts under investigation and in some cases will lack both familiarity with and interest in the products being studied. A four-year-old, for example, may know what cars are but may lack knowledge and interest in the product per se.
We believe the age range starting from adolescence to be more appropriate for investigating consumer socialization of parents. While babies and younger children may have a powerful influence in socializing adults into parental roles, this influence does not focus directly on acquisition of consumption s}ills. Furthermore, including children younger than adolescent age (less than 11 years old) is not appropriate for consumer socialization studies; such children are not fully cognitively developed (Elkind 1968; Mussen et al. 1969; Piaget 1970) and have been demonstrated not to understand economic concepts (Strauss 1952) and consumer s}ills related to information processing (Roedder 1981; Wackman and Wartella 1977).
It is apparent from previous research that one likely source of variance in children's decision influence will be the particular product under consideration. It will be desirable to gain some understanding of how perceived product importance (from both children's and parents' points of view) affects children's decision influence. It nay, however, not be advisable to aggregate children's influence measurements across these different product categories.
Data Collection Method
As with other dyadic family decision studies, observation would ideally yield an accurate picture of the decision making process. In fact, however, it has not in the past: its intrusiveness as a data collection method has sensitized respondents, and its poorly developed methodological procedures have made it difficult to use. Self-report methods have been most used in past family decision studies; they are likely to remain the method of choice for studies of children's influence in family decisions. However, the drawbacks of self-report methods, especially with younger children, must be taken into account.
SUMMARY AND CONCLUSION
To investigate family decisions between parents and children more realistically, we believe that it is necessary to adopt a reciprocal view of consumer socialization of how both parents and children may learn from each other. A significant amount of research exists regarding parents' influence on children. There is, however, a need for more research regarding children's influence on parents. The propositions advanced suggest that a child's influence differs under different circumstances. The type of family communication environment will affect a child's potential influence in family decisions. Family structure (i.e., typical two-parent or single-parent family, and sex-role) is also expected to affect a child's decision influence, as will various household socio-economic characteristics. We also propose that a child's influence will depend on his or her personal resources. The product's importance and the family's knowledge about the product to be purchased are expected to influence the child's involvement in family decisions as well. Finally, we believe that children who are more involved in family decisions will experience greater satisfaction with decision outcomes. The research we propose regarding children's influence on parents will advance the study of family decision making. It will do so by yielding a better understanding of reciprocal family interaction and hence of the way in which the household unit actually functions.
Atkin, Charles K. (1978), "Observation of Parent-Child Interaction in Supermarket Decision-Making," Journal of Marketing, 42 (October), 41-45.
Blood, R.O. Jr. and D.M. Wolfe (1960), Husbands and Wives: The Dynamics of Married Living, Glencoe, Ill: Free Press.
Brim, Orville G. (1966), "Socialization Through the Life Cycle," in O. Brim and S. Wheeler, eds., Socialization After Childhood, New York: John Wiley & Sons, Inc., 3-49.
Brim, Orville G. (1968), "Adult Socialization," in Socialization and Society, ed., J. Clausen, Boston: Little, Brown & Co.
(Additional references may be obtained from the second author).