Mothers' Attitudes and Perceptions of Children's Influence and Their Effect on Family Consumption

Mary Lou Roberts, Boston University
Lawrence H. Wortzel, Boston University
Robert L. Berkeley, Boston University
ABSTRACT - A model is proposed in which a range of mothers' attitudes affect the amount of influence they perceive their children to have and the amount of influence, in turn, affects the amount of family consumption. Findings indicate that et least three attitudinal dimensions -- economic, health-related and liberal versus conservative -- affect the degree to which mothers report children are influential. Degree of influence had a significant effect on the amount of family usage of a set of products, all of which are appropriate for consumption by a single individual. Areas in which further research is needed are discussed, as are managerial implications.
[ to cite ]:
Mary Lou Roberts, Lawrence H. Wortzel, and Robert L. Berkeley (1981) ,"Mothers' Attitudes and Perceptions of Children's Influence and Their Effect on Family Consumption", in NA - Advances in Consumer Research Volume 08, eds. Kent B. Monroe, Ann Abor, MI : Association for Consumer Research, Pages: 730-735.

Advances in Consumer Research Volume 8, 1981      Pages 730-735

MOTHERS' ATTITUDES AND PERCEPTIONS OF CHILDREN'S INFLUENCE AND THEIR EFFECT ON FAMILY CONSUMPTION

Mary Lou Roberts, Boston University

Lawrence H. Wortzel, Boston University

Robert L. Berkeley, Boston University

[The authors wish to thank Needham, Harper & Steers Advertising, Inc. for providing access to its 1979 Life Style Study data which were used in this project.]

ABSTRACT -

A model is proposed in which a range of mothers' attitudes affect the amount of influence they perceive their children to have and the amount of influence, in turn, affects the amount of family consumption. Findings indicate that et least three attitudinal dimensions -- economic, health-related and liberal versus conservative -- affect the degree to which mothers report children are influential. Degree of influence had a significant effect on the amount of family usage of a set of products, all of which are appropriate for consumption by a single individual. Areas in which further research is needed are discussed, as are managerial implications.

The role of children as influences in family purchasing decisions is an accepted part of the conventional wisdom of marketing. Empirical documentation of this role, however, is scant. Most studies have focused on only one product (Berey and Pollay 1968, Caron and Ward 1975, Frideres 1973), a small set of products (Jenkins 1978, Mehrota and Torges 1976; Moschis and Moore 1979; Popper 1979, Szybillo, Sosanie and Tenebein 1979, Szybillo and Sosanie 1976, Ward, Popper and Wackman 1970) or a single consumption decision (Nelson 1978), although at least one study did investigate a relatively large array of products (Ward and Wackman 1972).

The paucity of studies is not surprising in view of the difficulties of collecting valid and reliable data on family decision processes and consumption behavior (Davis 1977). The well-documented problems in collecting data on husband-wife influence processes are exacerbated when one tries to collect data on the input of children, especially young ones, to the process (Robertson and Feldman 1975, Rossiter 1978, Szybillo, Sosanie, and Tenebein 1979, Ward 1978). Hence, the important question of the effect of children's influence on the type and magnitude of family consumption remains empirically unanswered.

The existence of a data base from a large representative sample makes it possible to approach this question in an exploratory vein. In so doing, we are suggesting a simple model in which parental attitudes, in this case the mother's, will effect the amount of influence she perceives her children to have, and that this perception will, in turn, affect the amount of that product category consumed by the family.

In one of the earliest studies in this field Berey and Pollay (1968) identified the attitude child-centeredness as important in determining whether or not the mother yielded to the child's influence attempts. Ward and Wackman (1972) later found that the age of the child was a determinant of the frequency of parental yielding. The importance of the age variable is obvious, but other parental attitudes might prove to be of equal or greater value in explaining the success, or lack thereof, of children's influence attempts. Genera/ attitudes toward family financial matters, media and society all seem to be possible determinants of parental yielding. Specific attitudes toward food prices and nutrition would seem to have special relevance to yielding to requests for food products. The importance mothers place on consuming, or not consuming, certain types of food and on nutrition in general should affect their willingness to yield to children's requests. In terms of both the effect of attitudes on influence and of influence on consumption, it is necessary to control for the age of the child.

Looking at our basic model from the mother's viewpoint is appropriate for several reasons. The mother is clearly a primary socialization agent during the early years of a child's life, though it is equally clear that peer groups become increasingly important as the child grows older. Peer groups may achieve dominant influence during adolescence (Moschis and Moore 1979). Early in the child's life, the mother purchases most of the products consumed by the child. As they become older children obtain more discretionary spending power and adolescents may actually purchase many, if not most, of the products they consume singly. However, the preponderance of evidence indicates that, throughout the family life cycle, the mother is the chief purchasing agent for products consumed jointly by various members of the family. The provision of food, especially, is an important part of mothering throughout the family cycle. If however, other family members, in this case children, do act as influences, their influence may affect both the type and amounts of products actually consumed by the family.

Looking at the model from the mother's viewpoint suffers from the limitations of monomethod data collected from a single family member (Szybillo, Sosanie, and Tenebein 1979). However, these data have the advantages of robustness and a degree of replication which provides a useful reliability check.

SAMPLE AND METHODOLOGY

The data were obtained from the 1979 Needham, Harper and Steers Life Style Study, which uses the Market Facts mail panel. Needham, Harper and Steers describe the data base as follows:

"The Market Facts' mail panel is balanced for geographic region, age, income, and degree of urbanization. The very poor, the very rich, the transient, and minority populations are not well represented in the panel. Needham, Harper and Steers place a further restriction on the sample by requiring all individuals to be married. This latter restriction, coupled with the general characteristics of the Market Fact's mail panel, tends to confine the representativeness of the Life Style sample to stable, middle class households... (it) has proven to be an effective barometer of middle America."

The research questions to be answered using these data are:

1.  Do mother's attitudes toward a variety of family-related and social issues influence their perceptions of the amount of influence their children have on their brand choices in selected product categories?

2.  Does the amount of influence children have affect the amount of family consumption in that particular product category?

Data for mother's attitudes (Table 1) was obtained from the AIO Section of the questionnaire. Since the influence questions dealt primarily with food products, although they also included some items of children's clothing, we considered it desirable to include specific attitudes towards food (Concern for Nutrition), food prices, and money matters in general (Price Conscious, Family Financial Matters), and homemaking (Involved Homemaker). We also included more general measures of attitudes, social attitudes toward media (Concerned About Sex and Violence) and toward society. Some of the social attitudes seem to be focused inwardly on personal and family concerns (Traditional, Family Togetherness, Concerned About Our Society), while others suggest a reaching out for new ideas and experiences (Cosmopolitan, Venturesome).

The attitude scales were constructed by judgmentally selecting from the approximately 200 available AIO items, measured on a six-point agree-disagree scale; a set of items which appeared to tap the attitude of interest. Each set of items was then subjected to reliability analysis. Unreliable items were removed until examination of the item-total correlations and matrix indicated that no further improvement in alpha was possible. Cronbach's alpha is presented twice for each scale, once for all women in the sample (n=1601) and again for the subsample who had children or grandchildren aged 0-19 (n=1150). All remaining data deals with the subsample only.

Children's influence on brand choice for 35 products was measured on a four point 'Almost All the Time' to 'Never' scale. Because of the lack of precision in this measure, the responses were factor analyzed (Table 2)- After varimax rotation, four factors were obtained which explained over 47% of the variance, a quartimax rotation was also carried out to see if all loadings could be forced onto one factor. Quartimax rotation also produced four factors, all having high correlations with the original four (unrotated) factors. Results of the varimax rotations were therefore used in the analyses.

TABLE 1

MOTHERS' ATTITUDES SCALES

TABLE 2

CHILDREN'S INFLUENCE FACTORS

The four influence factors, which bear strong resemblance to those in an analysis of earlier Needham, Harper and Steers data (Mehrota and Torges 1976), were called Children's/Pet Foods, Gum, Clothing/Cereal/Cookies, and Sweets and Snacks. The analysis consists of cross-tabulations of the attitude scale scores against the influence factors followed by cross-tabulations of the influence factors against reported family use, measured on a six-point 'Less than once a Month' to 'More than once a Day' scale, of all available products in the influence factors. Limiting our analysis to a relatively few scales and factors tends to preclude finding a few significant relationships by chance alone.

RESULTS

In analyzing the data we used cross-tabulations rather than correlations because the influence categories and the use frequency categories do not, strictly speaking, constitute scales in which categories are equidistant. The predictor variables, which are the mean scores from each attitude scale, have been collapsed into three categories representing high, medium and low levels of agree-meat with the attitudes making up the scale. The criterion variables represent, respectively, four influence categories and three usage frequency categories.

The analysis resulted in the production of forty-four three by four tables representing attitude/influence relationships. Space limitations make it impossible to present all of these data; therefore we will present only the probability of significance ( p < ) chi-square statistic for each table. (A complete set of tables is available from the authors on request). The analysis was conducted using as the research population all mothers and grandmothers who reported children's influence.

The Relationship between Attitudes and Perceived Influence of Children

Table 3 shows the chi-square relationship between all of the attitude scales and each of the four product categories. All of the statistically significant relationships are linear. The pattern of results, while reasonable, is not always easily explainable.

High concern over Family Financial Matters predicted low levels of child influence for all product factors except Gum, yet Price Consciousness predicted low influence levels only for Children's/Pet Foods. Thus, one dimension which determines the extent of children's influence is economic, but it appears that general economic status is of more importance than specific price consciousness. Concerned about Nutrition predicted children's influence in three of the four categories. The fact that it did not predict for chewing gum may simply reflect mother's recognition that gum neither adds to nor subtracts from the child's nutritional well-being and is therefore of little consequence.

Some of the social-attitude-related scales also demonstrated significant predictive power. Although Traditional Orientation predicted a low degree of influence for Clothing/Cereal/Cookies and Children's/Pet Foods, high Venturesomeness predicted higher degrees of influence for Children's/ Pet Foods, Gum and Clothing/Cereal/Cookies. A higher degree of Concern about Our Society and Family Togetherness (conservative dimensions) predicted significantly lower degrees of influence for all four product categories. It is not unreasonable to characterize these scales as constituting a rough measure of a liberal versus a conservative orientation toward child rearing and thus to conclude that a conservative orientation is indicative of a smaller role for the child in purchase decision-making.

Thus, there are at least three dimensions which affect the extent to which children are described by these mothers as having influenced purchase decisions for four groups of products. One dimension is economic, the second is health-related (nutrition) and the third is a liberal versus a conservative orientation. We shall next explore, by looking at the relationships between influence factors and product use, whether this influence results in greater purchases of the products by the family.

Controlling for Children's Ages

A further analysis was performed, controlling for the presence or absence of children in four age categories. The categories used were 0-2, 3-7, 8-12 and 13-19 years of age, the first two of which approximate the stages of cognitive development. In the majority of cases where the two-way crosstabs were significant, a consistent pattern emerged in the three-way tables.

This is an example in which Concerned About Nutrition attitude scale is crosstabbed by the Children's/Pet Food influence factor and controlled by the presence and absence of children in the four age categories (only the significance level of the chi-square is shown). The presence of children aged 0-2 is highly significant; the direction of the relationship is that the more concerned mother is about nutrition the leas she allows her children to influence her, which makes good intuitive sense. However, it is the absence of children in the other three categories that shows significance, not their presence. In thane tables the relationships also follow the direction seen in the two-way crosstabs in which more concern led to less influence. It is difficult to explain how the absence of children would have this effect, and for this reason the complete set of controlled crosstabs is not presented for either attitudes/Children's influence or influence/family product usage in which the same pattern also tended to emerge.

TABLE 3

SUMMARY OF RELATIONSHIPS BETWEEN MOTHER'S ATTITUDE SCALES AND CHILDREN'S INFLUENCE FACTORS

TABLE

A further attempt was made to control for children's age by creating subsamples of families with children in specific age groups. However, only one group of families (those with children in the 0-2 age range) emerged as a separable subsample; all other families had children whose ages spanned two or more of the categories, even when age categories were re-defined as preschool (0-5), young (6-12) and teens (13-19).

Another controlling variable which separated families according to the age category into which the greatest number of children fell, or by the age of the oldest child(ren) when there were equal numbers of children in more than one category, was also used. This variable provided no further explanation of the variance in the data, and was discarded. Having children in other than the primary category appeared to create confounding effects.

The Relationships Between Childrens' Influence and Family Product Usage

Table 4 shows the chi-square relationships between influence factors and product usage. In this analysis, we will be comparing the extent to which mothers say their children influence purchases of each product factor with frequency of use of each of the available products within these factors.

This table shows that the relevant influence factor affected the household's usage of a substantial number of the products studied. In each cane where there was a significant effect it was positive; a higher degree of influence was accompanied by a higher rate of product usage by the family. The common denominator among all the products, with the exception of canned dog food, is that they can easily be consumed alone by one of the family members. It is reasonable, therefore, that influence should predict household demand.

A comparison of the results in this table with those in Table 3, which presented the relationship between attitudes and influence, suggests that it may be important not only to identify whether children are influential but whether their influence is directed primarily at generic product category demand or toward brand level demand.

Implications for Research and for Marketing Action

Our findings should be considered as tentative and exploratory. In this paper we have only begun to explore data available even in the data set we used. Next analytical steps should aim towards developing multidimensional classifications of attitudes, perception of children's influence and consequential purchase behavior. Two techniques that might be employed in this analysis are multivariate analysis of variance (MANOVA) and cluster analysis. Both these techniques should contribute to our understanding of the relationship. A MANOVA analysis would use the set of attitude factors as predictors and the individual products within product influence factors as criterion variables. A cluster analysis would use all three sets of variables, attitude scales, influence factors and use frequencies to develop behavioral market segments.

Future research should also concentrate on obtaining data which indicates the influence each child has on the family's decision making process. In developing and testing models of influence, attention ought to be paid not only to the age of the children, but to the spacing between children. It may well be that children grouped closely together exert more influence, due not only to greater numbers but to allied interests, than those with more years between. It may also be that differences, attitudinal as well as cognitive, in individual children affect the degree to which they attempt and are able to exert influence.

TABLE 4

SUMMARY OF RELATIONSHIPS BETWEEN CHILDREN'S INFLUENCE FACTORS AND FAMILY PRODUCT USAGE

The implications of our findings for marketing management are obvious. Evidently, children affect generic demand in some product categories as well as affecting brand choice. Therefore, it should be worthwhile to estimate the potential market size that cam result from children's demand and to direct advertising to them.

Evidently, certain children also affect brand level demand in some product categories. Our findings suggest possible marketing strategies based on these findings. For example, children seem to be more influential when their mothers have a liberal orientation, and less influential when their mothers are conservative. One might then advertise primarily to children in the liberal Northeast and advertise the same brand primarily to mothers in the Midwest. Or, one could limit advertising to mothers to conservative-oriented vehicles. In either instance the advertisements should take into account product-specific parental attitudes and emphasize how the products contribute to children's well-being.

CONCLUSION

This analysis carries the work on children as influencers and consumers one important step further by expanding the scope of previous research on the effect of parental attitudes on children's influence and by demonstrating that children's influence does affect the frequency, and presumably therefore the magnitude, of household consumption of certain products. Although the difficulties inherent in extending this stream of research are great, the conceptual and practical insight to be gained from it will certainly justify the effort.

REFERENCES

Berey, Lewis A. and Pollay, Richard W.(1968), "The Influencing Role of the Child in Family Decision Making," Journal of Marketing Research, 5, 70-72.

Caron, Andre and Ward, Scott (1957), "Gift Decisions by Kids and Parents," Journal of Advertising Research, 15, 15-20.

Davis, Harry L. (1977), "Decision Making Within the Household," in Robert Ferber ed., Selected Aspects of Consumer Behavior, Washington, D.C., U.S. Government Printing Office.

Frideres, James S. (1973), "Advertising, Buying Patterns and Children," Journal of Advertising Research, 13, 34-46.

Jenkins, Roger L. (1978), "The Influence of Children in Family Decision-Making: Parents' Perceptions," in William L. Wilkie ed., Advances in Consumer Research, Vol. VI, 413-418.

Mehrota, Sunil and Torges, Sandra (1976), "Determinants of Children's Influence on Mothers' Buying Behavior," in William D. Perreault, Jr. Ed., Advances in Consumer Research, Vol. IV, 56-60.

Moschis, George P. and Moore, Roy L. (1979), "Decision Making Among the Young: A Socialization Perspective," Journal of Consumer Research, 6, 101-112.

Nelson, James E. (1978), "Children As Information Sources in the Family Decision to Eat Out," in William L. Wilkie ed., Advances in Consumer Research, Vol. VI, 419-423.

Popper, Edward T. (1979), "Mothers Mediation of Children's Purchase Requests," in 1979 American Marketing Association Combined Proceedings, 645-648.

Robertson, Thomas S. and Feldman, Shel (1975), "Children As Consumers: The Need for Multitheoretical Perspectives," in Beverlee B. Anderson ed., Advances in Consumer Research, Vol. III, 508-512.

Rossiter, John R. (1978), "Children's Consumer Research: A Call for Rigor," in William L. Wilkie ed., Advances in Consumer Research, Vol. VI, 424-426.

Szybillo, George J., Sosanie, Arlene K. and Tenebein, Aaron (1979), "Family Member Influence in Household Decision Making," Journal of Consumer Research, 6, 312-316.

Szybillo, George J., Sosanie, Arlene K. and Tenebein, Aaron (1976), "Family Decision Making: Husband, Wife and Children," in William D. Perreault, Jr. ed., Advances in Consumer Research, Vol. IV, 469-49.

Ward, Scott (1974), "Consumer Socialization," Journal of Consumer Research, 1, 1-17.

Ward, Scott (1978), "Research on Marketing and Children: Upside or Downside on the Product Life Cycle?," in William L. Wilkie ed., Advances in Consumer Research, Vol. VI, 427-419.

Ward, Scott and Wackman, Daniel B. (1972), "Children's Purchase Influence Attempts and Parental Yielding," Journal of Marketing Research, 9, 316-319.

Ward, Scott, Popper, Edward and Wackman, Daniel (1977), Parents Under Pressure: Influences on Mothers' Responses to Children's Purchase Requests, Cambridge: Marketing Science Institute.

---------------------------------------