What Children Understand About Consumption Constellations: Differences Across Three Age Groups

ABSTRACT - This study focuses on whether children recognize consumption constellations (clusters of complementary products, specific brands and/or consumption activities used by consumers to define, communicate and enact social roles). Differences in the recognition of constellations across three age groups were considered. Consumption constellations were defined as part of the larger concept of consumption stereotypes. Pictorial occupational stimuli were used to determine if children could recognize which consumption constellation (consisting of branded goods) 'belonged’ to which owner. It was assumed that even young children socialized within a common cultural context would have, some knowledge of common consumption stereotypes. The children’s age was suggested as an explanatory factor for differences among the groups.

Citation:

Teresa Davis (2000) ,"What Children Understand About Consumption Constellations: Differences Across Three Age Groups", in NA - Advances in Consumer Research Volume 27, eds. Stephen J. Hoch and Robert J. Meyer, Provo, UT : Association for Consumer Research, Pages: 72-78.

Advances in Consumer Research Volume 27, 2000      Pages 72-78

WHAT CHILDREN UNDERSTAND ABOUT CONSUMPTION CONSTELLATIONS: DIFFERENCES ACROSS THREE AGE GROUPS

Teresa Davis, University of Wollongong

ABSTRACT -

This study focuses on whether children recognize consumption constellations (clusters of complementary products, specific brands and/or consumption activities used by consumers to define, communicate and enact social roles). Differences in the recognition of constellations across three age groups were considered. Consumption constellations were defined as part of the larger concept of consumption stereotypes. Pictorial occupational stimuli were used to determine if children could recognize which consumption constellation (consisting of branded goods) 'belonged’ to which owner. It was assumed that even young children socialized within a common cultural context would have, some knowledge of common consumption stereotypes. The children’s age was suggested as an explanatory factor for differences among the groups.

INTRODUCTION AND BACKGROUND

Most people would be able to describe a 'yuppie’ as a person with a collection of goods that relate particularly to a 1980’s lifestyle of consumer excess. Typically this description would include luxury cars, clothing and accessories creating a collective stereotype. Such a consumption stereotype really consists of two elementsBthe owner and a group of products that group together because of a clearly identifiable lifestyle.

Douglas and Isherwood (1978) considered that consumers’ social identities are defined within an identifiable consumption pattern or style.

McCracken (1988) calls this collective stereotype of owner and group of products/brands a 'diderot unity’ (Diderot (1713-1784) the French philosopher being the first documenter of such groupings). He explains the existence of such unities as necessary to define and describe cultural categories. Solomon and Buchanan (1991) name these symbolically complementary groupings associated with a specific social role as consumption constellations. Solomon and Assael (1987) taking a Gestalt view of the phenomenon, suggest that they occur because these products embody important role information. They defined them a "clusters of symbolic, complementary products, specific brands and /consumption activities used by consumers to define, communicate and enact social roles" (1987:191).

Such stereotypical descriptors are used and understood by adult consumers in a specific consumption context. How these stereotypes are recognized and used by child consumers is not fully understood. Most research in this area has concentrated on adult experiences. The notable exception to this has been the Belk, Mayer and Driscoll (1984) study. However, that study used pictures of individual products to see if children in the concrete operational and older stages made stereotypical inferences about the owners of these products. This research on the other hand, has studied the recognition of such collective consumption stereotypes by young children including 5 and 6 year olds. The collective consumption constellations were, moreover generated from an sample of adults in the same social/cultural context.

Much of the research that studies how consumers understand and use a product code has used methods adapted from social psychology and the study of social stereotypes. The association of specific products with particular social roles within a social context could lend itself to the description of a consensual stereotype ["of the coding people and behavior in terms of a few simple cognitive categoriesBto simplify what one’s to look for in people for categorization" Solomon 1988:237]. Such stereotypes, Stangor and Schaller (1996), suggest represent part of the collective knowledge of a society and is usually learned and transmitted among all its members.

It would seem then that all individuals enacting the same social role within the same cultural context should be identical in their need of product props. This cannot explain the number of individual variations of the same social role that one sees within a single society. Schenk and Holman (1980) suggest that while social role enactment is broadly culturally defined, most individuals combine these with individually defined goals or needs. Hogg and Michell (1995) also reinforce this explanation by highlighting the interdependence of the cultural context and the individual context in consumption patterns.

Belk, Mayer and Driscoll (1984) suggest that the recognition of consumption symbolism (the making of inferences about the owner based on the products s/he consumes) increases with age and experience. The approach that social psychologists use to explain individual differences in the interpretation of the collective cultural stereotypes, looks at how social stereotypes develop based on the individual’s direct contact or experience of the world.

It would seem that some part of the social role stereotype is culturally collective in definition, while there is another part of the stereotype that is individually determined. The ommon part of the stereotype is learned indirectly or non-experientially since it is part of the society’s collective knowledge. People in a common cultural context would, therefore, learn which specific consumption constellations go with a particular social role by the process of socialization. Detailed social role knowledge however, has to be learned experientially through belonging to, or being in constant contact with members of that particular social role.

If social role knowledge can be defined as a kind of stereotype, i. "a cognitive structure containing the perceiver’s knowledge, beliefs, and expectancies about some human social group" (Hamilton and Trolier 1986), then knowledge about these roles can be explained in terms of both cultural and individual components.

It would be consistent with the notion of a common cultural consumption stereotype that even young children could vicariously learn to recognize collective stereotypes. Hirschfield (1995) as well as Gelman, Collman and Maccoby (1986) have shown that in developmental research even pre-operational children acquire and use stereotypes (i.e labels and descriptions) before they really understand how to apply them. There has been successful research done with children in this age group in categorization studies with regard to racial and other types of social stereotyping ( Markman 1989).

If social role knowledge exists at the stereotypical level such as described by specific consumption constellations, then even children would have been socialized to recognize them. Since more detailed social role descriptions will be learned through experiential/direct sources over time, only the cultural level consensual stereotypical knowledge can be expected among young children. Older children will be able to recognize stereotypical consumption constellations for a social role better than younger children (having undergone more socialization). However, having more direct experience with members of the social role, will be able to describe alternative non-stereotypical consumption constellations.

Studies that have successfully used child samples have used visual stimuli. Perrachio (1990) lends support to this argument that pre-operational children can be successfully used in cognitive recognition and classification tasks by using non-verbal stimuli. She also suggests the use of tasks that the children are familiar and comfortable with. This research therefore, used primarily a visual, matching task in a familiar classroom environment.

TABLE 1

LEARNED COMPONENTS OF SOCIAL ROLE DEFINITION

MAIN RESEARCH QUESTIONS

A matching task was followed by an interview to complement the information gained from the grouping task. The independent variables of age and social role knowledge were considered. The dependent variables consisted of accuracy in identifying stereotypical consumption constellations (SCCS) and the description of alternative/non-stereotypical consumption constellations (ASCCS).

The main research questions addressed were related to age. Specifically stated these were:

1] Are social role stereotypes (as described by consumption constellations) recognised by children below the age of seven ( pre-operational stage)?

2] Does the accuracy with which such social role stereotypes (as described by consumption constellations) are recognised increase with age ?

3] Do older children have a better understanding of alternative/non-stereotypical consumption constellations than younger children

HYPOTHESES

While it would seem that pre-operational children (below the age of seven) are not cognitively developed enough to deal with the abstract concepts such as social stereotypes, some studies indicate that such children do have knowledge of such labels and categories (Hirschfield 1995) even if they cannot actually identify members of that category. They should have a stereotypical and culturally learned understanding of social roles. While they may not be able to identify or describe alternative/non-stereotypical consumption constellations associated with that role because of a lack of direct contact with members of that group, they will have knowledge learned from indirect sources.

The term consumption constellation is used in this research to mea "a cluster of complementary products, specific brands and or consumption activities associated with a social role" (Solomon and Assael 1987). This research adopts the view expressed by McCall and Simmons (1982) that a collection of products and activities are taken by society as defining social roles. For the purposes of this research, the Biddle and Thomas definition of a social role is to be used, which describes it a "a behavioral repertoire characteristic of a person or position a set of standards, descriptions, norms or concepts held (by anyone) for the behaviors of a person or a position" (1966:11). In this research the term stereotype is used in the same sense as Solomon does:

"the coding of people and behavior in terms of a few simple cognitive categories (to simplify) what one is to look for in people for categorization (social)"(1988:237).

H1  For all social roles, pre-operational children (below seven years of age) will identify accurately the stereotypical consumption constellation associated with a particular occupational role.

H1a  For all social roles, pre-operational children (below the age of seven) will provide no alternative, non-stereotypical consumption constellations.

With increasing age, the older child, having undergone more of the socialisation process, will have a better understanding of social/occupational roles and the stereotypical consumption constellations associated with them. This would make it easier for them to accurately identify these stereotypical consumption constellations as well as describe Alternative Non-stereotypical Consumption Constellations.

H2 For all social roles, the early concrete operational (eight and nine years of age) group of children will identify, more often than the pre-operational (below seven years of age) group, the accurate stereotypical consumption constellation.

H2a  For all social roles, the early concrete operational (eight and nine years of age) group of children will be able to provide alternative, non-stereotypical consumption constellations.

As the child grows and is socialised further, their experiential knowledge of the more atypical and non-stereotypical examples of occupational group members. Thus the oldest group of children considered (11-12 years) should be able to provide the most examples of non-stereotypical consumption constellations associated with the occupational roles than the younger children.

H3  For all socal roles, the late concrete operational (eleven and twelve years of age) group of children will identify, more often than the early concrete Boperational ( eight and nine years of age) and the pre-operational (below the age of seven), the accurate stereotypical consumption constellation.

H3a  For all social roles, the late concrete operational (between the ages of eleven and twelve) group of children will be able to provide more alternative non-stereotypical consumption constellations than the early concrete operational (ages of eight and nine years) group of children.

In cases where the child is constantly in the company of the members of a particular social role (e.g. parents are members of that occupation, or they know a lot of people in that occupation), social role knowledge is enhanced and a more detailed knowledge of that group results (Solomon1988). These are the main hypotheses to be considered in this research.

METHOD

The visual matching task was followed by a question and answer session with each subject. The sequence of research activities is presented in a below.

Sequence of Research Tasks

Part A Prerequisite Procedures

1] Sample Selection,

2] Product/brand Selection and

3] Materials and Stimuli Selection.

Part B Data Collection Task

1] Cognitive Developmental Stage Determination.,

2] Social Role Knowledge Scale Administration,

3] Visual Matching Task,

4] Follow-up interview.

Part A Prerequisite Procedures

1] Sample Selection: A quota sample was used drawing from children in three age-groups. Following the cognitive developmental stages defined by Piaget (1960), 66 children from the pre-operational (5-6 year olds), 48 children from the early concrete operational (8-9 year olds) and 61 children from the late concrete operational (11-12 year olds) were chosen to form the sample. This was closest to the 11:8:15 ratio of these age groups found in the school-going population of the state of New South Wales, Australia (Australian Bureau of Statistics: Schools and Schooling in NSW handbook 1999).This research did not explore differences across culture and, therefore, the sample was culturally homogenous. The sample was drawn from 5 different schools in different socio-economic suburbs ensuring a fair representation from the upper, middle and lower middle classes. Gender distribution was spread as equally as possible to minimise gender effects (97 female and 78 male subjects, comprising 55.4% female and 44.6% male). Thus, the total sample consisted of 175 students distributed across the three age groups.

2] Product/Brand Selection: A group list generation task was carried out to ensure that the 12 product brands could indeed be evoked by the occupational stimulus. A sample of 19 students, professionals and clerical people was used and consensus on the products listed was used to decide whether they were included in the visual sort task in the actual experiment. They all described themselves of Australians of Anglo-Celtic origin, this being the dominant cultural group in the Australian population (1991 Census of Population and Housing, ABS). The group were first to try and form a mental picture of a 'typical’ member of each of the three income classes and to describe in as much detai as possible, what occupation each belonged to. These classes were described occupationally as the Professionals, the Intermediates and the Working cluster based on the Australian Standard Classification of Occupations (Baxter, Emminson and Western 1991). Each respondent was asked to list only one occupation in each income class. Each respondent was further instructed to pick an occupation that in their opinion 'typifies’ that income class. The averages were calculated across the respondents and the three occupations were identified as a dentist, a school teacher and a garbage collector. Then each person was asked to describe in the greatest detail possible what kind of house each of these people lived in, what kind of car they owned, which local restaurant they frequented and where they would travel on a holiday.

The product categories used were value-expressive ones. This was done because, as Munson and Spivey showed in their 1981 study, product-brand user stereotypes hold only for such visible and classifiable products as cars, clothing, etc. Using 3 brands for each of 4 product categories (a set of 4 brands to correspond to each of the 3 occupational social roles), a total of 12 brands were used across the 4 product categories.

The responses were averaged (number of times any single brand was mentioned), and a final set of brands was arrived at. See table 2.

3] Materials and Stimuli: The actual stimuli consisted of pictures of 3 persons representing the 3 occupations (the dentist, the school teacher and the garbage collector) and separate pictures of the 12 product brands. The children used photographs or pictures which they matched; creating 3 sets; owner to 4 product brands.

Part B Evaluation Tasks

1] Cognitive Developmental Stage Determination: To ensure that the children of each of the three age groups did indeed belong to the cognitive developmental stage indicated by their age, two tests were carried out.

The tests used are part of the standard set of tests that measure cognitive developmental stage. The first test used was a 'seriation’ of length task. The second test was a 'conservation’ test (Inhelder and Piaget 1974).

The 2 tasks were scored 0, 1, or 2. Those scoring 2 (correctly on both tests) were categorised as early concrete operational. Those scoring 0 or 1 (seriation task correctly completed, conservation task incorrect) were categorised as pre-operational. Two 8-9 year olds could not complete the seriation task and their responses were excluded. None of the 5-6 group could correctly 'conserve’. It therefore seemed that age was an adequate proxy for cognitive development.

2] Social Role Knowledge Test: This consisted of the administration of the social-role knowledge scale that was based on Solomon’s 1988 scale. This was modified to suit the child sample. The final scale was a six-item scale relating to familiarity with members of a particular social role. The Social Role Knowledge score (SRKS) was calculated for each of the three occupations from this. The SRKS could range from 0B15 for each social/occupational role (scale and scoring method available on request from the author).

3] Visual Matching Task Administration: This task was based on the work of Jahoda (1959), consisted of a visual matching task where the subjects used pictures to group together a person of a particular occupation and the specific product brands to form owner based consumption constellations. This task was scored using a 4 point score for each occupational role. If the subject matched all four brands 'correctly’ to the owner they were awarded 4 points on their SCCS (Stereotypical Consumption Constellation Score). This was done for each of the three occupational roles used.

4] Follow-up interviews: These interviews were used to complement the information from the group-sort task. During this phase, subjects were asked the following questio "Can you tell me of another (other than the ones matched by the child in the firs task) house, car, restaurant or holiday that the dentist would own " After the child has responded, the same question was repeated for the next occupation and when that response was completed, for the third. Each group of owner with a set of four products (other than the 'correct’ stereotypical products) was scored as one point toward the ASCCS (Alternative/non-stereotypical Consumption Constellation Score). The children could use the pictures used in the earlier task, or describe/name the product.

TABLE 2

STEREOTYPICAL CONSUMPTION CONSTELLATIONS: AVERAGE PERCENTAGES ACROSS OCCUPATIONS AND PRODUCT BRAND

ANALYSES, RESULTS AND DISCUSSION

To find out if the main research questions were supported, three analytical steps were carried out. First stepwise regression analyses were carried out to see if age was an important explanatory variable for the dependent variables Stereotypical Consumption Constellation Scores (SCCS) and Alternative Non-Stereotypical Consumption Constellations (ASCCS). If age proved to be a significant independent variable, one-way ANOVAs were run to see if the differences between the age groups were significant. If significant, Post hoc tests were run to identify exactly where the differences lay. The results of the Step-wise partial regressions are shown in Table 3.

The following explanation applies to all abbreviations in tables and text:

SCCSBStereotypical Consumption Constellation Score (These could specifically relate to a particular occupation i.e. SCCS relating to the Dentist, Teacher and Garbage Collector is denoted by SCCSD, SCCST and SCCSGC.

ASCCSBAlternative/non Stereotypical Consumption Constellation Score ( These could specifically relate to a particular occupation i.e. ASCCS relating to the Dentist, Teacher and Garbage Collector is denoted by ASCCSD, ASCCST and ASCCSGC.

The stepwise regressions showed that age was an independent variable with fairly strong explanatory power in terms of the Stereotypical Consumption Constellation Scores (SCCS) for two of the occupational roles (dentist and garbage collector). It was low for the role of teacher (SCCST). For the Alternative Non-Stereotypical Consumption Constellation Scores (ASCCS), the age variable was considerably stronger, with the partial R ² for the occupational role of teacher being the strongest at.5153. For both dependent variables (SCCS and ASCCS), ANOVAS were run (excluding SCCST) to see if the differences across the age groups were significant. The results are shown in table 4.

Table 4 shows that there were significant differences between the age groups on the dependent variable of Stereotypical Consumption Constellation Scores (SCCS) for both the occupational role of the dentist and the garbage collector. The differences on the SCCST were not significant across age groups. For the dependent variable of Alternative Non- Stereotypical Consumption Constellation (ASCCS), that the differences between the age groups were all significant at the.0001 level. Therefore Post hoc tests were carried out on all the significant results (excluding the SCCST). The results of the Tukey’s Studentized Range (HSD) are shown in table 5 for the dependent variable SCCS and in table 6 for the dependent variable ASCCS. [Note: In the case of the SCCS for all occupations, the minimum score for correctly identifying the SCCS was calculated to be 2.02 when probability was factored in.]

TABLE 3

SUMMARY RESULTS OF THE STEPWISE PARTIAL REGRESSIONS

TABLE 4

SUMMARY RESULTS OF ANOVAS

The results of the Post hoc tests show fairly clear groupings in the case of the dependent variable SCCS. The means of the age groups are quite sequential for the occupational role of dentist and garbage collector. The 11 and 12 year olds (late concrete operational) showing the highest means on their scores with the 9 and 8 year olds (early concrete operational) showing means that are higher than the 5 and 6 year olds (pre-operational) but lower than that of the late concrete operational. Tukey’s groupings on these scores are interesting and quite consistent. The oldest and youngest age groups are clearly and significantlydistinct. The 8-9 year olds are inconsistent in their scores and seem to group with the older group for the role of Garbage collector, but group with the younger group on the role of dentist. The post hoc test was not carried out on the SCCST (teacher) role as the ANOVA did not show significant differences between the age groups for this role.

TABLE 5

RESULTS OF TUKEY'S (STUDENTIZED RANGE) HSD TESTS FOR SCCS

TABLE 6

RESULTS OF TUKEY'S STUDENTIZED RANGE (HSD) POST HOC TESTS FOR ASCCS

The results of the Post hoc tests for the dependent variable ASCCS showed a very clear and direct relationship with age. With the 11 and 12 year olds (late concrete operational) showing the highest mean scores for each occupational role (Dentist, teacher and Garbage collector). The groupings fell into three clear age related groups. In this it differs from the SCCS, where the groupings did not as clearly follow the Piagetian stages. The mean scores were sequential with age, the oldest group scoring the highest mean scores and the youngest group scoring the least. The results therefore show clear support for H1a, H2, H2a, H3 and H3a. H1a was not supported for 2 out of the 3 occupations. While the means of the youngest age group on the SCCSGC were higher than 2.02 (calculated as the random probability of matching correctly the stereotypical consumption constellation to the owner, in the other case (SCCSD) it was lower and therefore H1 was not supported.

GENERAL DISCUSSION

The results shows clearly that overall age appears to be a good predictor of how well children understand and use stereotypical consumption constellations in making inferences about product ownership amongst different occupational groups. There are some variations both in the case of the use of SCCS and ASCCS depending on the occupations themselves. For instance the occupation of teacher is clearly one that all ages of children were familiar with. The SCCS for this role did not clearly increase with age and was the only role where age was not a significant variable. The ASCCS for this role however showed clear support for the Piagetian age-wise grouping, indicating that while familiarity with an occupation tended to reduce the SCCS, It did not increase the ability of younger children to describe Alternative non-Stereotypical Consumption Constellations associated with that role. This might suggest that age may be a stronger predictor variable for ASCCS than occupational role familiarity.

It would be of interest to see if the age related results for the recognition and use of consumption constellations would hold true for other cultural contexts. The recognition and use of consumption constellations among other groups such as new migrants or new members of a culture would be avenues for possible future research. Such groups would be undergoing a process of consumer acculturation not different from young consumers and it would be of benefit to marketers and consumer behaviorists to see if there are differences in the way such consumption stereotypes are recognized and used.

REFERENCES

Alderson, Wroe (1957), Marketing Behavior and Exectutive Action: A Functionalist Approach to Marketing Theory. Homewood, IL:Richard.D.Irwin.

Belk, R; Mayer, R and Driscoll, A (1984), "Children’s recognition of consumption symbolism in children’s products." Journal of Consumer Research, vol 10 March, 386 -397.

Biddle, B.J & Thomas, E.J(1966), (eds) Role Theory: Concepts and Research. John Wiley & Sons, Inc, New York.

Bowen, R (1956), "Introduction" in Rameau’s nephew and other works by Denis Diderot Translated by Jacques Bazun and Ralph Bowen,NY Bobbs-Merill.

Bowen, R and Mischel, W (1979),"Prototypes in person perception." In L.Berkowitz Advances in Experimental Social Psychology, vol 12,4-52 NY Academic Press.

Douglas, M and Isherwood, B (1978), The World of Goods: Towards an anthropology of consumption. New York; W.W Norton.

Gelman, G; Collman, P & Maccoby, E.E.(1986),"Inferring properties from categories vs Inferring categories from properties: The case of gender" Child Development 57, 396-404.

Hamilton, D.L& Trolier, T.K(1986),"Stereotypes and stereotyping: An overview of the cognitive approach" in J.F Davidio & S.L. Gaetrner (eds) Prejudice discrimination and racism, 127-163. Orlando FL.Academic Press.

Hirschfield, L.A (1995), "Do children have a theory of race" in Cognition, 54, 209-252.

Jahoda, Gustav (1959), "Development of the perception of social differences in children from 6 to 10"The British Journal of Psychology 50, 59-175.

McCall, G.J and Simmons.J.C (1982), Social Psychology: A Sociological Approach The Free Press, New York.

McCracken, Grant (1988), Culture and consumption Indiana University Press: Bloomington IN.

Markman, E (1989), "Categorization and naming in children: Problems of Induction" MIT Press Cambridge, Mass.

Munson, M.J and Spivey, A.W (1981), "Product and brand user stereotypes across social class" Advances in Consumer Research, vol 8,696-699.

Piaget, J (1960), "General problems of the Psychological development of the child" in Discussions on child development: Proceedings of the World Health Organization study the group on Psychological development of the child; IV, eds J.M.Tanner & B.Elders, NY International Universties Press.

Perrachio, L (1990), "Designing Research to Reveal the Young Child’s Emerging Competence". Psychology and Marketing, 7 (Winter), 257-276.

Solomon, M.R (1988), "A social categorization approach to consumption symbolism" Journal of Psychology and Marketing, vol 5(3) 233-258.

Solomon, M.R and Assael, H (1987), "The Forest or the TreesBA Gestalt approach to symbolic consumption" in Marketing and Semiotics- New Directions in the study of signs for sale (ed) Umiker-Sebeok, Mouton-Gruyter, 189-218.

Solomon, M.R and Buchanan, B(1991), "A Role Theoretic approach to Product SymbolismBMapping a consumption constellation" Journal of Business Research, 22, 95-109.

(A complete set of references may be obtained from the author).

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