Advances in Consumer Research Volume 12, 1985 Pages 525-531
RACIAL AND SOCIOECONOMIC INFLUENCES ON THE DEVELOPMENT OF CONSUMER BEHAVIOR
George P. Moschis, Georgia State University
Roy L. Moore, Georgia State University
While most marketers and consumer behavior researchers are in general agreement that the consumer behavior of blacks differs from that of whites, there appears to be no consensus on the causes of the differences Information on cause of black-white differences in consumer behavior is lacking, and the few studies conducted in this area have produced contradictory results. The study results suggest that black-white differences in consumer behavior may be attributable to socialization processes that operate differently among the two subcultures.
Social scientists have traditionally been interested in understanding black-white differences in cognitions and behaviors, such as intelligence and competency, and whether such differences are the result of genetic or socioeconomic characteristics. Similarly, while most social scientists , marketers and consumer behavior researchers are in general agreement that the behavior of blacks in the marketplace differs from that of whites, there appears to be no consensus on the causes of the differences. The controversy seems to center around two issues: 1) whether differences in consumer behavior are the result of racial or socioeconomic factors; 2) how such differences develop. Generally, information on cause of black-white differences in consumer behavior is lacking and the few studies conducted in this area have produced contradictory results.
Previous studies focusing on the first issue suggest that black-white differences in consumer behavior may be the result of racial factors, socioeconomic factors, and an interaction of both (Feldman and Star 1968, Cicarelli 1974, Donahue et. al. 1978, Moschis and Moore 1981). However, these studies do not indicate how and why such differences may occur. Recent research suggests black-white differences may not be necessarily due to racial characteristics per se, but due to influences ( i.e., third variables) present in different socioeconomic and racial subcultures of society. Such influences say affect the learning of consumer behaviors. For example, black and lower-class youths are likely to respond to TV commercials and use the mass media differently than their white and upper-class counterparts (e.g., Christiansen, 1979) and such differences may result in the development of different consumer patterns.
The present study extends research reported by Moschis and Moore (1981) which found differences in the consumer skills, knowledge and attitudes between blacks and whites, such as knowledge of consumer affairs, ability to filter puffery, management of consumer finances, materialism and brand preferences, beyond those differences that can be accounted for by the individual's socioeconomic status. The focus is on whether these differences are due to antecedent variables associated with different socioeconomic and racial factors, or due to other intervening factors associated with different socioeconomic or subcultural characteristics.
Explanations for such possible differences are sought in socialization theories. The socialization perspective appears to be the most viable for addressing these particular research questions for at least three reasons: first, a great deal of consumer behavior is believed to be learned, especially during pre-adult years (Olshavsky and Granbois, 1979); second, learning of one's behavior and culture according to the culture in which he is raised (e.g., black vs. white) is a major subject of socialization theory and research (Engel, Blackwell and Kollat, 1978); finally, research findings on black-white differences in consumer behavior presented in this paper suggest that the socialization model is an appropriate framework for understanding these black-white difference i in consumer behavior.
BACKGROUND AND HYPOTHESES
A conceptual model of consumer socialization developed in previous research (Churchill and Moschis 1979; Moschis and Churchill 1978; Moschis and Moore 1978, 1979) incorporates five different types of variables: learning properties, age or life cycle, social structural variables, socialization agents, and learning processes (Moschis and Churchill 1978). The five are classified as "antecedent variables," "socialization processes," or "outcomes."
Antecedent variables include social structural and developmental variables that locate the individual in his or her social environment. Examples of social structural variables are social class, race, sex, and education; developmental variables include either age or life cycle. Socialization processes refer to agent-learner relationships, which incorporate the specific agent and learning process. Socialization agents include mass media, parents, peers, and school, while learning processes include modeling (imitation of learner's behavior), reinforcement (positive or negative), and social interaction (which may include both modeling and reinforcement). Outcomes in the model include the development of consumer knowledge , attitudes, and norms. Developmental and social structural variables may affect cognitions and behaviors directly as well as indirectly by impacting upon socialization processes.
With respect to the influence of antecedent variables, previous research suggest that race and socioeconomic status may directly affect the learning of some consumer behavior. Specifically, research by Moschis and Moore (1981) investigated the effects of race and socioeconomic status on the adolescent's ability to filter puffery in advertising, ability to manage consumer finances, consumer knowledge, brand preferences, materialistic attitudes, attitudes toward the marketplace, and consumer discontent. It was found that white adolescents were better able to filter puffery, but contrary to the researchers' expectations, they were less likely to have preferences for brands. Middle class adolescents, on the other hand, were better able to filter puffery in advertising, had greater knowledge about consumer matters, had stronger brand preferences, had weaker materialistic attitudes and more negative attitudes toward the marketplace than their lower class counterparts. Thus, while the effects of race, independent of those of socioeconomic status, were not particularly strong, the effects of socioeconomic status on consumer behavior were more significant and can be interpreted in the context learning and cognitive theories. For example, it can be argued from a learning theory perspective that low socioeconomic status adolescents have less experience with money and thus may be less aware of the range of consumer goods available in the marketplace than those from upper socioeconomic backgrounds who have more opportunities for and experiences with consumption (Riesman, Glazer and Denny, 1950; Ward, 1974). Additional research also appears to support this line of reasoning showing that upper socioeconomic class youths compared to those of lower socioeconomic background have greater knowledge of economic concepts (Williams, 1970) and brand preferences (Guest, 1942). It has also been argued that the cognitive field to the lower class youngster is believed to be unstructured, increasing the possibility of eventual disappointment and frustration (Hess, 1970). The latter notion may explain the inverse relationship found between social class and general attitudes toward the marketplace and is consistent with findings reported by Ward and Reale (1972).
The main research hypothesis in this study is that race and socioeconomic status are also likely to affect the acquisition of consumer behaviors indirectly by impacting upon socialization processes. Previous theory and research suggest that both the importance of socialization agents and learning process are likely to vary by race and socioeconomic status. For example, a commonly held belief among social scientists is that black children are sheltered by the family for a shorter period and exposed to peer interaction earlier than most white children (Hartup, 1970, pp. 409-410). Similarly, the black child may soon perceive his peers as having less status and power and he may feel that their judgments may be worth less simply because they are blacks (Hartup, 1970, p. 410), which may explain why blacks attempt to model after whites. Gerson (1966), for example, found that black adolescents are using the mass media to learn how to behave like whites. Others have also found that not only race, but also socioeconomic status may be differentially related to observational learning from television (e.g., Christiansen, 1979), with lower class and minority youths being most likely to be affected by television.
The preceding discussion suggests the following hypotheses:
H1: White adolescents compared to black adolescents are more likely to (a) discuss consumption with their parents, (b) observe parental consumer behavior; they are less likely to (c) discuss consumption with peers, (d) observe peer consumer behavior (e) be heavy television viewers, and (i ) to have motivations for television viewing.
H2: Middle class adolescents compared to lower class adolescents are more likely to (a) discuss consumption with their parents, ( b) observe parental consumer behavior; they are less to, (c) discuss consumption with peers, (d) observe peer consumer behavior, ( e) be heavy television viewers, and (f) to have motivations for television viewing.
H3: Consumer learning processes have different impact on the development of consumer orientations among (a) whites vs. blacks and (b) middle class vs. lower class adolescents.
Thus, race and social class are expected X indirectly influence the acquisition of consumer orientations ( attitudes toward products, brands, the marketplace and materialism) by affecting the socialization processes, and such processes are expected to directly affect consumer learning.
Since this study addresses different questions from those addressed in the Moschis and Moore (1981) study, the same data base was used. The sample of this study consisted of 784 adolescents from several cities and towns in six counties in urban, suburban, semirural, and rural Georgia. In each county, self-administered questionnaires were completed by respondents in one middle and one senior high school as part of a longitudinal study. The selection of schools was based on personal interviews with school officials, and attempts were made to select schools representative of their respective regions. The sample was very representative with respect to race (approximately 12 percent blacks), age (53% middle schoolers and 47% high schoolers), sex (49% males and 51% females) and socioeconomic status using Duncan's (1961) scale (mean = 45.15). [Using Duncan's two-digit occupational percentile code, the lower social class category consisted of adolescents whose father's occupation fell in the lowest 50% of the percentile scale (10% for the lower lower and 40% for the upper lower -- see Warner 1972), whereas the remaining respondents, which fell within the next 45% of the occupational characteristics, were treated as middle class. Since none of the respondents fell in the upper 3% of the code, no upper class category was constructed. Fifty-five percent of the respondents in the sample fell into the lower class category.] Duncan's scale has been used in previous studies of consumer socialization (e.g. Moschis and Churchill, 1978; Moschis and Moore, 1979).
The operationalization and reliability coefficients of the independent variables used in this research are shown in Table 1. These were constructed by summing responses to items and using item-to-total correlations to purify the interest and by using coefficient alpha to assess the resultant reliability. The criterion variables examined were limited to those which were found to be significantly related to race ant/or socioeconomic status in the Moschis and Moore (1981) study. Factor analysis revealed three factors relating to motivations for viewing of TV advertisements and programs: viewing TV ads for consumer decision making and social purposes, and viewing programs for social reasons. These motives were in line with previous theory and research (e.g., Moschis and Churchill, 1978; Lull, 1980), providing validity checks for these newly constructed variables.
Of main interest in this research was the examination of differences of teenagers' socialization processes as a function of the individuaL's socioeconomic and race characteristics. To investigate the independent effects of each of the two factors, a two-way analysis of variance was performed. The four possible treatment groups were defined by SES(2) x RACE(2). Table 2 shows F-ratios highlighting the significant effects produced by ANOVA.
In examining the effects of race and socioeconomic status on socialization processes, both variables produced main effects but no interactive effect on these processes (TabLe 2). Specifically, white adolescents do not appear to discuss consumption with their parents significantly more than black adolescents do, a finding that offers no support for hypothesis H1 (a). However, white adolescents tent to observe parental consumer behavior more frequently than their black counterparts, as posited [Hypothesis H1(b)] (F = 21.75, p < .001). White adolescents do not appear to be less likely than black adolescents to discuss consumption and observe peer consumer behavior, providing little support for Hypotheses H1(c) and H1(d), respectively. However, black teenagers are more likely than their white counterparts to spend more time watching television (F = 3.96, p < .05), as posited [Hypothesis H1(e) ]; and they are more likely to watch television ads and programs too for social reasons (F = 5.99, p < .01 and F = 5.05, p < 2, respectively, providing partial support for Hypothesis 1(f).
With respect to the effects of social class or socialization processes, there were no differences on the measures of overt and cognitive social interactions with significant others (parents and peers), providing no support for Hypotheses 2(a), 2(b), 2(c), and 2(d). Socioeconomic status, however, was found to be related to television viewing. Specifically, lower class adolescents compared with their middle class counterparts tend to spend more time watching television, and they are more likely to have motivations for watching television advertisements, such as to make Consumer decisions (F = 3.97, p < .04) and for social reasons (F = 6.19, p < .01) as well as to watch television programs for social reasons (F = 3.95, p < .04). These data provide support for Hypotheses 2(e) and 2(f).
To determine the extent to which such racial and socioeconomic differences on socialization processes affect consumer learning, separate analyses were performed on black and white as well as lower and middle-class adolescents. The results of these analyses are shown in Tables 3 and 4.
Table 3 shows relationships between significant socialization processes and consumption-related variables on which significant direct differences emerged among blacks and whites (see Moschis and Moore 1981). The table indicates that different learning processes may operate among black and white adolescents in the acquisition of various consumer skills and that the effectiveness of such processes may vary by consumer learning property. The regression coefficients were significantly different among the two groups as indicated by Chow's (1960) F-test of equality. [Chow's (1960) F-test can be used to test whether the same relationships (in a multiple regression model) hold for two different groups. A significant F-value suggests that the regression coefficients in the two groups differ, with the independent variables having different effects on the dependent variable.] For example, by observing their parents, black youths seem to develop preferences for brands, while whites seem to develop materialistic values watching TV ads for social reasons. These findings provide support for Hypothesis 3(a).
Although SES contributed to the understanding of differences in learning processes independently of race, there were fewer differences in the effectiveness of the learning processes among lower and middle class subjects as suggested by Chow's F-test (Table 4), providing inadequate support for Hypothesis 3(b). In addition, socioeconomic status appears to be a more important factor in the development of consumer behavior of whites rather than the consumer behavior of blacks, while race is a much more important variable in the consumer socialization of lower class rather than middle class youths (Table 4). Specifically, socioeconomic status is positively associated with consumer learning only among white adolescents, whereas in lower social classes relatively more white than black teenagers are likely to develop consumer behaviors in comparison with teenagers in middle social classes.
This study addressed the question of whether differences in consumer behavior between whites and blacks found in earlier studies can be attributed to socialization processes operating differently in these subcultures. Race and SES produced independent effects on consumer socialization processes, which differentially impacted the development of consumer cognitions. The influence of socialization processes, further, seems to vary by race and to a lesser extent by socioeconomic status. simultaneously with the impact of socialization processes using multivariate analysis, This analysis revealed that the effect of SES tends to disappear while racial effects remain strong. The impact of socialization processes on criterion variables is significant on both blacks and whites, as well as on lower and middle social classes. However, these influences are more likely to be different among blacks and whites than among lower and middle class adolescents. Thus, the development of consumer orientations-among blacks and whites occurs not only direct, but also indirect via socialization processes that operate differently among the two subcultures. Socioeconomic influences tend to be indirect via socialization processes, which appear to have a rather similar impact among the two social groups.
In the context of the general consumer socialization model, the antecedent variables race and social class examined in this study appear to differentially affect, both directly and indirectly, the development of consumer behavior. While adolescent's interaction with the television and his parents varied by his race and SES characteristics, the findings suggest that different learning processes may operate in the acquisition of various consumer behaviors among these groups. In sum, black-white differences in consumer behavior appear to be affected socialization processes operating differently among the two subcultures.
The findings of the present study should be viewed in the light of several limitations. First, one must keep in mind that the study dealt only with teenagers from specific counties in Georgia. Perhaps the consumer behavior of teenagers in other parts of the country may differ with respect to the effects of SES and race. Second, the dependent variables obviously do not represent all aspects of consumer behavior. Thirdly, because of the possible intercorrelations among the socialization processes, MANOVA (rather than ANOVA) could have produced more accurate findings. Finally, differences in consumer behavior between blacks and whites--beyond those accounted for by the socialization processes - may not necessarily be attributed to race per se, but they also may be due to third variables, perhaps other socialization processes not examined in this study. Future studies should address these questions using structural equations, using a more complete test with the explicit consideration of measurement error simultaneouslY.
Although the results of this study are limited, differences between black and white teenagers in the acquisition of consumer skills, knowledge and attitudes appear to be linked primarily to different socialization processes among the subcultural groups examined. Such socialization processes are unlikely to change until a strong effort is made by educators and others such as the family to teach both groups socially desirable consumer attitudes and behaviors. Consumer socialization both for blacks and whites appears to be a rather complex process that does not lent itself to easy solutions. Individualized instruction designed to fill in existing gaps in consumer knowledge and skills may be more appropriate than a massive education effort since levels of learning may vary considerably from one subgroup to another.
Also, the role of socioeconomic status should not be downplayed since higher SES usually provides greater opportunities for interaction with the marketplace and thus opportunities for appropriate learning. Since adolescence is apparently a particularly crucial period for consumer socialization, efforts at consumer education will probably be far more successful if begun at this stage or among younger children than if simply waiting until adulthood when it may be too late.
EFFECTS OF RACE AND SOCIOECONOMIC STATUS ON CONSUMER SOCIALIZATION PROCESSES
EFFECTS OF SOCIALIZATION PROCESSES AND SES ON CONSUMER ORIENTATIONS AMONG BLACK (B) AND WHITE (W) ADOLESCENTS
EFFECTS OF SOCIALIZATION PROCESSES AND RACE ON CONSUMER ORIENTATIONS AMONG LOWER-CLASS (L) AND MIDDLE-CLASS (M) ADOLESCENTS
Chow, Gregory C. ( 1960), "Tests of Equality Between Sets of Coefficients in Two Linear Regressions," Econometrica, Vol. 28 (July), 591-605.
Christiansen, John B. ( 1979), "Television Role Models and Adolescent Occupational Goals ," Human Communication Research, Vol. 5 (Summer), 335-337.
Churchill , Gilbert A., Jr. and George P . Moschis ( 1979), "Television and International Influences on Adolescent Consumer Learning ," Journal of Consumer Research, Vol. 6 (June), 23-35.
Cicarelli, James ( 1974 ), W On Income , Race, and Consumer Behavior," American Journal of Economics and Sociology, (July), 243-248.
Donahue, Thomas R., Timothy P. Meyer and Lucy L. Nenke (1978), "Black and White Children: perceptions of TV Commercials," Journal of Marketing, Vol. 42 (October), 34-40.
Duncan, Otis D. (1961), "A Socioeconomic Index of All Occupations," in Albert J . Reiss, Jr . ( ed . ), Occupations and Social Status (New York: The Free Press.
Engel, James, Roger Blackwell, and David Kollat (1978). Consumer Behavior, Third Edition. New York: Holt, Rinehart and Winston, Inc.
Feldman, Laurence P. and Alvin D. Star (1968), "Racial Factors in Shopping Behavior ," in Keith Cox and Ben Enis ( eds.), A New Measure of Responsibility for Marketing (Chicago: American Marketing Association), June, 216-226.
Gerson, W. M. (1966), "Mass Media Socialization Behavior: Negro-White Differences ," Social Forces, Vol. 45, 40-50.
Guest, Lester P. ( 1942), "The Genesis of Brand Awareness," Journal of Applied Psychology, Vol. 26 (December). 800-808.
Hartup, Willard (1970), "Peer Interaction and Social Organization," in Paul H. Mussen (ed.), Manual of Child Psychology, Vol. 2, Third Edition (New York: John Wiley & Sons, Inc.), Chapter 24, 361-456.
Hess, Robert D. (1970)," Social Class and Ethnic Influences Upon Socialization ," in Paul Mussen (ed.), Manual of Child Psychology, Vol. 2, Third Edition (New York: John Wiley & Sons, Inc.), 457-459.
Lull, James (1980), "The Social Uses of Television," Human Communication Research, Vol. 6 (Spring), 197-209.
Moschis, George P. and Gilbert A. Churchill, Jr. ( 1978), "Consumer Socialization: A Theoretical and Empirical Analysis ," Journal of Marketing Research, Vol. 15 (November), 599-609.
Moschis, George P. and Roy L. Moore (1979) "Decision-Making Among the Young: A Socialization Perspective," JournAl of Consumer Research, Vol. 6, ( September), 101-112.
Moschis, George P. and Roy L. Moore (1981), "Racial and Socioeconomic Influences on Adolescent Consumer Behavior", Proceedings of the American Marketing Association. Chicago, Illinois.
Olshavsky, Richard W. and Donald H. Granbois (1979), "Consumer Decision-Making -- Fact or Fiction?" Journal of Consumer Research, Vol. 6 (September), 93-100.
Riesman, David and Howard Roseborough (1955), "Careers and Consumer Behavior," in L. Clark (ed.), The Life Cycle and Consumer Behavior (New York: New York University Press).
Ward, Scott (1974), "Consumer Socialization," Journal of Consumer Research, Vol. 1 (September), 1-16.
Ward, Scott and Greg Reale (1972), "Students Attitudes Toward Business and Marketing Institutions and Practices," MSF Working Paper, May.
Warner, Lloyed W. (1972), "Classes Are Real" in Issues and Social Equality, G.W. Thubar and S. Feldman, eds. Boston: Little, Brown.
Williams, J. W. (1970), A Gradient of the Economic Concepts of Elementary School Children and Factor s Associated with Cognition," Journal of Consumer Affairs, Vol. 4 (Summer), 113-123.