Self-Esteem, Susceptibility to Interpersonal Influence, and Fashion Attribute Preference in Early Adolescents

ABSTRACT - This study examined the impact of self-esteem and consumer susceptibility to interpersonal influence on the attribute preference of middle school children (grades 6-8). The results indicate that self-esteem is positively associated with the importance given to the utilitarian aspects of clothing, and susceptibility to interpersonal influence is positively associated with the importance placed on the display aspects of clothing. Gender differences in the relationships are also described.

Citation:

Gregory M. Rose, David M. Boush, and Marian Friestad (1998) ,"Self-Esteem, Susceptibility to Interpersonal Influence, and Fashion Attribute Preference in Early Adolescents", in E - European Advances in Consumer Research Volume 3, eds. Basil G. Englis and Anna Olofsson, Provo, UT : Association for Consumer Research, Pages: 197-203.

European Advances in Consumer Research Volume 3, 1998      Pages 197-203

SELF-ESTEEM, SUSCEPTIBILITY TO INTERPERSONAL INFLUENCE, AND FASHION ATTRIBUTE PREFERENCE IN EARLY ADOLESCENTS

Gregory M. Rose, University of Mississippi, U.S.A.

David M. Boush, University of Oregon, U.S.A.

Marian Friestad, University of Oregon, U.S.A.

ABSTRACT -

This study examined the impact of self-esteem and consumer susceptibility to interpersonal influence on the attribute preference of middle school children (grades 6-8). The results indicate that self-esteem is positively associated with the importance given to the utilitarian aspects of clothing, and susceptibility to interpersonal influence is positively associated with the importance placed on the display aspects of clothing. Gender differences in the relationships are also described.

INTRODUCTION

Changes in the structure of American society have greatly increased the consumption power of adolescents. With the growth of single parent and dual income families, children have assumed greater responsibility in the home and increasingly participate in a variety of consumption decisions (Stipp 1988).

Research on adolescents, however, has generally not been commensurate with their importance as consumers. More specifically, vrtually no research has addressed what motivates the buying decisions of early adolescents (aged 12-14). The developmental psychology literature suggests that peer evaluations strongly influence adolescents (Harter 1990; Mack and Ablon 1983). Early adolescence is a crucial period in which the social aspects of the self become dominant and peer relationships become paramount. Normative influence, therefore, should have a particularly strong effect on the behavior of early adolescent consumers. Typically in studies of adults, normative influence has been investigated as an individual difference variables such as consumer susceptibility to interpersonal influence (CSII) or attention to social comparison information (ATSCI), which is related to more global individual difference variables such as self-esteem (Bearden, Netemeyer, and Teel 1989; Bearden and Rose 1990).

In contrast, studies of adolescent consumers have typically focused on the environmental influence of various socialization agents, such as parents and the media, and on the development of broad, general traits, such as materialism (Moschis 1987). Thus, little is known about the influence of individual difference variables, particularly normative influence, on early adolescent consumers.

This study examines the influence of self-esteem and interpersonal influence on the importance early adolescents place on specific aspects of clothing. Early adolescence represents a critical period in the development of social sources of self-esteem. Self-esteem is fundamental to adolescent development and should be related to more specific individual difference variables. Although normative influence in consumer behavior has been examined among adults, interpersonal influence should be particularly important among early adolescents. Thus, this study examines the effect of self-esteem and consumer susceptibility to interpersonal influence (CSII) (Bearden, Netemeyer, and Teel 1989) on the relative importance placed on the utilitarian (e.g. ease of care, value) and display aspects of clothing (e.g. brand name and what others wear).

Clothing was chosen as the focal product in this study because attribute preferences in clothing are likely to reflect strong normative influences. Clothing is a particularly visible means of consumption and self-presentation is both a preoccupation of adolescents (Harter 1990) and a central motivation in clothing selection (Back 1985).

MODEL DESCRIPTION

Overview

There are four basic constructs in this study: self-esteem, CSII, and the display and utilitarian aspects of clothing. CSII measures the degree to which a person is influenced by others in consumption decisions. Past studies have found a negative relationship between self-esteem and CSII (Bearden et al. 1989). The display aspects of clothing examines the importance of the external and status dimension of dress (such as brand name and what others wear). We expect a positive relationship between the importance attached to the display aspects of clothing and CSII because both are concerned with the opinions of others. The utilitarian aspects of clothing measures the importance placed on the functional aspects of what people wear (such as value and durability). Self-esteem is expected to be positively associated with this dimension because it reflects personal judgments, and high self-esteem is associated with an independence in judgment (McGuire 1968). This argument forms the basis of the model presented in Figure 1. Self-esteem is positively related to the utilitarian aspects of clothing, CSII is positively related to the display aspects of clothing, and CSII and self-esteem are negatively related.

Self-Esteem

Conceptions of self-esteem have always had a social dimension. James (1892 distinguished between the self as knower and thinker (the I) and the self as an object of reflection (the Me). Modern conceptions of self-esteem also include an interpersonal dimension. Rosenberg (1965) focused on the social development of self-image during adolescence. Self-esteem was derived from a series of self-estimates and the evaluation of self was inextricably linked to the reflected appraisals of others (Wells 1976). Social comparison theory (Festinger 1954) explicitly encompasses the comparative aspect of self-evaluation. People have a basic need to evaluate their opinions and abilities and this need can be satisfied by comparing oneself with others. In the absence of an objective or physical basis of comparison, people utilize a social basis of comparison, such as their peers.

In short, self-esteem is linked to the reflected appraisals of others, and can be seen as a prism through which an individual views the world. It is a broad global construct which influences both self-perception and social behavior.

Self-Esteem and Gender. Gender differences in sources of self-esteem have been studied among both adults and adolescents. Generally, the adult male’s self-esteem is rooted in instrumental action, while the adult female tends to derive her self-esteem from relationships (Spence, Deaux, and Helmreich 1985). Men not only derive self-esteem from their location in a status hierarchy but they are also more likely than women to create such hierarchies to enhance their self-esteem. Women have been found to be more relationship oriented, to possess a greater desire for affiliation, and to be more concerned with promoting interpersonal harmony (Schwalbe and Staples 1991).

FIGURE 1

CONCEPTUAL MODEL

As children move into adolescence they are concerned with conforming to adult-typed sex roles (Allgood-Merten and Stockard 1991). Girls suppress individualistic tendencies, while boys suppress their communal, expressive, and tender qualities (Huston and Alvarez 1990). Boys age 12-14, tend toward instrumentalism, and emphasize success, achievement, and being at the top. Girls this age place great value on promoting and maintaining relationships. They are highly self-conscious, more vulnerable to criticism, and more concerned with promoting interpersonal harmony (Rosenberg and Simmons 1975, 157). Thus, early-adolescent girls and boys may have different sources of self-esteem. Girls emphasize external, interpersonal sources and attach more importance to the reaction and opinions of others in forming their self-concept, while boys are more concerned with their position in a status hierarchy (Rosenberg and Simmons 1975). These gender differences become evident in early adolescents and may be maintained throughout adulthood.

Self Esteem And Influencability. Self-esteem was one of the first individual difference variables studied in persuasion research (Hovland and Janis 1959). McGuire (1968) posited a general trait of influencability, which included an individual’s tendency to conform to normative pressure and to yield to persuasive communications. Conformity research has demonstrated the importance of social influence on behavior and found extensive differences in the tendency of individuals to yield to group influence (Asch 1956).

Developmental research reveals that adolescents are extremely concerned with what others think. During early adolescence the scope and number of significant others expands and there is a substantial amount of effort devoted to placing oneself in this expanded social network (Mack and Ablon 1983). Just prior to early adolescence, comparisons with others are first invoked as a barometer of skills and attributes (Harter 1990). Early adolescents are particularly sensitive to the opinions of others and increasingly rely on their peers for information (Mack and Ablon 1983). Peer popularity becomes an important measure of self-worth and early adolescents become pre-occupied with impression management (Harter 1990).

Studies of consumer behavior also support the link between normative influence, self-esteem, and behavior. In a series of studies, Bearden and Rose (1990) describe the impactof attention to social comparison information (ATSCI) on buyer behavior among college students. They found that ATSCI was negatively correlated with self-esteem and positively related to a tendency to comply with normative pressure.

CSII is conceptually related to ATSCI. It measures the extent to which a consumer (1) uses social comparison information to choose products and brands; and (2) makes purchases to conform to the expectations of others. It has two basic dimensions: normative and informational. The normative dimension measures the propensity of an individual to use products or brands to comply with the expectations of others, and the informational component measures an individual’s tendency to learn about products by observing others. Both the normative and the informational components of the scale were negatively correlated to self-esteem (Bearden, Netemeyer, and Teel 1989).

HYPOTHESES

Both the psychological and consumer behavior literature suggest a negative correlation between self-esteem and susceptibility to interpersonal influence. Moreover, the normative component of consumer preference is well established and studies have found a negative correlation between self-esteem and social comparison information (Bearden and Rose 1990; Bearden, Netemeyer, and Teel 1989). These findings lead to the first hypothesis:

H1: Self-esteem will be negatively related to consumer susceptibility to interpersonal influence (CSII).

If as the literature suggests girls place more emphasis on interpersonal sources of self-esteem, and girls are more concerned with the evaluation and reflected appraisals of others:

H2: The negative relation between self esteem and consumer susceptibility to interpersonal influence will be stronger for girls than for boys.

The psychology literature provides a strong theoretical rationale for an association between self-esteem, susceptibility to interpersonal influence, and dress. A central tenet of social comparison theory is that external interpersonal cues are used as a referent for comparison. Because clothing is an especially distinct personal cue, individuals high on the need for social comparison should attend more vigilantly to the dress of others. Consumers susceptible to interpersonal influence should utilize dress as an important normative and informational cue, and as a means of indicating the acceptance of group norms (Bearden and Rose 1990). In short, consumers susceptible to interpersonal influence will place a greater value on the display aspects of clothing.

H3: Consumer susceptibility to interpersonal influence (CSII) will be positively related to the display aspects of clothing.

TABLE 1

CONSTRUCTS AND ITEMS

Adolescent girls are particularly concerned with their self-presentation. Girls generally report higher levels of intimacy with their peers (Blyth and Foster-Clark 1987), derive their self-esteem from the reflected appraisals of others, and are more self conscious and other-oriented than boys (Rosenberg and Simmons 1975).

H4: The association between consumer susceptibility to interpersonal influence (CSII) and the display aspects of clothing will be stronger for girls than for boys.

Individuals with high self-esteem are less influenced by the opinions of others, more independent, and less likely to conform to group pressure than individuals with low self-esteem (McGuire 1968). The utilitarian aspects of clothing are directly related to the ability of the clothes to meet an individual’s functional needs, and individuals with high self-esteem should focus on how the clothes serve their needs, not the message their clothes impart to others. That is:

H5: Self-esteem will be positively related to the utilitarian aspects of clothing.

Finally, differences in the relation between self-esteem and the utilitarian aspects of clothing will be examined between boys and girls; this examination, however, will be exploratory and no hypothesis will be advanced.

METHOD

Respondents were drawn from two middle schools in a medium-sized city in the pacific northwest. The two schools were selected for convenience and respondents were not intended to be representative of middle-school students in general. A total of 346 boys and 412 girls completed a survey that included questions on self-esteem, consumer susceptibility to interpersonal influence (CSII), and the motivation for selecting clothes. Additional items that measured media habits and attitudes toward advertising were contained in the questionnaire but are not included in this analysis. In total, 758 respondents completed the survey with approximately 35% in the sixth grade, 34% in the seventh grade, and 31% in the eighth grade.

Three global self-esteem items were taken from Rosenberg (1965). These items represented the students’ general self-contentment, optimism, and perceived capabilities. Consumer susceptibility to interpersonal influence was assessed using three items from the scale developed by Bearden, Netemeyer, and Teel (1989). CSII has been extensively validated among adults and provides a reliable measure of interpersonal influence in the consumer domain. A pre-test of 33 sixth through eighth graders resulted in a slight wording modification in one item in this scale. Finally, the 9 fashion related items were based in part on a list developed by Prakash (1984). The items in this study are contained in Table 1. A five point scale was used for all measures (1=strongly disagree to 5=strongly agree).

Data AnalysisBSplit-Sample

Data analysis was conducted with LISREL VII and split-multi-sample analysis was employed. Initially, the sample was split by gender. Then both the girls’ and the boys’ samples were randomly split to form a by-gender analysis and a by-gender hold-out sample. Each model was fit twice, first for the analysis and then for the hold-out sample, and each model contained two groups: boys and girls. After this analysis revealed similar structures in the sub-samples the data were recombined.

Table 2 outlines the measures of fit obtained for both the analysis and the holdout samples for the four models. The first, is our final model, as shown in Figure 1. The second is a null model which was run as a basis of comparison. In the third model, the lambdas were constrained to be equal across groups. This was done to check the reliability of the factor structure; that is, to insure that the same factor structure fit both the boys and the girls. Finally, in the fourth model the factor loadings (lambdas), factor variance (phis and psis), and error variances (the theta deltas and theta epsilons) were all constrained to be equal across groups. This provides a test of both factor and error invariance.

TABLE 2

MEASURE OF OVERALL MODEL FIT

Chi square difference tests indicate that the factor loadings were invariant between groups. No statistical difference, p<.05, was found between Model 3 and our final model for either the analysis sample (c² difference=15.51 for 11 degrees of freedom) or the hold-out sample (c² difference=19.26 for 11 degrees of freedom). Constraining the error and factor loadings to equality, as well as the factor loadings (Model 4), produced a significantly worse fit (p<.01, c² difference=60.58 for the analysis sample and c² difference=65.73 for the hold-out sample with 31degrees of freedom) than when these estimates were allowed to remain free between groups. Constraining the factor loadings, error variances, and factor variances across groups, however, is an extremely stringent test and there is no reason to expect the error variances of boys and girls to be similar for all the constructs in our study.

Final Model Fit

Although based on the chi-square test the model should be rejected (p<.000), the sensitivity of this test to sample size is well documented (Bagozzi and Yi 1988; Joreskog and Sorbom 1983). LISREL’s Goodness-of-Fit-Index (GFI) and the Tucker-Lewis Index (TLI) were therefore utilized as alternative measure of fit. Both indicated an acceptable level of fit (about .90) for both the analysis and the hold-out sample.

Tables 3 summarizes the Common Metric Standardized Loadings and the T-values for the analysis sample. These loadings were also examined for the hold-out sample. All the t-values of our model were significant at the .05 level for both the analysis and hold-out sample. Furthermore, the pattern of loadings was similar across samples.

The structural coefficients were also examined for both the analysis and the hold-out samples. All structural coefficients were significant (p<.05) and in similar directions across samples. At this point the analysis and hold-out samples were combined. The relative stability of the results had been demonstrated and the split-sample analysis had served its purpose.

RESULTS

Full Sample Analysis

Examining the coefficients of determination for both the x and y variables indicated an acceptable level of fit. The total coefficient of determination for the x variables was .945 for the boys and .929 for the girls, and the total coefficient of determination for the y variables was .980 and .979, respectively. The percent of variance extracted and the construct reliability were also calculated. These values appear in Table 4.

Hypotheses Tests

The hypotheses were tested utilizing the full sample. The goodness of fit improved slightly from the split-sample estimates (from .896 to .934 for the boys and from .914 to .925 for the girls). Path coefficients for the final model are listed in Table 5.

Hypothesis one was not confirmed. Self-esteem was negatively related to CSII only for the girls’ sample (p<.01). For the boys’ sample, the relation was opposite the hypothesized direction; self-esteem was positively related to CSII (p<.01). Therefore, contrary to hypothesis, self esteem was not more negatively related to CSII for girls than for boys. Instead, the strength of the association was similar but in opposite directions.

Hypotheses three through five were confirmed. Consumer susceptibility to interpersonal influence was positively related to the display aspects of clothing (hypotheses 3, p<.01) for both boys and girls, and the relationship was significantly stronger for girls (p<.05). Hypothesis four was tested by constraining the path between CSII and the display aspects of clothing to be equal across genders an comparing the chi-square to the final model without this constraint. Allowing the path between CSII and the display aspects of clothing to be free resulted in a significant improvement in chi-square (4.05 for 1 degree of freedom, p<.05). Although this improvement was sufficient to be statistically significant, the magnitude of the chi-square indicates that support for hypotheses 4 should be regarded as marginal.

TABLE 3

COMMON METRIC STANDARDIZED LOADINGS AND (T-VALUES)-ANALYSIS SAMPLE

TABLE 4

CONSTRUCT RELIABILITY AND VARIANCE EXTRACTED-FULL SAMPLE

TABLE 5

COMMON METRIC PATH ESTIMATES AND (T-VALUES)-FULL SAMPLE

As predicted in hypothesis five, self-esteem was positively related to the utilitarian aspects of clothing (p<.01). This relationship was especially strong for boys. A significant improvement in chi-square (7.84 for 1 degree of freedom, p<.01) was obtained by allowing different path estimates for the boys and the girls.

In short, three of the five hypotheses were confirmed and the two paths between the endogenous and exogenous variables were as hypothesized. Self esteem was positively related to the utilitarian aspects of clothing and CSII was positively related to the display aspects of clothing. Moreover, the correlation between the display aspects of clothing and CSII was stronger for girls, and the relation between self-esteem and the utilitarian aspects of clothing was stronger for boys. The major surprise in this study was the different association between self-esteem and CSII by gender. As expected, self-esteem was negatively related to CSII for girls, but, contrary to our hypothesis, the two variables were positively related for boys. This finding will be examined further in the discussion section.

DISCUSSION

As expected consumer susceptibility to interpersonal influence (CSII) was positively associated with the display aspects of clothing. This is consistent with previous research that linked attention to social comparison information with a tendency toward conformity (Bearden and Rose 1990) and CSII with a tendency to engage in normative behavior (Bearden, Netemeyer, and Teel 1989). Adolescents with high self-esteem, on the other hand, placed more value on the non-normative (utilitarian) aspects of clothing. This can be viewed as consistent with past research on interpersonal influence (Bearden, Netemeyer, and Teel 1989; McGuire 1968) and dress (Lennon, Fairhurst, and Peatross 1991). Thus, our study substantiates the fundamental link between susceptibility to interpersonal influence and attribute-preference found in adults and extends these findings to adolescents.

The influence of gender on self-esteem, CSII, and clothing preference is more complex than first thought. As hypothesized, the association between CSII and the display aspects of clothing was stronger for girls. Consistent with these findings, we find that susceptibility to interpersonal influence is more strongly associated with the display aspects of clothing for girls than for boys. Moreover, self-esteem and the utilitarian aspects of clothes were more strongly related for boys.

The major surprise in this study: that self-esteem and CSII were negatively related for girls but positively related for boys may be explained as follows. CSII and self-esteem may be negatively related for girls because the stronger a girl’s self-esteem, the less she needs to compare herself to others, and the less influenced she is by the opinions of others. For boys, however, the need to compare oneself to others may stem from the need to determine ones’ relationship in the status hierarchy (Schwalbe and Staples 1991). For adolescent boys, high CSII may be an indication of high self-esteem because social comparisons are only made after a boy believes he is an accepted member of a status hierarchy. Boys with high self-esteem may be more interpersonal and compare themselves to others more frequently than low self-esteem boys. This would suggest that the nature of CSII is fundamentally different for boys than for girls.

Another explanation may have to do with the validity of the self-esteem and CSII measures for girls (see Table 4). Current measures of self-esteem may de-emphasize the relational aspects of self-esteem favored by girls.

In summary, this study helps to clarify the purchase motivations of middle school children and illustrates the normative aspect of adolescent preference for clothing. The association between susceptibility to interpersonal influence and the display aspects of clothing was confirmed for both girls and boys. Differences, however, were found in the strength of this relationship. These findings extend previous adult research on normative influence to middle-school children and show the importance of self-esteem and interpersonal influence in clothing choice.

REFERENCES

Allgood-Merton, Betty and Jean Stockhard (1991), "Sex Role Identity and Self-Esteem: A Comparison of Children and Adolescents," Sex Roles, 25 (3/4), 129-138.

Asch, Solomon E. (1956), "Studies of Independence and Conformity: A Minority of One Against the Majority," Psychological Monographs, 70, No. 9 (Whole No. 416).

Back Kurt W. (1985), "Modernism and Fashion: A Social Psychological Interpretation," The Psychology of Fashion, Michael R. Solomon (Ed.), Lexington, MA: Lexington Books.

Bagozzi, Richard P. and Youjae Yi (1988), "On the Evaluation of Structural Equation Models," Journal of the Academy of Marketing Science, 16(1), 74-94.

Bearden, William O. and Randall L. Rose (1990), "Attention to Social Comparison Information: An Individual Difference Factor Affecting Consumer Conformity," Journal of Consumer Research, 16 (March), 461-471.

Bearden, William O., Richard G. Netemeyer, and Jesse E. Teel (1989), "Measurement of Consumer Susceptibility to Interpersonal Influence," Journal of Consumer Research, 15 (March),473-81.

Blyth, D. A., and F.S. Foster-Clark (1987), "Gender differences in perceived intimacy with different members of adolescents’ social networks," Sex Roles, 17, 689-719.

Festinger, Leon (1954), "A Theory of Social Comparison Processes," Human Relations, 7, 117-140.

Harter, Susan (1990), "Processes underlying adolescent self-concept formation." In R. Montemayor, R. R. Adams, and T. P. Gullotta (Eds.), From childhood to adolescence: a transitional period? Newbury Park, CA: Sage Publications.

Hovland, Carl, and Irving L. Janis (1959), Personality and persuasibility. New Haven: Yale University Press.

Huston, A. C. and M. Alvarez (1990), "The socialization context of gender role development in early adolescence." In R. Montemayor, R. R. Adams, and T. P. Gullotta (Eds.), From childhood to adolescence: A Transitional Period? Newbury Park, CA: Sage Publications.

James, William (1892), Psychology: The Briefer Course. New York: Holt.

Joreskog, Karl J. and Dag Sorbam, (1983), LISREL User Guide Version V and VI, Uppsala, Sweden: University of Uppsala.

Lennon, Sharron J., Ann Fairhurst, and Freida Peatross (1991), "Apparel and Furniture Attribute Importance as a Function of Self-Monitoring," Home Economics Research Journal, 19 (4), 292-302.

Mack, J. E. and S. L. Ablon, (1983). The development and sustenance of self-esteem in childhood. New York, NY: International University Press, Inc.

McGuire, William J. (1968), "Personality and Susceptibility to Social Influence," Handbook of Personality Theory and Research, Edgar F. Borgatta and William W. Lamber (Eds), Chicago: Rand McNally, 1130-1188.

Moschis, George P. (1987), Consumer Socialization: A Life-Cycle Perspective, Lexington, MA: Lexington Books.

Prakash, Ved (1984), "Personal Values and Product Expectations," in Personal Values and Consumer Psychology, Robert E. Pitts, Jr. and Arch G. Woodside (Eds). Lexington, MA: Lexington Books, 145-154.

Rosenberg M. (1965) Society and the Adolescent Self-Image. Princeton , N.J: Princeton University Press.

Rosenberg, Florence R. and Robert.G. Simmons (1975), "Sex Differences in the Self-Concept of Adolescence," Sex Roles, 1 (2), 147-159

Schwalbe, Michael L. and Clifford L. Staples (1991), "Gender Differences in Sources of Self-Esteem," Social Psychology Quarterly, 54 (2), 158-168.

Spence, Janet T, Kay Deaux, and Robert Helmreich (1985), "Sex Roles in Contemporary American Society," The Handbook of Social Psychology, 3(2), Elliot Aronson and Gardner Lindzey (Eds.)New York, NY: Random House.

Stipp, H.H. (1988), "Children as consumers," American Demographics, 10 (February), 27-32.

Wells, L. E. (1976) Self-Esteem. Beverly Hills, CA: Sage Publications.

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