Opinion Leadership and Self-Concept: an Empirical Examination

ABSTRACT - Over the years, opinion leadership has been studied mostly in the context of new product diffusion models. Self-concept has been studied mostly in the context of matching product and self perceptions or the relationships between ideal, real, personal and social self-concepts. We examine these two concepts and develop hypotheses about the role of some facets of self-concept in explaining product-type opinion leadership, another facet of self-concept. The model is tested empirically and the data mostly support the research hypotheses.

Citation:

Aviv Shoham, Gregory M. Rose, and Lynn R. Kahle (1998) ,"Opinion Leadership and Self-Concept: an Empirical Examination", in E - European Advances in Consumer Research Volume 3, eds. Basil G. Englis and Anna Olofsson, Provo, UT : Association for Consumer Research, Pages: 204-210.

European Advances in Consumer Research Volume 3, 1998      Pages 204-210

OPINION LEADERSHIP AND SELF-CONCEPT: AN EMPIRICAL EXAMINATION

Aviv Shoham, TechnionBIsrael Institute of Technology, Israel

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

Lynn R. Kahle, University of Oregon, U.S.A.

ABSTRACT -

Over the years, opinion leadership has been studied mostly in the context of new product diffusion models. Self-concept has been studied mostly in the context of matching product and self perceptions or the relationships between ideal, real, personal and social self-concepts. We examine these two concepts and develop hypotheses about the role of some facets of self-concept in explaining product-type opinion leadership, another facet of self-concept. The model is tested empirically and the data mostly support the research hypotheses.

OPINION LEADERSHIP AND SELF-CONCEPT: A PRODUCT-TYPE EXAMINATION

Opinion leaders are individuals to whom others look for information (Childers 1986; Feick and Price 1987). Most prior research was devoted to understanding the role of opinion leaders in the diffusion of new innovations (Gatignon and Robertson 1985), but the relationship between opinion leadership and other facets of self-concept has ben largely ignored (Chan and Misra 1990). Self-concept has been defined as one’s perception of oneself and has been studied in the context of the congruity of ideal and real self-concepts and the congruity of self-, brand-, and product-concept (Sirgy 1982). Understanding the relationship between opinion leadership and other facets of self-concept is important because these facets can provide a rich "menu" for targeting opinion leaders. For example, these facets can be used to develop advertising messages and to select models and celebrities with perceived profiles that are similar to opinion leaders’. Additionally, prior research has established that opinion leadership is product-specific (King and Summers 1970). Thus, knowledge about opinion leaders is mostly at the specific product level. We study opinion leadership in a more general, product-type level. The purposes of this paper are to bring opinion leadership and other facets of self-concept into a joint framework.

Opinion Leadership

An opinion leaders can influence the attitudes and actions of others (Rogers 1983) and have been studied in the context of diffusion of new innovations (Gatignon and Robertson 1985). Research based on a two-step diffusion model has examined the role played by opinion leaders in the diffusion of new innovations (Arndt 1967; Katz and Lazersfeld 1955; King and Summers 1970). Opinion leadership varies by productBa generalized opinion leadership does not exist (King and Summers 1970; Langeard, Crousillat, and Weisz 1977; Myers and Robertson 1972). Opinion leaders tend to be category-, rather than product- or brand-specific (King and Summers 1970; Myers and Robertson 1972) making the trait monomorphic rather than polymorphic (Merton 1957). There is another type of consumerBmarket mavenBthat is less a leader and more an information provider for multiple types of products and services (Feick and Price 1987; Lichtenstein and Burton 1990; Slama and Williams 1990).

Identification of opinion leaders is problematic because their demographic and life-style profiles are similar to their followers’ (Myers and Robertson 1972). Opinion leaders are technically competent and have high exposure to mass media (Higie, Feick, and Price 1987; Rogers 1983; Summers 1970). They are socially active, so they are likely to hold offices in community activity groups (Baumgarten 1975; Venkatraman 1989). Opinion leaders are appearance-conscious, self-centered (Baumgarten 1975), and self-confident (Summers 1970) and they exhibit public individuismBthey feel and act differently (Chan and Misra 1990).

Other Facets of Self-Concept

Studies of self-concept (e.g., Gardner and Levy 1955; Levy 1959) have used single- or multiple-component perspectives (Sirgy 1982). Multi-component studies have examined combinations of ideal versus actual and social versus private self-concepts (Kassarjian and Sheffet 1991; Sirgy 1982). Self-concept theory is based on an individual’s desire to attain self-consistency and enhance self-esteem. To achieve self-consistency, consumers purchase products that fit their self-concept. This behavioral pattern is based on the self- and product-image congruity theory (Sirgy 1982). Most research on self-concept examined its influence on product choice. Dolich (1969) found a greater similarity between self-concept and images of preferred than of less preferred beer, soap and toothpaste brands. Grubb and Grathwohl (1967) found that consistent brand- and self-concept results in highest consumer preference and Grubb and Hupp (1968) reported consistent perceptions of owned automobiles and self-concept. White (1967) identified low, medium, and high discrepancy segments and reported that medium-discrepancy consumers tend to own compact cars more than the other two groups. Guttman (1973) noted that light television viewers tend to perceive themselves as achieving and active in contrast to hevy viewers who perceive themselves as more sociable.

Self-concept and opinion leadership. The research hypotheses outlined below are based on the congruity argument. Whereas previous research examined the relationship between self-concept and behavior, the argument for congruity applies equally to relationships within self-concept. In other words, an individual’s self-concept profile should be internally consistent. Furthermore, while opinion leadership can be seen as a trait, it has a behavioral facet as well (such as providing information to others). In either case, we focus on congruity between opinion leadership and other facets of self-concept. Such internally-consistent relationships have been documented in previous research. Prince (1993) found consistent relationships between self-concept, money beliefs, and money values. For example, a distrustful self-concept was related positively to the belief that "in business, everyone has a price" and a low self-esteem self-concept was related to the belief that "it is important that I wear expensive clothes" (Prince 1993, p. 171). Thus, the general research proposition is:

P: There exists a relationship between an individual’s self-concept and the extent to which (s)he is an opinion leader.

Opinion leaders are narcissistic, exhibitionists, and appearance-oriented (Baumgarten 1975; Summers 1970). Thus, a positive relation between an appearance-based self-concept and opinion leadership is expected (we discuss appearance-based self-concept in a later section of this article). Similarly, because they emphasize their appearance, we expect a positive relationship between the self-concept of being beautiful and opinion leadership:

H1: The more positive the internal and external self-concept and the more beautiful the self-perception of an individual, the higher the individual’s opinion leadership.

Opinion leaders have high incomes and occupational status (Summers 1970). Because of its role in adoption of new innovations (Myers and Robertson 1972), opinion leadership requires that one has the resources to purchase new products. Thus, a positive relationship is expected between the self-concept of spender and opinion leadership. Individuals also have to hold positive expectations about the future if they are to spend money. Kahle, Shoham, Rose, Smith, and Batra (1995) establish a link between optimism and attitudes and behavior. Thus, a positive relationship is expected between being optimistic and opinion leadership:

H2: The higher an individual’s self-concept of being a spender and expecting the best, the higher the individual’s opinion leadership.

Opinion leadership and innovation are not synonymous, but traits of individuals in both groups are similar (Dutton, Rogers, and Jun 1987; Myers and Robertson 1972). Adoption of new innovations requires that individuals be modern, leaders, and adventurous rather than traditional, followers, and cautious (Goldsmith and Goldsmith 1980; Jacoby 1971). Adoption of new products may also be associated with a sense of excitement and sophistication. Thus:

H3: The higher an individual’s self-concept of being sophisticated, exciting, modern, a leader, and adventurous the higher the individual’s opinion leadership.

Prior research has established that opinion leaders are nrcissistic, socially active, and gregarious (Baumgarten 1975; Summers 1970). Outgoing individuals, who are more gregarious and active, will be able to provide leadership to other people.

H4: The higher an individual’s self-concept of outgoing, the higher the individual’s opinion leadership.

METHOD

Sample and Procedure

The data for this research are from a large-scale study of US female household heads. Data were collected over 11 months in 1993 by a professional market research firm. Respondents were selected to be nationally representative on age and education. Age quota was 35% for the 18-34, 35% for the 35-54, and 30% from 55 and over, not more than half over 65. Education quota was 20% with some high school, 40% high school graduates, 25% with some college, and 15% college graduates. Respondents did not participate in similar interviews in the preceding year.

Respondents were recruited on the basis of the age and education quota. They were invited to a hotel for a three hours’ data collection session (see Table 1 for descriptive statistics). In each month, 130 individuals were recruited to increase the probability of at least 100 shows at the session. A reminder letter was mailed a few days before the session. In all, 1078 individuals participated in the 11 sessions.

Measures

Opinion Leadership. The questionnaire included 12 items measuring opinion leadership modeled after one of King and Summers (1970) questions, which was also included in Childer’s scale (1986). Respondents were asked to mark, on ten-point scales (0=less likely to 10=more likely): "Compared with most other adults you know, how likely are you to be asked for your ides or advice on each of the following topics?" This was followed by a list of 12 items, such as "the latest clothing fashions", "sports", and "food nutrition" (Table 2).

The self-designating method of measuring opinion leadership was preferred to sociometric techniques because of the need to administer the questionnaire to a large number of respondents. The main criticism of self-designated measurement is that respondents might inflate their scores (Darden and Reynolds 1972). An examination of Table 2 shows that this is not the case. Mean values on opinion leadership vary between 2.87 (for sports) and 6.50 (for cooking). Most of the means are close to the mid-point on the scales (5.0). In fact, the only three cases for which the mean exceeds 6.0 are for cooking, health, and nutrition. One would intuitively expect females heads of households to be knowledgeable about these issues.

Four sub-scales were created to capture opinion leadership for health, beauty, home, and money. The four sub-scales exhibit acceptable levels of reliability (Nunnally 1967). The first included three health items: health issues, food nutrition, and exercise and fitness. Coefficient a for the scale was 0.71. The second included two beauty items: the latest clothing fashions and beauty aids (such as cosmetics). Coefficient a was 0.66. The third sub-scale included four home-related items: raising children and babies, cooking methods and recipes, health issues, and nutrition. Scale reliability was 0.65. The money sub-scale included two items: money matters and business issues. Coefficient alpha was 0.62.

To establish convergent validity, the relationships of the scores on the four sub-scales with related constructs were examined. Opinion leadership should be correlated with experience with new products (Childers 986). Respondents sub-scales’ scores were correlated with a question designed to measure the extent to which they are likely to try new products (respondents indicated, on a ten-point scale whether they are "...always the first among [their] friends to try out new products). Given Childers’ mild correlations, we expected a similar pattern. The correlations between trying out new products and opinion leadership were .29 (p<.01), .14 (p<.01), .08 (p<.02), and .04 (p<.17) for beauty, health, home, and money opinion leadership respectively. Additionally, opinion leadership has been linked to magazines readership (Baumgarten 1975; Summers 1970). The correlations of the four sub-scales with three printed media readership items were used. In these items, respondents were asked to rate the extent to which they read magazines, the daily, and the Sunday papers. Nine of the possible 12 correlation coefficients were positive and significant as expected. In sum, the four scales exhibit convergent validity.

Self-Concept. Two sets of variables were used to measure self-concept. The first set included 14 bi-polar items, such as modern versus traditional and conservative versus liberal. All items were rated on ten-point scales. These items cover the content dimension of self-concept (Sirgy 1982), and, more specifically, the aspects of non-physical disposition and social-identity. The direction dimension of self-concept was also covered by these items. Of the 14 items, we only used the nine self-concept items covered in the research hypotheses.

The second set includes six items covering self-feelings that are concerned with outer-physical self-concept (e.g., feelings about one’s weight and sex appeal) and with inner-physical self-concept (e.g., feelings about one’s overall energy level and overall health). The six items were measured on ten-point scales (0=have strong negative feelings to 10=have strong positive feelings). Confirmatory factor analysis resulted in two interpretable factors with eigen-value>1.0. The three items for internal physical self-concept loaded on one and the three for external physical self-concept loaded on the other. Therefore, two averaged scales were formed. The first scale ("External") includes three itemsBoverall physical appearance, weight, and sexual appeal. The scale’s a is 0.74 and minimal inter-item correlation coefficient is 0.39. The second scale ("Internal") includes three itemsBoverall energy level, overall health, and overall stress level. The scale’s a is 0.67 and lowest inter-item correlation is 0.35.

TABLE 1

SAMPLE DESCRIPTIVE STATISTICS

TABLE 2

OPINION LEADERSHIP DESCRIPTIVE STATISTICS

TABLE 3

REGRESSION RESULTS

To assess the convergent validity of the two scales, their correlations with the self-concept bi-polar items were measured. Eighteen of a total 22 correlations were significant. High scores on the scales are associated with individuals who are exciting, beautiful, forgiving, outgoing, leaders, optimistic, adventurous, and spontaneous. As none of the correlations exceeded 0.36, the two sets capture related, but distinct dimensions of self-concept, supporting the argument for discriminant validity of the two sets. To further establish the convergent validity of the external dimension of the physical self-concept scale, its correlation with an item measuring dieting behavior was calculated. The correlation coefficient was -0.21, indicating that individuals with low external physical self-concept tend to diet more.

Self-concept and opinion leadership (as operationalized here) may both be self-traits. Thus, there is a need to establish discriminant validity for the two constructs. Earlier, we argued that the two sets should be related. In fact, the research proposition is based on congruity between self-concept and opinion leadership. Therefore, the two sets should be related, but correlations between the two sets should be moderate. Of the 44 correlation coefficients, 39 (88.6 percent) were significant (p<0.05). The largest correlation coefficient was for beauty opinion leadership and boring/exciting self-concept (r=0.31). The mean correlation coefficient was a moderate 0.14. These coefficients, in combination, suggest that the two sets are not only theoretially, but also empirically discriminant.

Findings

Multiple regression analysis was used as a statistical technique. A separate multiple regression was run for each of the four opinion leadership scales (Table 3). A product-class opinion leadership measure was the dependent and the 11 self-concept items and scales were the explanatory variables in each regression. The four regression models explain 7-24 percent of the variance in opinion leadership. Arguably, the relationships uncovered in these models (discussed below) may be due to the large sample size in this study. To assess the impact of this argument, we re-analyzed the data after splitting the sample into five sub-sets with approximately 200 respondents in each. We note here that the results were substantively similar to those reported for the complete sample. Therefore, it appears that the findings are substantive and not an artifact of sample size.

H1 posited that positive internal and external physical self-concept and perception of beauty are associated with higher opinion leadership scores. The data provide support to H1, in most cases. Internal and physical self-concept has a significant (P<0.05) and positive impact on opinion leadership on beauty, health, and money. External and physical self-concept has a significant (P<0.05) and positive impact on beauty, health, and money. The effect of one’s self-concept of being beautiful has a significant (P<0.01) positive impact on leadership for beauty products. In sum, 7 of the 12 effects are significant and in the expected direction.

H2 posited a positive relationship between one’s self-concept of being a spender and optimist and opinion leadership. The data mostly support H2 with regards to the effect of optimism on health, home, and money opinion leadership. However, whereas being a spender is associated significantly (P<0.01) and positively with opinion leadership for beauty as expected, the association for the other three dimensions is negative and reaches significance (P<0.01) for health and money opinion leadership.

An individual’s self-concept of being sophisticated, exciting, modern, and leader was expected to have a positive impact on her opinion leadership (H3). The data support some of these relationships. Sophistication has a positive and significant (P<0.01) effect on beauty and money leadership. Being exciting has a positive and significant (P<0.01) effect on beauty leadership. Being modern has a positive impact (P<0.01) on beauty leadership as expected. However, contrary to H3, being modern has a negative and significant (P<0.05) effect on health and home opinion leadership. Being a leader has a positive and significant (P<0.01) effect on opinion leadership for health, home and money as expected, but not on beauty. Being adventurous has a positive effect on leadership, but only for money issues (P<0.01).

Being an out-going rather than shy was hypothesized to have a positive effect on opinion leadership (H4). This indeed was the case for beauty aids.

DISCUSSION

The results of this research mostly support the general proposition (P) and four research hypotheses. Both internal and external dimensions of physical self-concept affected opinion leadership similarly. The more positive one’s feelings about aspects such as physical appearance, weight, health and stress, the more likely she will be an opinion leader. Both internal and external physical self-concepts have similar effects on leadership for beauty aids and health issues. Thus, the higher the internal and external physical self-concept, the higher the level of opinion leadership. The significant effect of the internal physical self-concept (but not the external) on money leadership is intuitively appealing. One’s appearance should not be related to knowledge about money matters. On the other hand, hih energy, good health, and low stress levels may provide individuals with the physical resources needed to learn about and become opinion leader on business issues.

Contrary to expectations, physical self-concept was insignificant in predicting opinion leadership on home issues, such as raising children and babies and cooking. It may be that one’s appearance is unrelated to opinion leadership on home issues. A woman can become knowledgeable about such issues regardless of how good she feels about her appearance.

A woman’s self-concept of being beautiful was significant in predicting her leadership on issues such as cosmetics and beauty aids. This pattern is similar to findings reported by Baumgarten (1975) and Summers (1970). Interestingly, feeling beautiful did not have a significant impact on the other three opinion leadership scales. Apparently, beauty only enhances the probability of being an opinion leader for issues related to itBclothing and cosmetics. It can be argued that this finding may be trivial because both self-concept of being beautiful and opinion leadership on beauty products may be due to one’s involvement with beauty. We share this concern, but believe that our results are not trivial. In fact, a convincing argument can be made that beauty products are more important to less beautiful individuals. Such individuals may have a stronger need to enhance and improve on the "cards dealt by nature". Thus, less fortunate individuals (appearance-wise) may need such products more than beautiful ones. Apparently, this was not the case in our sample.

Being a spender was related to beauty opinion leadership as expected. The higher the self-concept of being a spender, the higher the opinion leadership on beauty issues. This is similar to Goldsmith et al.’s finding (1991). Fashion and beauty products require high expenditures compared to health, home, and money. In fact, being able to save moneyBget the best dealBon health and banking is an important part of becoming an opinion leader on these topics. This argument is supported by the data as the higher the self-concept of being a saver the higher the self-reported opinion leadership for health and money.

Optimism was expected to affect opinion leadership through its effect on spending (Kahle et al. 1994). The data support this expectationBthe more optimistic the individual, the more she is an opinion leader on health, home, and money. Expectations about the future did not affect a woman’s opinion leadership for beauty products. Individuals in this study know about and use such products regardless of their self-concept of being an optimist.

Being sophisticated had the hypothesized effect on opinion leadership for beauty products, and money. These three industries probably require sophistication. In fact, "a sophisticated look" is a term used often in describing fashion and beauty products. It is also used frequently to describe the "sophisticated investor". It is not associated, in this sample or in everyday terminology, with expertise on home or health issues.

The excitement associated with new fashion trends carries through to opinion leadership on beauty products as expected. The higher the self-concept of being exciting rather than boring, the higher the opinion leadership score for fashion and beauty. This facet of self-concept did not carry through to the more mundane issues of health, food, and money. Individuals in these three product categories can be opinion leaders regardless of their self-ascribed level of being exciting or boring.

Being modern is important in predicting opinion leadership on fashion products. The more modern the individual, the higher her level of leadership on beauty. Conversely, for domestic opinion leadership, the relationships are inverse. A traditional person is more likely to be an opinion leader for education, cooking, and nutrition.

Not surprising, being perceived as a leader, rather than a follower, is associated with opinion leadership for health, home, and money. Beyond its supportive role for the validity o the measures for opinion leadership, the positive effect of perceived leadership suggests that this general trait also carries through to the more specific product category level. Stated differently, individuals who perceive themselves as leaders in general can also be expected to be opinion leaders for specific products and services.

Being cautious had the expected negative effect on opinion leadership. The data support the argument that opinion leaders are also likely to be innovators (Summers 1970; Schrank 1973). Being an innovator requires that an individual also accept the risks inherent in the purchase and trial of new products. The data support this argument for money and business issues. The insignificant effect for opinion leadership on beauty, home, and health may be due to the lower risks associated with trial of new products in these categories.

Outgoing individuals were expected to score highly on opinion leadership (H4). The data provide support for H4 for the beauty sub-dimension. In other words, individuals who perceive themselves as outgoing are leaders for issues such as cosmetics, but not for health, home, and money. It may be that shyness is detrimental only to products that can be observed directly. If this is the case, then the trait of being outgoing will only be related to cosmetics, fashion accessories, and clothing, but not for the other dimensions of opinion leadership.

SUMMARY AND CONCLUSIONS

Based on the findings of this research, the profiles of opinion leaders for the four types of products and services are similar, but not identical. Table 4 summarizes the profiles of opinion leaders for each product type. In general, opinion leaders exhibit self-concepts that are mostly, but not always, associated with positive labels. "Exciting", "beautiful", and "healthy" exemplify positively labeled traits. However, other labels are neutral. For example, "down to earth" or "traditional" may be positive for some individuals, whereas "sophisticated" or "modern" may be positive for others.

An opinion leader on beauty products typically has strong positive feelings about her physical self. These include both the internal and external sub-dimensions of the physical self-concept. Furthermore, such females perceive themselves as beautiful, sophisticated, and exciting. Being modern, outgoing, and spenders round up the profile by suggesting traits that are necessary for fashion leadership.

TABLE 4

PROFILES OF OPINION LEADERSHIP

Leaders for health issues share the same positive feeling on the two physical self-concept sub-dimensions. In most other respects, they differ from fashion opinion leaders. They consider themselves as savers and traditional. They also tend to be intense, leaders and optimistic. Opinion leaders for the home sub-dimension share most of the non-physical traits of health opinion leaders. They also tend to be traditional and optimistic leaders.

Opinion leaders for money and business exhibit a self-concept that is a combination of the other three groups. They share a positive physical internal (but not external) self-concept with opinion leaders on beauty and health. They are sophisticated, rather than down to earth, much like leaders for beauty products. They see themselves as savers like health opinion leaders. They are leaders and optimistic like leaders for health and home products. They also exhibit the distinctive trait of being adventurous.

In sum, some self-concept traits are universal or almost universal (e.g., the two sub-dimensions of physical self-concept, being a leader and being an optimistic). These traits can be used in positioning new products and services in most categories. Other traits are important for one category or affect the categories differently. These should be used with care based on the specific product marketed.

REFERENCES

Arndt, Johan (1967), "The role of Product-related Conversations in the Diffusion of a New Product," Journal of Marketing Research, 4 (August), 291-295.

Baumgarten, Steven A. (1975), "The Innovative Communicator in the Diffusion Research," Journal of Marketing Research, 12 (February), 12-18.

Chan, Kenny K. and Shekhar Misra (1990), "Characteristics of the Opinion Leader: A New Dimension," Journal of Advertising, 19 (3), 53-60.

Childers, Terry L. (1986), "Assessment of the Psychometric Properties of an Opinion Leadership Scale," Journal of Marketing Research, 23 (May), 184-188.

Darden, William R. and Fred D. Reynolds (1972), "Predicting Opinion Leadership for Men’s Apparel Fashions," Journal of Marketing Research, 1 (August), 324-328.

Dolich, Ira J. (1969), "Congruence Relationships Between Self Images and Product Brands," Journal of Marketing Research, 6 (February), 80-84.

Dutton, William H., Everett Rogers, and Sukho Jun (1987), "Diffusion and Social Impacts of Personal Computers," Communication Research, 14 (2), 219-250.

Feick, Lawrence F. and Linda L. Price (1987), "The Market Maven: A Diffuser of Market Information," Journal of Marketing, 51 (January), 83-97.

Gardner, Burleigh B. and Sidney J. Levy (1955), "The Product and the Brand," Harvard Business Review, 33 (March/April), 33-39.

Gatignon, Hubert and Thomas S. Robertson (1985), "A Propositional Inventory for New Diffusion Research, Journal of Consumer Research, 11 (March), 849-867.

Goldsmith, Elizabeth B. and Ronald E. Goldsmith (1980), "Dogmatism and Confidence as Related actors in Evaluation of New Products," Psychological Reports, 47 (December), 1068-1070.

Goldsmith, Ronald E., J. R. Heitmeier, and Elizabeth B. Goldsmith (1991), "Social Values and Fashion Leadership," Clothing and Textiles Research Journal, 10 (Fall), 37-45.

Grubb, Edward L. and Harrison L. Grathwohl (1967), "Consumer Self Concept, Symbolism and Market Behavior: A Theoretical Approach," 31 (October), 25-26.

Grubb, Edward L. and Gregg Hupp (1968), "Perception of Self Generalized Stereotypes and Brand Selection," Journal of Marketing Research, 5 (February), 58-63.

Guttman, Jonathan (1973), "Self-Concepts and Television viewing Among Women," Public Opinion Quarterly, 34 (Fall), 388-397.

Higie, Rogin A., Lawrence F. Feick, and Linda A. Price (1987), "Types and Amount of Word-of-Mouth Communications About Retailers," Journal of Retailing, 63 (Fall), 260-277.

Jacoby, Jacob (1971), "Personality and Innovation Proneness," Journal of Marketing Research, 8 (May), 244-247.

Kahle, Lynn R., Aviv Shoham, Gregory M. Rose, Malcolm Smith, and Rajeev Batra (1994), "Economic Versus Personal Future-Oriented Attitudes as Consumer Indicators," a working paper, the University of Oregon.

Kassarjian, Harold H. and Mary Jane Sheffet (1991), "Personality and Consumer Behavior: An Update," in Perspectives in Consumer Behavior, eds. Harold H. Kassarjian and Thomas S. Robertson, Englewood Cliffs, NJ: Prentice Hall, 281-303.

King, Charles W. and John O. Summers (1970), "Overlap of Opinion Leadership Across Consumer Product Categories," Journal of Marketing Research, 7 (February), 43-50.

Langeard, E., M. Crousillat and R. Weisz (1977), "Exposure to Cultural Activities and Opinion Leadership," in Advances in Consumer Research, 5, ed. H. Keth Hunt, Ann Arbor, MI: Association for Consumer Research.

Levy, Sidney J. (1959), "Symbols for Sales," Harvard Business Review, 37 (July/August), 117-124.

Lichtenstein, Donald R. and Scot Burton (1990), "An Assessment of the Moderating Effects of Market Mavenism and Value Consciousness on Price-Quality Perception Accuracy," in Advances in Consumer Research, 17, eds. Marvin E. Goldberg, Gerald Gorn, and Richard W. Pollay, Provo, UT: Association for Consumer Research.

Merton, Robert (1957), Social Theory and Social Structure, Glencoe, IL: The Free Press.

Myers, James H. and Thomas S. Robertson (1972), "Dimensions of Opinion Leadership," Journal of Marketing Research, 9 (February), 41-46.

Prince, Melvin (1993), "Self-Concept, Money Beliefs and Values," Journal of Economic Psychology, 14, 161-173.

Rogers, Everett M. (1983), Diffusion of Innovations, NY: The Free Press.

Schrank, H. L. (1973), "Correlates of Fashion Leadership: Applications for Fashion Process Theory," The Sociological Quarterly, 14 (Autumn), 534-543.

Sirgy, Joseph M. (1982), "Self-Concept in Consumer Behavior: A Critical Review," Journal of Consumer Behavior, 9 (December), 287-300.

Slama, Mark E. and Terrel G. Williams (1990), "Generalization of the Market Maven Information Provision Tendency Across Product Categories," in Advances in Consumer Research, 17, eds. Marvin E. Goldberg, Gerald Gorn, and Richard W. Pollay, Provo, UT: Association for Consumer Research.

Summers, John O. (1970), "The Identity of Women’s Clothing Fashion Opinion Leaders," Journal of Marketing Research, 7 (May), 178-185.

Venkatraman, Meera P. (1989), "Opinion Leaders, Adopters, and Communication Adopters: A Role Analysis," Psychology and Marketing, 6 (Spring), 51-68.

Venkatraman Meera P. (1990), "Opinion Leadership, Enduring Involvement and Characteristics of Opinion Leaders: A Moderating or Mediating Relationship?" in Advances in Consumer Research, 17, eds. Marvin E. Goldberg, Gerald Gorn, and Richard W. Pollay, Provo, UT: Association for Consumer Research.

White, Irving S. (1967), "The Perception of Value in Products," in On Knowing the Consumer, ed. Joseph W. Newman, NY: Wiley, 90-106.

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