Asia Pacific Advances in Consumer Research Volume 1, 1994 Pages 203-207
OPINION LEADERSHIP AND SELF-MONITORING: EVIDENCE FOR THE TWO-WAY FLOW OF COMMUNICATIONS
Gulden Asugman, Bogazici University
Eser Borak, Bogazici University
Muzaffer Bodur, Bogazici University
In order to assess its role in the flow of communications opinion leadership is investigated as to its correlation with self-monitoring as a personality trait. Results reveal that in some areas opinion leaders are high self-monitoring individuals. Our findings provide support as to the presence of a two-way flow of information between opinion leaders and others since high self-monitors are people who seek out and consult social comparison information. Results have been discussed in terms of possible contributions to diffusion of innovations and consumer behavior literatures.
"Who says what to whom and with what effect? (Katz and Lazarsfeld 1955) is the crucial question that researchers try to answer for decades now. The concept of opinion leadership has been developed thus, to be able to answer this question, even if partly. This study focuses on opinion leadership and assesses the nature of opinion leaders in terms of communicative tendencies.
People receive messages both from impersonal and personal sources and to interpret these they rely on other people. In this context, opinion leaders and their characteristics have been of major interest to researchers and marketers. Rogers (1983) defined opinion leadership as "the degree to which an individual is able informally to influence other individuals' attitudes or overt behavior in a desired way with relative frequency (p. 271)." As explained by Katz and Lazarsfeld (1955), opinion leaders are the major elements of the two-step flow of communications. The first step in this model of communications involves the flow of information from mass-media sources to opinion leaders. In the second step, information is transmitted from the opinion leaders to others in the form of word-of-mouth. Thus, an important goal for the researchers in this area has been to identify common characteristics of opinion leaders so as to enable the marketers to reach these individuals with ease. The two-step flow of communications model, however, was criticised by several researchers as Turnbull and Meeneghan (1980) pointed out. The idea of active opinion leaders and a passive audience was questioned (Myers and Robertson 1972; Troldahl 1971). In this regard, Aaker and Myers (1975) stated that this model ignored "the exchange or two-way transaction nature of communications and influence (p. 372)." Arndt (1968) who found the opinion leaders to be active receivers of word-of-mouth as well as transmitters, interpreted this as an unexpected finding and attributed it to the opinion-sharing nature of the communication situation rather than to the nature of opinion leaders. Lack of conclusive evidence as to the nature of communications between opinion leaders and followers (i.e., one-way from opinion leaders to followers or two-way), suggests the need for further research in this area.
Thus, the present study attempts to examine this aspect of opinion leadership by focusing on its potential relationship with self-monitoring. Delineating the relationship between opinion leadership and self-monitoring can shed light to the nature of communications in the two-step flow model. Specifically, if opinion leaders are high self-monitoring individuals then it would mean that opinion leaders are not mere transmitters of information but also are recievers of information from their group due to their nature. Findings of Snyder (1974) revealed that "individuals with high self-monitoring scores were more likely than those with low scores to seek out and consult social comparison information about their peers (p. 536)."
Characteristics of opinion leaders have been researched extensively. Robertson and Myers (1969) attempted to define the personality characteristics of opinion leaders. Examining a sample of housewives they found out that none of the 18 personality characteristics under investigation significantly correlated with opinion leadership. In an attempt to profile the women's clothing fashion opinion leaders, Summers (1970) also investigated a number of demographic, sociological, and attitudinal variables. Regarding the demographic characteristics, findings implied that opinion leaders were more likely to be found among people who were younger, more educated, had higher incomes and higher occupational status. Physical mobility, social communications, affiliations with organizations and participation in social activities were the sociological characteristics which seemed to determine women's clothing fashion opinion leaders. These people were more progressive, outgoing, and susceptible to change.
Silk (1971) focused on the relationship between interest and opinion leadership. For the fashion involvement factor, research indicated that women's clothing fashion opinion leaders were more interested and knowledgable in the topic compared to non-leaders.
Darden and Reynolds (1972) studied men's apparel fashion opinion leaders. For the subsample of suburban men, results showed a strong relationship between fashion interest and fashion leadership. These findings, then, are in line with that of Silk (1971). According to Darden and Reynolds (1972), fashion interest and fashion venturesomeness were context-free characteristics of fashion opinion leaders.
Among the correlates of opinion leadership analyzed in the reviewed literature, personality variables were not found to be strong predictors of opinion leadership. Examples of personality traits investigated in relation to opinion leadership were dominance, self-acceptance, self-control, tolerance, flexibility, and intellectual efficiency. Self-monitoring, however, has not been researched as a potential correlate of opinion leadership in previous research.
Besides investigating the correlates of opinion leadership, several studies in the area focused on the overlap in opinion leadership. While some researchers claimed the presence of generalized opinion leaders (King and Summers 1970; Marcus and Bauer 1964; Myers and Robertson 1972;), others argued that an opinion leader in one area need not be an opinion leader in other areas as well (Katz and Lazarsfeld 1955; Silk 1971; Sudman 1971). Upon this issue, a middle of the road approach seems to dominate which maintains the view that opinion leadership overlap is most likely to take place across topics of common interest.
The concept of opinion leadership was also examined in cross-national settings (Cosmas and Sheth 1980; Yavas, Riecken and Haahti 1982). Cosmas and Sheth (1980) investigated opinion leadership across different cultures living in the United States. Yavas, Riecken and Haahti (1982) carried out their research across three subsamples, namely, Turkish, Finnish, and American college students. In this study, researchers' aim was to find out if the self-designated opinion leadership scale of King and Summers was reliable or not as a cross-national concept. Areas of opinion leadership investigated were clothing fashions, elective classes, academic majors, professsors, and career opportunities. Results of this research indicated that opinion leadership was a cross-nationally reliable concept.
MEANS AND STANDARD DEVIATIONS FOR OPINION LEADERSHIP SCORES (OLS)
OPINION LEADERSHIP AND SELF-MONITORING
Considering the findings of previous studies, this study basically aimed at investigating the correlation of opinion leadership with an individual's self-monitoring degree.
Self-monitoring of expressive behavior, as put forward by Mark Snyder (1974), refers to controlling and modifying one's social behavior and emotional expression out of a concern for social appropriateness. Snyder (1974) explained that a high self-monitoring individual modified his/her behavior according to the cues in the situation whereas the low self-monitoring individual was more consistent across situations. The role of self-monitoring as a moderating variable in consumer behavior was examined by Becherer and Richard (1978). Their findings confirmed their hypothesis that the low self-monitoring group behaved differentially than the high self-monitoring group in the case of private brand proneness.
Self-monitoring which indicates the presence of "self-observation and self-control guided by situational cues to social appropriateness (Snyder 1974, p. 526)" can be used to test the nature of opinion leadership phenomenon. Specifically, the presence of a two-way influence between the opinion leaders and their followers can be assessed. If opinion leaders are individuals with high self-monitoring levels, then it means that an opinion leader acts as a mediator and represents a typical group-mind since the very nature of a high self-monitoring person requires his/her undivided attention to social and situational cues and attentiveness to others' behavior. Tentative findings regarding the possibility of a two-way flow between opinion leaders and their followers (Arndt 1968; Myers and Robertson 1972) suggest a need to further elaborate this aspect of opinion leadership. In the present study it is hypothesized that opinion leadership will be positively related to the level of self-monitoring.
The respondents were 154 undergraduate students of both sexes in a Turkish university where the instruction is in English. Questionnaires were administered to sophomores in the Administrative Sciences faculty in various courses during classtime. Utilizing student subjects should not affect the results since research in both opinion leadership and self-monitoring also used students as subjects (Becherer and Richard 1978; Jacoby 1974; Yavas, Riecken, and Haahti 1982).
Opinion leadership was assessed using the self-designated opinion leadership scale developed by King and Summers (1970). Due to the evidence regarding its reliability (e.g, Yavas and Riecken 1982; Yavas, Riecken, and Haahti 1982), this scale was utilized in the present study. Considering the findings of Childers (1986) regarding the increased internal consistency reliability when five response categories are used for all items, this version of the response format was utilized in the present research. Opinion leadership was investigated for; packaged foods, clothing items, personal care items, travel, entertainment, and clubs on the campus as well as for general opinion leadership.
The measure utilized for the self-monitoring construct was the self-monitoring scale developed and tested by Snyder (1974). As indicated by Becherer and Richard (1978), this scale demonstrates internal consistency, stability over time, and discriminant validity. The statements in this scale had a true/false response format.
For each respondent 6 total opinion leadership scores were computed based on the topics covered. The range for these scores was between 6 and 30. The question which dealt with the overall evaluation of being an opinion leader or not, was treated separately as it designated generalized opinion leadership. The range of this variable was between 1 and 5. Frequency results for the opinion leadership scores are presented in Table 1. These results indicate that to the present sample travelling and entertainment are more relevant areas followed by clothing.
To construct a single index score for self-monitoring, 25 true-false items of this scale were summed and a total self-monitoring score was calculated for each subject. The range of the total scores was between 0 and 25. The frequency distribution of these total scores appeared to be between 2 and 21 with a mean of 10.8.
Both self-monitoring and opinion leadership seem to exist in every individual in varying degrees. Thus, a number of Pearson correlation analyses were performed to see the extent of association between self-monitoring and opinion leadership.
The results indicate that self-monitoring is correlated with opinion leadership in some areas but not in others (see Table 2). Specifically, self-monitoring correlates significantly with opinion leadership in travelling, and with opinion leadership in entertainment (p<0.01). There is a marginal correlation between self-monitoring and generalized opinion leadership (p<0.1). For other areas of opinion leadership, however, the level of correlations are not significant. One reason for this observation maybe that for the present sample areas other than travelling and entertainment were not of much interest. Further, it is noteworthy that high correlations between opinion leadership and self-monitoring occurred in areas of group activities rather than in personal consumption topics. This finding can be related to the social visibility of consumption situations and the role of self-monitoring in these as suggested by Becherer and Richard (1978).
DEGREE OF CORRELATION OF OPINION LEADERSHIP SCORES (OLS) WITH SELF-MONITORING
The objective of this paper was to examine the relationship between opinion leadership and self-monitoring to assess the nature of communications between opinion leaders and others. Findings suggest important implications for the opinion leadership concept. First, by assessing the relationship between opinion leadership and self-monitoring it is shown that at least in some areas of opinion leadership, self-designated opinion leaders are also high self-monitors. Specifically, high self-monitors are likely to be opinion leaders in areas of travel and entertainment which are group activities. If opinion leaders are also high self-monitors, as suggested in the present findings, then these individuals are likely to play a much more important and different role in consumer behavior than originally assumed in opinion leadership and diffusion of innovations literature. Opinion leaders were described as people whose opinions are asked and who influence others' opinions (Katz and Lazarsfeld 1955). Thus, marketing literature treated opinion leaders as transmitters of information within the flow of communications from mass media sources to consumers in general. Findings of this study, however, point out another possibility. Since the very nature of a high self-monitoring person requires his/her attention to social and situational cues and attentiveness to other's behavior, our findings indicate the presence of a two-way flow of influence between opinion leaders and followers rather than a one-way flow from leaders to followers.
Furthermore, since "high self-monitoring individuals have richer, better articulated, and more informative images of prototypic persons in the same large set of behavioral domains than do low self-monitoring individuals (Snyder 1979, p. 106)", opinion leaders in some areas who are also high self-monitors represent their groups. These individuals not only act as mediators of information but reflect a typical group-mind which maybe very important for consumer behaviorists and marketers. In this sense, opinion leaders can be utilized to find out the characteristics and attitudes of their followers. Utilizing such individuals in focus groups for example, can provide marketers with enhanced information regarding their target market. These people can also be useful to researchers advocating the relativist approach due to their ability in reading others and different situations.
Studies which attempted to find personality correlates of opinion leadership have failed to do so (Robertson and Myers 1969). This can be due to the possibility that opinion leaders investigated were high self-monitoring individuals. Characteristics of opinion leaders listed by Robertson (1971) supports this view. Robertson stated that opinion leaders were individuals with no distinguishing personality features and with greater norm adherence. Both of these points happen to characterize high self-monitoring individuals as well. Due to their nature, high self-monitors modify their behavior across situations and it may be difficult to detect dispositional characteristics of such individuals. Becherer and Richard (1978) pointed out that "low self-monitoring individuals will be associated with dispositional variables such as personality traits, while the behavior of high self-monitors will relate more closely to situational cues (p.159)."
It seems that self-monitoring level of an individual is correlated with the opinion leadership level. Further research is needed to investigate this relationship with other representative samples so that it becomes possible to generalize this finding.
Aaker, D. A. and Myers, J. G. (1975), Advertising Management, Englewoodd Cliffs, NJ: Prentice-Hall.
Arndt, Johan (1968), "A Test of the Two-Step Flow in Diffusion of a New Product," Journalism Quarterly, 45, 457-65.
Becherer, Richard C. and Lawrence M. Richard (1978), "Self-Monitoring as a Moderating Variable in Consumer Behavior," Journal of Consumer Research, 5, 159-63.
Childers, Terry L. (1986), "Assessment of the Psychometric Properties of an Opinion Leadership Scale, Journal of Marketing Research, 23, 184-88.
Cosmas, S. C. and J. N. Sheth (1980), "Identification of Opinion Leaders Across Cultures: An Assessment for Use in the Diffusion of Innovations and Ideas," Journal of International Business Studies, 11, 65-72.
Darden, William R. and Fred D. Reynolds (1972), "Predicting Opinion Leadership for Men's Apparel Fashions," Journal of Marketing Research, 9, 324-28.
Jacoby, Jacob (1974), "The Construct Validity of Opinion Leadership," Public Opinion Quarterly, 38, 81-89.
Katz, Elihu and Paul F. Lazarsfeld (1955), Personal Influence, The Free Press: Glenco,Ill.
King, Charles W. and John O. Summers (1970), "Overlap of Opinion Leadership Across Consumer Product Categories," Journal of Marketing Research, 7, 43-50.
Marcus, A. S. and R. A. Bauer (1964), "Yes! There are Generalized Opinion Leaders," Public Opinion Quarterly, 28, 628-32.
Myers, James H. and Thomas S. Robertson (1972), "Dimensions of Opinion Leadership," Journal of Marketing Research, 9, 41-46.
Robertson, T. S. (1971), Innovative Behavior and Communication, Holt, Rinehart, and Winston.
Robertson, Thomas S. and James H. Myers (1969), "Personality Correlates of Opinion Leadership and Innovative Buying Behavior," Journal of Marketing Research, 6, 164-68.
Rogers, Everett M. (1983), Diffusion of Innovations, 3rd ed., The Free Press: New York.
Silk, Alvin J. (1971), "Response Set and the Measurement of Self-Designated Opinion Leadership," Public Opinion Quarterly, 35, 383-97.
Snyder, Mark (1974), "Self-Monitoring of Expressive Behavior," Journal of Personality and Social Psychology, 30, 526-37.
Snyder, Mark (1979), "Self-Monitoring Processes," Advances in Experimental Social Psychology, 12, 85-122.
Sudman, Seymour (1971), "Overlap of Opinion Leadership Across Consumer Product Categories," Journal of Marketing Research, 8, 258-61.
Summers, John O. (1970), "The Identity of Women's Clothing Fashion Opinion Leaders," Journal of Marketing Research, 7, 178-85.
Troldahl, Verling C. (1971), "A Field Test of a Modified 'Two-Step Flow of Communication' Model," Public Opinion Quarterly, 35, 609-23.
Turnbull, P. W. and A. Meenaghan (1980), "Diffusion of Innovation and Opinion Leadership," European Journal of Marketing, 14, 3-33.
Yavas, Ugur and Glen Riecken (1982), "Extensions of King and Summers' Opinion Leadership Scale: A Reliability Study," Journal of Marketing Research, 19, 154-55.
Yavas, Ugur, Glen Riecken, and Antti Haahti (1982), "Reliability Assessment of Opinion Leadership Concept: A Cross-National Study," The Finnish Journal of Business Economics, Special Edition, 3, 239-46.